Garfield Co

COMMUNITY HEALTH RISK ANALYSIS
OF
OIL AND GAS INDUSTRY IMPACTS

IN
GARFIELD COUNTY
Teresa Coons, Ph.D. Saccomanno Research Institute
Russell Walker, Ph.D. Mesa State College
ii
CommunityHealthRiskAnalysisofOilandGasIndustryImpactsinGarfieldCounty
ACKNOWLEDGMENTS
The individuals and organizations listed below provided valuable data, information and/or technical supporttothisstudy. SACCOMANNORESEARCHINSTITUTE KathyPetefish SusanKeizer MaryNichols TearaCarr KelliJackson ST.MARYSHOSPITALANDREGIONALMEDICALCENTER ROCKYMOUNTAINHEALTHPLANS GARFIELDCOUNTY GarfieldCountyPublicHealth GrandRiverHospitalDistrict ValleyViewHospital Concernedcitizensandorganizations COLORADODEPARTMENTOFPUBLICHEALTHANDENVIRONMENT NATURALGASINDUSTRYPRODUCERSOPERATINGINGARFIELDCOUNTY
iii
TABLE OF CONTENTS
PAGE EXECUTIVESUMMARY xiv I NTRODUCTIONTOTHE S TUDY 1 PART I: R ISK A SSESSMENT 9 Descriptionofstepsandcomponentsofnaturalgasoperations Descriptionofpollutantsassociatedwithspecificnaturalgasoperations Overviewofriskassessmentandriskmanagementprocess Qualitativedescriptionoftransportprocesses Dataonpollutantconcentrationsatknownorpossibleexposurepoints Resultsofpollutantconcentrationmodeling Descriptionoftoxicityofmostimportantpollutants Resultsofriskassessmentforselectedpollutants Resultsinterpretation P ART II: H EALTH S TUDY Premisesofthestudy Methodology Descriptionofstudyareaandcomparisoncounties PerceptionsandConcerns(Qualitativedatacollection) Quantitativedatacollection Householdsurvey Results PerceptionsandConcerns(Qualitativedatafromfocusgroups,interviews, publicmeetingsandcomplaintreports) QuantitativeHealthData Birthstatistics(1996,20022006) Deathstatistics:ratesandleadingcausesofdeath(19902006) Birthdefects:typesandrates(20002005) Childandadolescenthealthmeasures(20012005) Injuryhospitalizationanddeath:causesandrates(20012003) BehavioralRiskFactorStudySurvey(BRFSS)data(20042005) WestNilevirus(20022007) Cancerstatistics(19922005) Reportableconditions(19982006) Sexuallytransmitteddiseases(20002007) HospitalandOutpatientdata DRGbasedhospitaldischargedata(2000through1stquarter2007) Emergencyroomdata(GrandRiverHospitalDistrict;RockyMountain HealthPlansmemberdata;ValleyViewHospital) 49 49 50
63 63 66
iv
PAGE RockyMountainHealthPlans(RMHP)inpatientandoutpatienthospital visitsandambulatorymemberdata:4countycomparison St.MarysHospitalCareFlightdata HouseholdSurveyResults 178 Descriptionofsurveypopulationdemographics Surveyoutcomes P ART III: R ECOMMENDATIONSAND S TUDY L IMITATIONS A PPENDICES A. Equationsusedinexposureassessment B. GaussianPlumemodel C. DeterminationofwindspeedforuseinGaussianPlumemodel D. Mostfrequentwinddirections E. Durationoftimesunisineachinsolationcategory F. Fractionofmonthreportedasclear,partlycloudy,orcloudy G. Occurrenceofeachairstabilityclassintypicalyear H. Calculationofpollutantconcentrationsinair IA.Benzene,toluene,andm,pxyleneconcentrationsduringflowback IB.Benzene,toluene,andm,pxyleneconcentrationsfromglycoldehydrationemissions IC.Benzene,toluene,andm,pxyleneconcentrationsfromcondensatetankemissions J.Riskassessmentoftotalpetroleumhydrocarbonresidueinsoilat1,000mg/kg K.CausesofdeathbyCounty20032006 L.Causeofdeathbyrace/ethnicityforGarfieldCountyresidents2006 M.CancerincidenceandmortalityratesbyCountyfor1992through2005 N.SexuallyTransmittedDiseasesRatesbyCountyfor200020007 O.ColoradoHospitalAssociationutilizationdata2000through1stquarter2007 P.RockyMountainHealthPlans(RMHP)memberdataforhealthplanusage2000through1st quarter2007 Q.GrandRiverHealthDistrictemergencyroom(ER)admissions20042006,includingERdata groupings R.ValleyViewHospitalemergencyroomadmissions20042006 S.CareFlightdata T.Householdsurveyinstrumentandotherdocuments U.Householdsurveydataanalysisandoutcomes
226
LIST OF TABLES AND FIGURES

TABLES PAGE I NTRODUCTIONTOTHE S TUDY 1. Annualpermitstodrill 4 2. Coloradocoalminestatistics,2006(ColoradoDivisionofReclamation, 8 MiningandSafety) PART I: R ISK A SSESSMENT 3. SummaryofPM10monitoringresults 4. SummaryofstatisticsfordetectedVOCs 5A.Benzeneconcentrationinairresultingfromselectednaturalgas developmentoperations 5B.Tolueneconcentrationsinairresultingfromselectednaturalgas developmentoperations 5C.m,pXylenesconcentrationsinairresultingfromselectednaturalgas developmentoperations 6.Referenceconcentrations 7A.Benezenecancerriskresultingfromflowbackwithnogasrecovery 7B.Benezenecancerriskresultingfromflowbackwith93%recoveryofnaturalgas 7C.Benzenecancerriskresultingfromoperationofwellheadglycoldehydrationunits 7D.Benzenecancerriskresultingfromcondensatetanksemitting20tonsperyear ofvolatileorganichydrocarbons 7E.Benzenecancerriskresultingfromcondensatetankswith98%removalof emissionsof20tonsperyearofvolatileorganichydrocarbons P ART II: H EALTH S TUDY Descriptionofstudyareaandcomparisoncounties 8.Comparisoncountyprofiles:2000and2006data Perceptionsandconcerns 9.Qualitativedata:Intervieweedemographics Quantitativehealthdata 10.Vitalstatisticsbycounty:DeltaCounty1996,20022006 11.Vitalstatisticsbycounty:GarfieldCounty1996,20022006 12.Vitalstatisticsbycounty:MesaCounty1996,20022006 13.Vitalstatisticsbycounty:MontroseCounty1996,20022006 14.SelectedbirthcharacteristicsbyCounty,2006
21 23 30 31 32 34 36 37 38 39
40
51
63
67 69 71 73 80
vi
15.AgespecificdeathratesbyCounty,2006 16.NumberofresidentlivebirthsinGarfieldCounty:20002005 17.Ratesofbirthdefects,20002005 18.MajorcongenitalanomalyratesinGarfieldCounty:20002005 19.ChildandadolescenthealthbyCounty 20A.InjuryhospitalizationsbyCounty,20012003.Totalintentioncategories 20B.InjuryhospitalizationsbyCounty,20012003.Totalinjurytype 21A.InjurydeathsbyCounty,20012003.Totalintentioncategories 21B.InjurydeathsbyCounty,20012003.Totalinjurytype 22.BehavioralRiskFactorStudySurvey(BRFSS)data,20042005 23.Comparisonofcountycancerincidencerateswithstaterates,19922005 24.Comparisonofcountycancermortalityrateswithstaterates,19922005. 25AK.ReportableconditionsratesbyCounty,(20002006)andSTDrates,20002007 26.Hospitalutilization,GarfieldCountyresidents,20052006 27.Top10DiagnosisRelatedGroup(DRG)categoriesbyyearinGarfieldCounty 28.Emergencyroomadmissionsdata (RockyMountainHealthPlansmemberdata,20002007;GrandRiver HospitalDistrictdata,20042006;ValleyViewHospitaldata,20042006) 29.Inpatienthospitaladmissionsdata (RockyMountainHealthPlansmemberdata,20002007) 30.Outpatienthospitalvisitsorlabworkdata (RockyMountainHealthPlansmemberdata,20002007) 31.Ambulatoryofficevisitsdata (RockyMountainHealthPlansmemberdata,20002007) Householdsurvey 32.Householdsurveydemographics 33.Householdsurveyresponserate
PAGE 82 85 86 88 92 100 101 102 103 104 107 108 133 142 144 163
169 171 174
178 179
vii
FIGURES PAGE I NTRODUCTIONTOTHE S TUDY 1. Relationshipbetweenexposurepotentialandhealthassessment 2. DrillingrigsrunninginColoradobyCountyeachweekin2007 3. ActiveoilandgaswellsallColoradoCounties,asof030608 4. Locationofcoalmines,powerplants,railroadsandcoalbearingregionsin Colorado,2006 5. LocationofRifleuraniumprocessingsitesanddisposalsite PART I: R ISK A SSESSMENT 6. PiceanceBasin,WilliamsForkFormation 7. AconceptualmodelforcontainmenttransportfromGarfieldCountynaturalgas Operations P ART II: H EALTH S TUDY Descriptionofstudyareaandcomparisoncounties 8. WorldHealthOrganizationdefinitionofhumanhealth 9. ColoradoCountiesmap 10. PopulationestimatesbyCounty,20002006 11. Averageannualpercentchangeinpopulation,20002006 12. Populationageasapercentageoftotalpopulation,Year2000censusdata 13. Raceandethnicityasapercentageoftotalpopulation,Year2000censusdata 14. MedianhouseholdincomebyCounty,Year2000censusdata 15. Medianhouseholdincomebycensustract,Year2000censusdata 16. PercentofpersonsbelowpovertybyCounty,Year2000censusdata 17. Percentofpersonsbelowpovertybycensustract,Year2000censusdata 18. Percentofchildrenbelowpovertybycensustract,Year2000censusdata
1 5 5 6
9 14
49 50 53 53 54 54 55 55 56 57 57
Perceptionsandconcerns 19. Communityinterviews/focusgroupcomments 65 Quantitativehealthdata 20. FirsttrimesterprenatalcarebyCounty,2005 21. LowbirthweightbyCounty,20032005 22. VerylowbirthweightbyCounty,20032005 23. NeonatalmortalitybyCounty,20012005 24. InfantmortalitybyCounty,20012005 25. Crudetotaldeathrates:Delta,Garfield,MesaandMontroseCounties,19962006 26. Averageannualageadjusteddeathratesbycause,20022006 27. DeathratesforGarfieldCountymalesbyagegroups,19902006 28. DeathratesforGarfieldCountyfemalesbyagegroups,19902006 29. LeadingcausesofdeathbyCounty,2006 75 76 77 78 79 81 81 83 83 84
viii
30.TeenfertilityratesbyCounty,20032005 31.TeenhospitalizationratesbyCounty,20022005 32. TeenmotorvehicledeathratesbyCounty,20042005 33. TeensuicideratesbyCounty,20012005 34. ChildabuseratesbyCounty,2005 35. ChilddeathratesbyCounty,20012005 36. InjuryhospitalizationswithintentioncategoriesbyCounty,20012003 37. InjuryhospitalizationsbyinjurytypeandCounty,20012003 38. InjurydeathswithintentioncategoriesbyCounty,20012003 39. InjurydeathsbyinjurytypeandCounty,20012003 40AD.Allcancers:incidenceandmortalityratetrendsformales&females byCounty,19922005 41AB.Prostatecancer:incidenceandmortalityratetrendsbyCounty,19922005 42AB.Breastcancer:incidenceandmortalityratetrendsbyCounty,19922005 43AD.Colorectalcancer:incidenceandmortalityratetrendsformales&females byCounty,19922005 44AD.Lungcancer:incidenceandmortalityratetrendsformales&females byCounty,19922005 45AB.Bladdercancer:incidenceratetrendsformales&femalesbyCounty, 19922005 46AB.Melanoma:incidenceratetrendsformales&femalesbyCounty,19922005 47AB.Leukemia:incidenceandmortalityratetrendsformales&females byCounty,20032005 48.Thyroidcancer:incidenceratetrendsformales&femalesbyCounty,20032005 49.Cervicalcancer:incidenceratetrendsbyCounty,20032005 50.WestNileVirusinfectionsbyCounty2003 51.WestNileVirusinfectionsbyCounty2004 52.WestNileVirusinfectionsbyCounty2005 53.WestNileVirusinfectionsbyCounty2006 54.WestNileVirusinfectionsbyCounty2007 55.ColoradoWestNileVirushumancases2002 56.ColoradoWestNileVirushumancases2003 57.ColoradoWestNileVirushumancases2004 58.ColoradoWestNileVirushumancases2005 59.ColoradoWestNileVirushumancases2006 60.ColoradoWestNileVirushumancases2007 61.ColoradoWestNileVirussurveillancemap2002 62.ColoradoWestNileVirussurveillancemap2003 63.ColoradoWestNileVirussurveillancemap2004 64.ColoradoWestNileVirussurveillancemap2005 65.ColoradoWestNileVirussurveillancemap2006 66.ColoradoWestNileVirussurveillancemap2007
PAGE 94 95 96 97 98 99 100 101 102 103 109
111 112 113 115 117 118 119 120 120 122 123 123 124 124 125 125 126 126 127 127 128 128 129 129 130 130
ix
PAGE 67.AllhospitalizationsbyCountyofresidenceCombinedDiagnosticRelated 147 Group(DRG)Categories,20001stquarter2007(ColoradoHospitalAssociationdata) 68.AlltraumahospitalizationsbyCountyofresidence,20001stquarter2007 147 (ColoradoHospitalAssociationdata) 69.AlcoholanddrugdisordershospitalizationsbyCountyofresidence,2000 147 1stquarter2007(ColoradoHospitalAssociationdata) 70.BirthingandpregnancydisordershospitalizationsbyCountyofresidence,2000 148 1stquarter2007 (ColoradoHospitalAssociationdata) 71.NeonataldisordershospitalizationsbyCountyofresidence,20001stquarter2007 149 (ColoradoHospitalAssociationdata) 72.NormalbirthshospitalizationsbyCountyofresidence,20001stquarter2007 149 (ColoradoHospitalAssociationdata) 73.FemalereproductivedisordershospitalizationsbyCountyofresidence,2000 150 1stquarter2007(ColoradoHospitalAssociationdata) 74.MalereproductivedisordershospitalizationsbyCountyofresidence,2000 150 1stquarter2007(ColoradoHospitalAssociationdata) 75.BoneandjointdisordershospitalizationsbyCountyofresidence,20001stquarter2007 151 (ColoradoHospitalAssociationdata) 76.CirculationdisordershospitalizationsbyCountyofresidence,20001stquarter2007 151 (ColoradoHospitalAssociationdata) 77.DigestivedisordershospitalizationsbyCountyofresidence,20001stquarter2007 152 (ColoradoHospitalAssociationdata) 78.Ear,nose,andthroatdisordershospitalizationsbyCountyofresidence,2000 153 1stquarter2007(ColoradoHospitalAssociationdata) 79.InfectiousandparasiticdiseaseshospitalizationsbyCountyofresidence,2000 153 1stquarter2007(ColoradoHospitalAssociationdata) 80.MentaldisordershospitalizationsbyCountyofresidence,20001stquarter2007 154 (ColoradoHospitalAssociationdata) 81.NervoussystemdisordershospitalizationsbyCountyofresidence,20001stquarter2007 155 (ColoradoHospitalAssociationdata) 82.Nutritional,metabolicandthyroiddisordershospitalizationsbyCountyofresidence, 155 20001stquarter2007(ColoradoHospitalAssociationdata) 83.PancreasandliverdisordershospitalizationsbyCountyofresidence,2000 156 1stquarter2007(ColoradoHospitalAssociationdata) 84.RedcellandclottingdisordershospitalizationsbyCountyofresidence,2000 156 1stquarter2007(ColoradoHospitalAssociationdata) 85.AllRespiratorydisordershospitalizationsbyCountyofresidence,20001stquarter2007 157 (ColoradoHospitalAssociationdata) 86.RespiratorydisordershospitalizationsbyCountyofresidenceBronchitis,asthma, 158 COPDandrespiratoryinfections&inflammation,20001stquarter2007 (ColoradoHospitalAssociationdata) 87.SkindisordershospitalizationsbyCountyofresidence,20001stquarter2007 158 (ColoradoHospitalAssociationdata)
PAGE 88.Skingrafts,traumaandpoisoninghospitalizationsbyCountyofresidence,2000 159 1stquarter2007(ColoradoHospitalAssociationdata) 89.UrinarytractdisordershospitalizationsbyCountyofresidence,20001stquarter2007 159 (ColoradoHospitalAssociationdata) 90.Hospitaldischarges:asthmaasprincipaldiagnosisbyCounty 160 91.CareFlightsemergencyservicesbyCounty:allemergencies,20042007 177 92.CareFlightsemergencycallstodrillrigs:GarfieldandRioBlancoCounties, 177 20042007 Householdsurvey 93.Qualityofhouseholdsurveyinterviews 180 94.LengthofresidenceinGarfieldCounty 181 95.LengthofresidenceinGarfieldCounty:Lengthoftimeincurrentresidence 181 96.LocationofresidenceinGarfieldCounty 182 97.Householdsize(residentsperhousehold) 182 98.Householdethnicity 183 99.Meanageofhouseholdresidents 183 100.Genderofhouseholdresidents 183 101.Percentageofchildreninrespondenthouseholds 184 102.Educationlevelofsurveyrespondent 184 103.Percentageofhouseholdswithhealthinsurance 184 104.Currenthealthofhouseholdresidents 185 105.Healthofhouseholdresidentsoneyearago 185 106.PercentageofindividualsreportingselecteddiseasesandsymptomsCountyoverall 186 107.Percentageofindividualssufferingfromfrequentheadachesormigrainesbyzipcode 187 108.Mainwatersupplyandnumberofindividualsreportingfrequentheadaches 188 109.Primarysourceofdrinkingwaterathome 188 110.Percentageofindividualsconcernedabouthealthproblemsrelatedtowatersupply 189 111.DayspermonthofalcoholconsumptionbyageCountyoverall 189 112.Numberofdrinksperdayforindividualswhoconsumedalcoholduringthe 190 past30daysCountyoverallbyage 113.Percentageofhouseholdmemberswhoreportedusingrecreationaldrugsduring 190 theprevious30daysbyzipcode 114.Percentageofindividualswhohavesmokedmorethan100cigarettesbyzipcode 191 115.Percentageofindividualswhohavesmokedmorethan100cigarettesbyagegroup 191 Countyoverall 116.Frequencyofsmokingforindividualswhohavesmokedmorethan100cigarettes 192 Intheirlivesbyzipcode 117.Smokingfrequencyofindividualswhohaverespiratoryconditions:zipcode81601 193 118.Smokingfrequencyofindividualswhohaverespiratoryconditions:zipcode81623 193 119.Smokingfrequencyofindividualswhohaverespiratoryconditions:zipcode81635 193 120.Smokingfrequencyofindividualswhohaverespiratoryconditions:zipcode81647 194 121.Smokingfrequencyofindividualswhohaverespiratoryconditions:zipcode81650 194
xi
122.Smokingfrequencyofindividualswhohaverespiratoryconditions:zipcode81652 123.Percentageofhouseholdmemberswithasthmabyzipcode 124.PercentageofhouseholdmemberswithCOPDbyzipcode 125.Percentageofhouseholdmemberswithemphysemabyzipcode 126.Percentageofhouseholdmemberswithotherlungproblemsbyzipcode 127.Percentageofindividualswithdiabetesbyzipcode 128.DiabetescomplicationsPercentageofindividualswithdiabetesandpain inhandsorfeet 129.DiabetescomplicationsPercentageofindividualswithdiabetesandlossoffeeling 198 inhandsorfeet(comparisonwithlossoffeelinginnondiabeticsurveypopulation) 130.DiabetescomplicationsPercentageofindividualswithdiabetesandkidney 199 problems/proteininurine(comparisonwithkidneyproblemsin nondiabeticsurveypopulation 131.DiabetescomplicationsPercentageofindividualswithdiabetesandeyeproblems 199 132.Percentageofindividualswithcoronarydiseasebyzipcode 200 133.PercentageofindividualswithcoronarydiseaseineachagegroupCountyoverall 200 134.Ofwomenwhobecamepregnant,numberofpregnanciesperwomanbyzipcode 201 135.Percentageofpregnanciesthatresultedinmiscarriagesbyzipcode 201 136.Percentageofchildrenbornwithbirthdefects(selfdefined)byzipcode 202 137.Percentageofchildrendevelopinghealthordevelopmentalproblemswithin5years 202 ofbirthbyzipcode 138.Percentageofindividualsreportingcancersbyzipcode 203 139.PercentageofreportedcancersbyageCountyoverall 204 140.ReportedcancertypesCountyoverall 204 141.CurrentoccupationCountyoverall 205 142.Currentoccupationbyzipcode 206 143.CurrentoccupationbyzipcodePrimarilySpanishspeakinghouseholds 206 144.Currentindustryemploymentbyzipcode 207 145.CurrentindustryemploymentbyzipcodePrimarilySpanishspeakinghouseholds 207 146.Longestoccupationbyzipcode 208 147.LongestoccupationbyzipcodePrimarilySpanishspeakinghouseholds 208 148.Longestindustryemploymentbyzipcode 209 149.LongestindustryemploymentbyzipcodePrimarilySpanishspeakinghouseholds 209 150.Percentageofindividualswhodidntknoworrefusedtoansweroccupationor 210 industryemploymentquestionsbyzipcode 151.Percentageofindividualsemployedintheagriculturalindustryreportingselected 211 diseasesandsymptomsCountyoverall 152.Percentageofindividualsemployedintheminingindustryreportingselected 211 diseasesandsymptomsCountyoverall 153.Percentageofindividualsemployedinthemanufacturingindustryreportingselected 212 diseasesandsymptomsCountyoverall 154.Percentageofindividualsemployedinthewholesaletradeindustryreportingselected 212 diseasesandsymptomsCountyoverall
PAGE 194 195 196 196 197 197 198
xii
PAGE 155.Percentageofindividualsemployedinthefinance,insuranceorrealestate 213 industriesreportingselecteddiseasesandsymptomsCountyoverall 156.Percentageofindividualsemployedinthepersonalservicesindustryreportingselected 213 diseasesandsymptomsCountyoverall 157.Percentageofindividualsemployedintheprofessionalandrelatedservices 214 industriesreportingselecteddiseasesandsymptomsCountyoverall 158.Percentageofindividualsemployedintheforestryorfisheriesindustriesreporting 214 selecteddiseasesandsymptomsCountyoverall 159.Percentageofindividualsemployedintheconstructionindustryreportingselected 215 diseasesandsymptomsCountyoverall 160.Percentageofindividualsemployedinthetransportation,communicationsorpublic 215 utilitiesindustriesreportingselecteddiseasesandsymptomsCountyoverall 161.Percentageofindividualsemployedintheretailtradeindustryreportingselected 216 diseasesandsymptomsCountyoverall 162.Percentageofindividualsemployedinthebusinessandrepairserviceindustries 216 reportingselecteddiseasesandsymptomsCountyoverall 163.Percentageofindividualsemployedintherecreationandentertainmentindustries 217 reportingselecteddiseasesandsymptomsCountyoverall 164.Percentageofadultswhoseindustryemploymentcouldnotbeclassifiedreporting 217 selecteddiseasesandsymptomsCountyoverall 165.Percentageofcurrentlyunemployedadultsreportingselecteddiseasesandsymptoms218 Countyoverall 166.Percentageofcurrentlydisabledadultsreportingselecteddiseasesandsymptoms 218 Countyoverall 167.Percentageofcurrentlyretiredadultsreportingselecteddiseasesandsymptoms 219 Countyoverall 168.Percentageofadultswhodidntknoworrefusedtoprovideanindustryaffiliation 219 reportingselecteddiseasesandsymptomsCountyoverall 169.Percentageofindividualsconcernedabouthealthproblemsrelatedtohome 220 drinkingwatersupplybyzipcode 170.Percentageofhouseholdsconcernedthathealthproblemsmayberelatedtochemicals 221 inornearthehomebyzipcode 171.Concernthatchemicalsinornearthehomemayberelatedtohealthproblems 221 correlationwitheducation 172.Percentageofhouseholdsconcernedthathealthproblemsmayberelatedto 222 environmentalorchemicalhazardsintheirneighborhoodsbyzipcode 173.Concernthatenvironmentalorchemicalhazardsintheneighborhoodmayberelatedto 223 Healthproblemscorrelationwitheducation 174.Concernthatenvironmentalorchemicalhazardsmayberelatedtohealthproblems 224 correlationwithhouseholdincome 175.Percentageofhouseholdsconcernedthatthenaturalgasindustrymaybe 225 relatedtohealthproblemsbyzipcode
xiii
PAGE 176.Concernthatthenaturalgasindustrymayberelatedtohealthproblems 225 correlationwitheducation
xiv
EXECUTIVE SUMMARY
INTRODUCTION This report, both in its format and its content, is intended to meet the needs of several audiences: professional scientists and risk assessors; policy makers at the state, County and local levels; and community members or other interested individuals who may or may not be familiar with this geographic region and/or the natural gas industry. Thus, some of the information included in this report is not typically found in an environmental or health risk assessment study report. We hope, however, that the additional information will add to the understanding of the diverse audience that thestudywasdesignedtoserve.Similarly,inPartII,HealthStudy,wehavepresentedstatisticalhealth outcomesdatafromseveraldifferentsourcesasameansofprovidingvalidationand/orcontrastwith the results of the selfreported household survey data and among the various hospital, insurance providerandstatebaseddatasourcesthatwereavailabletous. Oil and gas activity within Garfield County has generated public concern with regard to impacts on both the environment and public health. As a result, the Garfield County Commissioners approved usingfundingfromafineleviedagainstEnCanaCorporationbytheColoradoOilandGasConservation Commission to conduct a comprehensive study of health and environmental risks to residents of Garfield County. The study, conducted over a period of approximately three years, had two major components: a risk analysis based on environmental exposure data and modeling and a comprehensive health study that creates a baseline assessment of the health of Garfield County residents. TheprimaryfocusofthishealthandriskassessmentstudyisGarfieldCounty.Mesa,Montrose,and DeltaCountieswereselectedascomparisoncountiesforthehealthassessmentportionofthestudy. All four counties are located on the Western Slope of Colorado and share similar social and political cultures and population demographics. All four counties have experienced energy and mining activitiesandtheaccompanyingcyclesofeconomicgrowthandrecession(boomandbust),aswell as environmental and social impacts. Currently, however, Mesa, Montrose and Delta Counties are experiencing relatively fewer impacts from natural gas industry drilling and processing activities. In addition, the four counties have overlapping healthcare networks and service areas, which made it easiertoacquirecomparativehealthdata.
xv
PARTIRISKASSESSMENT Risk was evaluated for human exposure to pollutants associated with natural gas operations in air, water, and soil in Garfield County. A general lack of data on pollutant concentrations in the environment limited the assessment. With respect to air, pollutant concentrations were generated throughamathematicalmodeltosupplementtheconcentrationscollectedbysamplingandanalysis duringtheGarfieldCountyAmbientAirQualityMonitoringReport,June2005May2006.Withrespect to water and soil, a more qualitative evaluation was performed in lieu of data on pollutant concentrations. AGaussianplumemodelwasusedtomakeaplausiblepredictionofairpollutantconcentrationsthat may occur during natural gas operations. The model was based, to the degree possible, on the meteorological conditions specific to Garfield County, and was applied to five specific natural gas operations: flow back during well completion with no recovery of natural gas, flow back with 93% recoveryofnaturalgas,wellheadglycoldehydration,uncontrolledemissionsfromcondensatetanks, and condensate tank emissions controlled by a combustion device. The pollutant concentrations generated by the model were then used in risk calculations based on the U.S. Environmental Protection Agencys Risk Assessment Guidance for Superfund, Volume 1. Human Health Evaluation Manual(PartA). TheEnvironmentalProtectionAgency(EPA)advocatesasagoalthatcancerriskforaspecificexposure benomorethan1chancein1million.TheEPAfurtherspecifiesthatacancerriskofupto1chancein ten thousand is considered acceptable. The results of the risk assessment indicate that the EPAs acceptablevalueforcancerriskcanbeexceededforbenzeneinairforthefollowingsituations: For flow back with no gas recovery, the 70year exposure exceeds the acceptable range for distancesupto500meters(550yards)downwindofthewell. Forflowbackwith93%recoveryofgas,the70yearexposureexceedstheacceptablerangefor distancesupto75meters(82yards)downwindofthewell. For emissions from wellhead glycol dehydration units, the 70year exposure exceeds the acceptablerangefordistancesupto50meters(55yards)downwindofthewell. ForVOCemissionsoftwentytonsperyearfromcondensatetanks,the70yearexposureexceeds theacceptablerangefordistancesupto100meters(110yards)downwindofthetank. Benzene emissions during uncontrolled flow back present the greatest cancer threat. However, the riskofcancerexceedstheEPAacceptablerangeonlyforaseventyyearexposure.Anexposureofthat duration to uncontrolled flow back appears unlikely. A 70year exposure to dehydration unit and condensatetankemissionsmaybemoreplausible,dependingontheactualproductionlifeofnatural gasresourcesinGarfieldCounty.
xvi
The results of the risk assessment for air also indicate that reference concentrations for noncancer effectsmaybeexceededforsomesituations.Referenceconcentrationsarepollutantconcentrations in air representing thresholds below which health effects are very unlikely to occur. Reference concentrations have been developed for acute exposures (one to fourteen days), intermediate exposures (fifteen to 364 days), and chronic exposures (seven years to a lifetime). Modeling results indicatethatreferenceconcentrationscanbeexceededinthefollowingsituations: Forflowbackwithnogasrecovery,thebenzeneacutereferenceconcentrationof30micrograms percubicmeterisexceededfordistancesupto250meters(275yards)downwind. For flow back with no gas recovery, the benzene intermediate reference concentration of 20 microgramspercubicmeterisexceededfordistancesupto300meters(330yards)downwind. Forflowbackwithnogasrecovery,thebenzenechronicreferenceconcentrationforbenzeneof 30 micrograms per cubic meter is exceeded for distances up to 250 meters (275 yards) downwind. For flow back with no gas recovery, the m,pxylenes chronic reference concentration of 100 microgramspercubicmeterisexceededfordistancesupto100meters(110yards)downwind ForemissionsofVOCsattwentytonsperyearfromcondensatetanks,them,pxyleneschronic referenceconcentrationof100microgramspercubicmeterisexceededfordistancesupto50 meters(55yards)downwind. These results suggest that emission of benzene during uncontrolled flow back is the situation that presentsthegreatestthreatofnoncancereffects.Theseeffectsmayoccurinpeoplewhospendone day or more within a distance 250 meters downwind of the natural gas well when this operation is takingplace.Thenoncancereffectsofbenzeneincludeneurotoxicityanddepressionofbonemarrow function,resultinginblooddisorders(decreasedcountsofspecificbloodcells,suchaserythorocytes, leukocytes,andthrombocytes)andimpairmentoftheimmunesystem. There are a number of uncertainties in the noncancer threat and cancer risks determined using pollutantconcentrationsbasedontheGaussianplumemodel.Someoftheseuncertaintieshavethe effectofunderstatingthreatandrisk;othershavetheeffectofoverstatingthreatandrisk.Thereport includesadiscussionoftheseuncertainties. GarfieldCountyresidentshaveexpressedconcernaboutpossibleeffectsongroundwaterandsurface waterusedfordrinkingandotherpurposes. Naturalgasoperationsdohavethepotentialtocreate water contamination, and have done so in certain wellpublicized instances. However, because of a lack of water data representative of broad areas of the county, a quantitative risk assessment for ingestionofcontaminatedgroundwaterorsurfacewaterwasnotperformed.Contaminantpathways anddrinkingwaterstandardsarediscussedinaqualitativemannerinstead.Similarly,giventhelimited dataonsoilpollutantconcentrations,onlyaqualitativeevaluationwasperformedforpossiblehuman healtheffectsfromexposuretocontaminatedsoil.
xvii
Severalrecommendationsareofferedbasedontheriskassessment.Therecommendations,detailed in the report and summarized at the end of the Executive Summary, focus on filling in data gaps in ordertomakeabetterevaluationofrisk,andontheuseofbestmanagementpracticessuchasgreen completions. PARTIIHEALTHASSESSMENT Twoquestionsguideddatacollectionforthispartofthestudy: 1. Is the health of residents of Garfield County different than the health of residents of Delta, MesaorMontrosecounties? 2. IsthehealthofresidentsofareasofGarfieldCountythatareheavilyimpactedbythenatural gasindustrydifferentfromthehealthofresidentsoflessimpactedareasofGarfieldCounty? Toanswerthesequestions,datawerecollectedinfourways: Public perceptions, concerns and experiences were identified through focus groups, public meetings,interviewswithkeyinformants,andbyreviewinglogsofcomplaintsthathadbeen receivedbytheGarfieldCountyHealthDepartment. Quantitative health indicators and outcomes were obtained through two complementary methods: o collecting and analyzing health data that are reported annually to the Colorado DepartmentofPublicHealthandEnvironmentandareavailablebycounty,and o collectingandanalyzinghospitalandmedicalinsurancedataforthefourcountyregion. o Selfreported health outcomes and risk factor information was collected for a representative, random sample of residents from throughout Garfield County (by zip code),usingatelephoneormailbasedhouseholdsurvey. Thehealthassessmentwasintentionallybroadandcomprehensive,innature,andonemightaskwhy thestudylookedatdiseaseanddiseasesymptomsthatpresumablywouldnotberelatedtoexposures from natural gas industry operations. However, many health conditions have symptoms that are similarand/orhavemultiplecauses,anditisawellestablishedfactthatapersonsgeneralhealthcan influence susceptibility to toxins or disease agents. Thus, this study was designed to also look at factorsthatcouldcontributeto,confound,orexacerbatehealthconditionsorsymptoms(i.e.,lifestyle, health insurance, residence, occupational history, concurrent disease, etc.). This comprehensive approachalsoprovidesameansoflookingatcrosscountydifferencesthatmayormaynotberelated tonaturalgasindustryactivitiesthatoccurmoreintensivelyinsomepartsofGarfieldCountythanin otherparts. ThisstudyprovidesasnapshotintimeofthehealthofGarfieldCountyresidents.Itisapopulation based, descriptive study, providing correlations and comparisons. The nature of the study and the
xviii
available data make it impossible to provide definitivecausal relationships between observedhealth andexposures,particularlyattheindividuallevel.Itdoes,however,provideacomprehensivedataset that may be used as a starting place for monitoring health trends or more quickly identifying new trends. It also provides source data for more specific analyses, should other researchers or public health officials wish to delve more deeply into any aspect of the health outcomes described in this report. Perceptions and Concerns. Healthrelated perceptions and concerns that were expressed by individuals who participated in public meetings, focus groups, individual interviews or registered complaintswithGarfieldCountyEnvironmentalHealthcouldbegenerallygroupedandsummarizedas follows: PhysicalHealthIssues/Concerns (Note:Thefollowingconcernsmayormaynothavebeenassociated withenvironmentalexposures.) Increase in or exacerbations of allergies and asthma and related concerns such as coughing, wheezing,andotherrespiratorycomplaints Generalizedchemicalsensitivities Fibromyalgia/chronicpainandrelatedconcernssuchaschronicfatigueandlethargy Chroniccoldsandconcernaboutcompromisedimmunesystems Headaches, dizziness, burning/itching eyes, nausea/vomiting, sinus problems most often attributedtoodors Burning/itchingskin Mentalhealthissuessuchasstress,depression,anger,inabilitytosleep Cancer (adrenal cancer, brain tumors, unknown/presumed cancers or fear of developing cancer) Lossofvoiceorspeechproblems Trauma/workrelatedinjuries Agerelatedillnesses Diabetes Obesity Perceptions that preexisting health conditions have beenexacerbated; people feeling worse thaninthepast. Social/CommunityIssuesandConcerns Increaseinchildandspousalabuse;childneglect;stressedfamilyrelationships Alcoholabuse(especiallyamonghighschoolstudents) Drugabuse(especiallymethamphetamineuse) Highsuiciderate Increaseinsexuallytransmitteddiseasesrelatedtoincreaseintemporaryworkers Lack of health insurance and related concerns such as lack of dental care for children and preventivecare Lackofaccesstohealthcareandmentalhealthservices Numberoflowincomefamilies
xix
Growthissuessuchastheavailabilityofhousingandcommunityservices,increaseinlowincome families,culturalclash(longtimeresidents,industryworkers),traffic,publicsafety EnvironmentalConcerns Noise Odors Dust Toxicchemicalsinairandwater Impactsondomesticanimalsandwildlifesuchasreportedchangesinherdanimalreproductive patterns/illnessesanddecreasesinbird,insect,anddeerpopulations. Quantitative Health Data. Health outcomes data were collected from a number of sources for Garfield County and comparison counties. The search for health outcomes data was driven by two objectives:1)tocompleteasnapshotintimepictureofthehealthofGarfieldCountyresidentsin comparisontothehealthofresidentsinthecomparisoncounties,and2)toobtainstatisticaldatathat couldbeusedtorespondtotheconcernsvoicedbyGarfieldCountyresidentsduringthequalitative data collection process. Thus, to the extent that the data were available, we collected statistical informationontheprevalenceofconditionssuchascancerandasthmaandthepredominantcausesof mortalityandmorbidityinGarfieldCounty. The following data were obtained from the Colorado Department of Public Health and Environment (CDPHE;datasetsrepresentthemostcurrentdataavailableatthetimethisreportwaswritten): Deathstatistics:ratesandleadingcausesofdeath(19902006) Birthdefects:typesandrates(20002006) Adolescenthealthmeasures(20002005) Reportableconditions(19982006) WestNilevirus(20022007) Cancerstatistics(19922005) BehavioralRiskFactorStudySurvey(BRFSS)data(20002005) Generalhealthstatus(physical&mental) Diabetes,asthma Smoking,weight Healthinsurance Injuryhospitalizationanddeath:causesandrates(20012003) Hospitalandoutpatientdatawereobtainedfromthefollowingsources: Colorado Hospital Association (CHA), Diagnosis Related Group (DRG)based hospital discharge data(2000through1stquarter2006) Emergencyroomdata GrandRiverMedicalCenter(locatedinRifle,COandservingwesternGarfieldCounty) ValleyViewHospital(locatedinGlenwoodSpringsandservingeasternGarfieldCounty) RockyMountainHealthPlans(RMHP)memberdatafor4counties St.MarysHospitalCareFlightdata
xx
RMHP hospital inpatient, outpatient and ambulatory member data 4 county comparison by DRGcategory OutcomesHighlights. Pleaseseethefullreportforadditionalmeasuresandcomparisonsandactualrates.Inthefollowing section, references are made to Garfield County disease or condition rates being lower, higher or similartothoseofDelta,MesaandMontroseCounties.Forthefollowingreasons,wespecificallydid not attempt to designate whether or not the rates themselves should be considered high or low. First,whetherornotarateishighorlowvarieswiththediseaseorcondition,andhastodowithwhat theexpectedfrequencyofthatconditionwouldbeforaparticularpopulation(andhencewhetheror nottheactualfrequencyishigherorlowerthanexpected).Secondly,whetherarateishighorlowis oftenamatterofindividualperceptionandmaydependonwhyanindividualislookingattheratein question.Thus,forthepurposesofthisreport,themostobjectivewaytoreportratesistoprovidethe actualnumericalrate(inthefullreport)andtoproviderelativecomparisonswithothercommunities. Thosewhousethisinformationmayhaveothermeasuresagainstwhichtojudgewhetherornotthe rateofaparticulardiseaseorconditionisacceptablylowortoohigh. Cardiovasculardiseaseisthebiggestcauseofdeathinallfourcounties,followedbyotherheart relateddiseaseandcancer.Theratesamongthefourcountiesaresimilar. o The ageadjusted, total death rate for Garfield County was comparable to the age adjusted,totaldeathrateforDeltaCounty,andlowerthanthoseforMesaorMontrose Countiesovertheperiodof19902006. o GarfieldCountyscrudedeathratehasbeenthelowestamongthefourcountiesforthe past2decades. o GarfieldCountysrateofneonatalandinfantdeathswassimilartothatofMesaCounty, andhigherthanthatofDeltaandMontroseCounties. o Injurydeathrates(CDPHEdata)for20012003weresignificantlyhigherinGarfield CountythanforthestateoverallandforMesaCounty. Birth defect rates in Garfield County were not different from those seen in the comparison countiesfortheyears20022006. For selected measuresof child and adolescent health, Garfield County had lower rates of teen pregnancy, teen suicide and child abuse than were documented for the other three counties overthetimeperiodssurveyed.ChilddeathswerehigherinGarfieldCountythanintheother counties. According to the Behavioral Risk Factor Study Survey (BRFSS) data for 2000 2005, the percentageofGarfieldCountyresidentsthathavebeendiagnosedwithdiabetesorasthma,are smokersorareoverweight/obese,andhavehealthinsuranceiseitherlowerorsimilartothe percentages in Delta, Mesa and Montrose Counties. (Please see table below.) However, in everycase,theconfidencelimitsforGarfieldCountyareverybroad,indicatingthatthereare widevariationswithinthecountyonthesemeasures.(The95%confidencelimitistherange withinwhichwecanbe95%certainthattheactualratefortheentirepopulationofthecounty
xxi
not just the rate for the individuals sampled in the survey will fall.) This observation is supportedbythehouseholdsurveydata.Forexample,tobaccouseishigheramongsurveyed residentsofzipcodes81635,81650and81652thanamongresidentsofzipcodes81601,81623 and 81635. The percentage of individuals with health insurance is lower among surveyed residents of zip code areas 81623, 81647, 81650 and 81652 and primarily Spanishspeaking householdsthanitisamongresidentsofzipcodeareas81601and81635.
xxii
DeltaCounty 95% Topic % CI* Diagnosedwith 0.0 diabetes 2.8 5.8 14.1 Currentsmoker 23.9 33.7 Currentlyhavehealth 56.8 insurance 67.6 78.4 7.9 Everhadasthma 17.2 26.5 49.2 Overweight 60.7 72.2 BMI**25.029.9 5.4 Obese 14.1 22.7
BMI**>30
Garfield County % 2.6 18.7 79.2 16 41.8 12.6
95% CI* 0.0 5.1 9.6 27.7 20.8 87.6 8.1 23.9 31.1 52.5 4.9 20.2
MesaCounty 95% % CI* 5.4 20.7 80.1 10.6 57.3 20.9
3.07.8 15.2 26.3 74.3 85.9 7.0 14.2 50.8 63.8 15.5 26.3
Montrose County 95% % CI* 1.6 6.4 11.2 12.5 21.8 31.1 60.3 71.3 82.3 5.2 15 24.7 44.8 56.3 67.8 12.2 21.3 30.3

Colorado 95% % CI* 4.6 19.9 84.4 17.2 60.7 14.1
3.45.0 18.9 20.9 83.4 85.3 7.9 26.5 49.2 72.2 5.4 22.7
*ConfidenceInterval **BodyMassindex(BMI)isdefinedasweightinkilogramsdividedbyheightinmeterssquared Percentsareweightedtothetotalpopulation. Duringthesixyearperiod,20002005,withtheexceptionof2003,circulationdisorders andboneandjointdisorderswereeitherthefirstorsecondmostcommonreasonsfor hospitalization (in 2003, birthing and pregnancy disorders was number one, and circulation disorders was third). Respiratory disorders were either the fourth or fifth most common reason for hospitalization in all years; nervous system disorders were rankedeitherseventhoreightheachyear.Digestivedisordersrankedthirdinfourout ofthesixyears.Neonataldisordersweresixthinfrequencyduringtheentiresixyear period. Injury hospitalization rates obtained from CDPHE for 20012003 showed that Garfield Countys rates were similar to those for Colorado overall, and lower than those documented for Delta and Montrose Counties. These data correlated with trauma hospitalization data obtained from the Colorado Hospital Association and Rocky Mountain Health Plans. According to CHA and RMHP data, Garfield Countys injury hospitalizationratesincreasedforadults(>18years)after2005,butstillremainedlower thantheratesforMesaandDeltaCounties.Accident,injuryandtraumahospitalization
xxiii
ratesforchildren(<18years)residinginGarfieldCountyarethelowestamongthefour counties. The following tables provide a summary of data for hospitalization, outpatient and emergency room visits by diagnosisrelated categories for Garfield County, in comparisonwithDelta,MesaandMontroseCounties.Thedatawereobtainedfromthe Colorado Hospital Association, Rocky Mountain Health Plans (member data), Grand RiverHospitalDistrict(locatedinRifle;dataforresidentsofRifle,Parachute/Battlement Mesa and Silt only), and Valley View Hospital (located in Glenwood Springs).
xxiv
CommunityHealthRiskAnalysisofOilandGasIndustryImpactsinGarfieldCounty CHAdata (20001stQ2007) GarfieldCountys injuryhospitalization ratesincreasedfor adults(>18years) after2005,butstill remainedlowerthan theratesforMesa andDeltaCounties. Accident,injuryand trauma hospitalizationrates forchildren(<18 years)residingin GarfieldCountyare thelowestamongthe fourcounties. Notapplicable. RMHP(20002007) Outpatient/Ambulatory EmergencyRoom Sameasforinpatient Adultemergency visitrates. roomvisitsrelatedto accident,injuryor traumaincreased from2003through 2005,thenslightly decreasedthrough 2007. GRHDER (20042006) Emergencyroomvisit dataforGrandRiver HospitalDistrictshow increasingratesfor Silt, Parachute/Battlement Mesa,andRifle residentsforthe period20042006. ValleyViewER* (20042006) AdultERvisitrates werestable throughoutthe period.Ratesfor children/youngadults decreasedbetween 2004and2005,then remainedstablefor therestoftheperiod.
DRGCategory
Accident,Injury, Trauma
InpatientHospital GarfieldCountys rateswerethelowest amongthefour countiesuntil2005 forbothadultsand children,and remainedthelowest forchildrenthrough 2007.
Notapplicable.
AnnualPhysicals/Well ChildCheckups
Birthing,Pregnancy, Reproductive,and Neonatal
Ratesareconsistently higherforGarfield Countythanfor Delta,Mesaand MontroseCounties forbirthingand pregnancydisorders, butlowerformale andfemale reproductiveand neonataldisorders.
GarfieldCounty residentshadlower utilizationratesthan didresidentsofDelta, MesaandMontrose Counties.Thisis likelyareflectionof thedifferencein utilizationpatternsof insuredanduninsured populations.
GarfieldCounty'srates arethelowestamong thefourcounties. Note:TheseareRMHP dataforinsured individuals! Sameasforinpatient visits.
Notapplicable.
Notapplicable.
Notapplicable.
Sameasforinpatient visits.
Datashowan increaseinbirthing andpregnancy relatedpatientvisits for2005through 2006forSilt, Parachute/Battlement MesaandRifle residents.
AdultERvisitrates werestableoverthe period;ERvisitrates forthiscategory decreasedamong youngadults.ER visitsforconditionsin theperinatalperiod increasedthroughthe timeperiod.
Abbreviations: DRG = Diagnosis-Related Group; CHA = Colorado Hospital Association; RMHP = Rocky Mountain Health Plans; GRHD = Grand River Hospital District. *Data presented for Glenwood Springs zip code areas only.
xxv
DRGCategory
CHAdata (20001stQ2007) GarfieldCounty's ratesarethelowest amongthefour counties.
Circulation/Cardiac
InpatientHospital GarfieldCounty's ratesarethelowest amongthefour countiesforboth adultsandchildren. Ratesshowan increasingtrendfrom 20032006.
RMHP(20002007) Outpatient/Ambulatory EmergencyRoom Sameasforinpatient Sameasforinpatient visits. visits.
GRHDER (20042006) RatesforSilt, Parachute/Battlement MesaandRifle residentsshowan increasingtrendfor thetimeperiod.
ValleyViewER* (20042006) AdultERvisitsfor thiscategorywere stableoverthe period;child/young adultvisits decreased.
Endocrine/Metabolic
Overall,Garfield County hospitalizationrates forthesedisorders arenotdifferentthan theratesforthe otherthreecounties. However,diabetes related hospitalizationrates forbothadultsand childreninGarfield Countyarethelowest amongthefour counties.
Ratesareconsistent withtheCHAdatafor bothadultsand children.
SameasforERvisits.
ERvisitratesfor thesedisordersare generallylowerfor bothadultandchild residentsofGarfield Countythanforthe otherthreecounties; ratesshowaslight increasingtrendover thetimeperiod.
RatesforSilt, Parachute/Battlement MesaandRifle residentsshowan increasingtrendfor thetimeperiod.
AdultERvisits decreasedbetween 2004and2005,then remainedstablefor theremainderofthe period.Child/young adultratesincreased between2004and 2005,thendecreased from2005through 2006.
xxvi
CommunityHealthRiskAnalysisofOilandGasIndustryImpactsinGarfieldCounty DRGCategory CHAdata (20001stQ2007) GarfieldCountyhas thehighestrate amongthefour countiesforotitis media,upper respiratory infections,bronchitis, andasthmain children.Asthma hospitalizationshave beenhigherthanin eitherMesaorDelta Countiessince1993, butwerelowerthan inMontroseCounty fortheperiod,1993 2001.Garfieldadult hospitalizationrates fortheseconditions werelowestamong thecounties,aswere theratesforCOPD andotherrespiratory infectionsand inflammationsfor bothadultsand children.Exceptions aresimple pneumoniaand pleurisyinchildren, which,until2005, werehighestamong thefourcounties. RMHP(20002007) Outpatient/Ambulatory GarfieldCounty outpatient/ambulatory visitsforadultswere thelowestamongthe fourcounties,but highestamongthefour countiesforchildren. GarfieldCountyrates decreasedforboth adultsandchildren from2005through 2007. GRHDER (20042006) ERvisitratesforSilt, Parachute/Battlement Mesa,andRifle residentsarevariable, butgenerally increasingoverthe timeperiod. ValleyViewER* (20042006) Adultvisitsforotitis mediaincreased between2004and 2005,thendecreased from2005through 2006.Ratesfor respiratory conditions,ingeneral, andasthma, specifically,were stablethroughoutthe timeperiod. Child/youngadultER visitsforotitismedia decreasedsteadily throughthetime period.Ratesfor respiratory conditions,overall, werestable,however asthmarates increased.
InpatientHospital Hospitalizationrates foradultENTand respiratory conditionswerethe lowestamongthe fourcountiesfor adultsandsimilarto theothercounties forchildren.
EmergencyRoom GarfieldCountyER visitratesforadults werethelowest amongthefour countiesandwere stableoverthe8year period.Ratesfor childreninGarfield Countyincrease steadilyfrom2000 2003,butaresimilar tothosefortheother threecounties.
Ear,Nose,Throat, Respiratory
CHAdata (20001stQ2007) Nodataforthis category. RMHP(20002007) Outpatient/Ambulatory EmergencyRoom Sameasfor Sameasfor hospitalizationrates. hospitalizationrates.
xxvii GRHDER (20042006) RatesforSilt increasedsharply between2004and 2005;rateincreases forParachuteand Rifleweremore gradualoverthesame period.Rifle'srates continuedtoincrease through2006,while ParachuteandSilts ratesdecreased. RatesforSilt, Parachute/Battlement MesaandRifle residentsshowan increasingtrendfor thetimeperiod. ValleyViewER* (20042006) AdultERvisitswere lowandstable throughoutthe period.Child/young adultratesshoweda steadydecreaseover thetimeperiod.
DRGCategory
InpatientHospital GarfieldCounty's ratesarethelowest amongthefour countiesforboth adultsandchildren.
Eye
Gastrointestinal/Urinary
GarfieldCounty's ratesareeitherthe lowestamongthe fourcounties,orare similartothoseof theothercounties forthisgroupof disorders.
Ratesareconsistent withtheCHAdata forbothadultsand children.
ERvisitratesfor thesedisordersare lowerforGarfield Countyadultsthan fortheotherthree counties,andalso generallylowerfor children,butincrease overthetimeperiod. ERratesforGarfield Countyadultswere inconsistent,without asustainedtrendin eitherdirection,but similartoother counties.ChildER visitsforinfections weregenerally higherthanother counties.
AdultERvisitsfor thiscategorywere stablethroughout theperiod. Child/youngadult visitsincreased between2004and 2005,then decreasedbetween 2005and2006. AdultERvisitsfor infectionsdecreased throughoutthe period,asdid child/youngadult visits.
Nodataforthis category.
GarfieldCountyrates aresimilartothose fortheother counties.
Infection
GarfieldCounty outpatient/ambulatory visitsforadultswere thelowestamongthe fourcounties,but highestamongthefour countiesforchildren duringthetimeperiod 2000through2002.
Ratesforinfection relatedERvisitswere variableoverthetime period,without consistenttrendsfor thecommunitiesof Parachute/Battlement Mesa,SiltandRifle.
xxviii
CommunityHealthRiskAnalysisofOilandGasIndustryImpactsinGarfieldCounty CHAdata (20001stQ2007) GarfieldCounty's ratesrethelowest amongthefour counties. RMHP(20002007) Outpatient/Ambulatory EmergencyRoom Sameasfor Sameasfor hospitalizationrates. hospitalizationrates. GRHDER (20042006) RatesforRifleandSilt showagradual increase;ratesfor Parachute/Battlement Mesaincreased steeplybetween2005 and2006. ValleyViewER* (20042006) OveralladultERvisit ratesdecreased between2004and 2005,thenincreased to2004levels between2005and 2006.Drug/alcohol relatedvisitsshowed asteadyincrease overtheperiod.The samepatternwas seenforchildren andyoungadults. AdultERvisitswere stableoverthetime period,aswere child/youngadult visits. AdultERvisits, includingthose relatedtoheadaches ormigraines,were stableoverthe period.Child/young adultvisitsoverall increasedbetween 2004and2005,then decreasedthrough 2006.Visitsrelated toheadachesor migrainesincreased.
DRGCategory
Mental
InpatientHospital GarfieldCountyrates arelowerthanfor theothercounties andsteadily decreasingoverthe timeperiod.
Musculoskeletal
GarfieldCounty's ratesarethelowest amongthefour counties. Overall,Garfield hospitalizationrates forthesedisorders arelowerthanfor theothercounties. Forthesubcategory ofseizureand headache,ratesfor childreninGarfield Countyhave increasedsince2004, whiledecreasingin adults.
Ratesareconsistent withtheCHAdata forbothadultsand children. Ratesareconsistent withtheCHAdata forbothadultsand children.
GarfieldCountyER visitrates,overall, werethelowest amongthefour counties.
NervousSystem
CHAdata RMHP(20002007) (20001stQ2007) InpatientHospital Nodataforthis Ratesforthese conditionsforDelta category. Countywere consistentlythe highestamongthe fourcounties. GarfieldCounty's rateswerenot differentfromthose ofMesaand MontroseCounties.
xxix GRHDER (20042006) Nodataforthis category. ValleyViewER* (20042006) Includedwith Gastrointestinal/urinary category.
DRGCategory
Outpatient/Ambulatory EmergencyRoom Nodataforthis Nodataforthis category. category.
Pancreas/Liver
RedCell/Clotting
GarfieldCounty's ratessteadily decreasedfrom 2001through2006. Ratesforthese conditionsforDelta Countywere consistentlythe highestamongthe fourcounties. GarfieldCounty's rateswerenot differentfromthose ofMesaand MontroseCounties.
Skin/Allergy
GarfieldCounty's ratesarethelowest amongthecounties, showingperiodsof moderaterate decrease(2001 2003)andincrease (20032005).
Garfield County's rates are the lowest among the four counties, increasing slightly from 2003 to 2004, then decreasing from 2005 through2007.
Garfieldratesare lowestamongthe counties,andwere generallystable throughoutthe8 yearperiod,except foranincrease between2003and 2004,decreasingto previousratesin 2006.(Ratesfor childreninGarfield Countydecreased steadilyfrom2005 through2007.)
AdultERratesincreased between2004and2005. Child/youngadultrates werestablethrough 2005,thendecreased through2006. RatesforSilt, AdultERrateswere Parachute/Battlement stable.Child/youngadult MesaandRifle ratesdecreasedthrough residentsshowan theperiod. increasingtrendfor thetimeperiod.
xxx
Cancer Oneofthedifficultiesofobservingrelativelyrecenttrendsincancerincidencestatistics andtryingtodrawconclusionsaboutpossiblechangesinriskfactorsisthat,withrespect tocarcinogenicexposures,thesetrendsreflecteventsthathappened1020yearsagoor arecumulativeoveralifetime.Generallyspeaking,theappearanceofclinicalcancerhas a lag time of up to two decades following initiation of carcinogenesis. (Childhood cancersandsomerarecancersareexceptions.)Thus,withtheexceptionofchangesin cancerscreeningpracticesthatoftenartifactuallyinflatecancerrates,shorttermtrends maynotreflectchangesinthepotentialforexposuretocarcinogenicmaterials. CountyspecificcancerratesforDelta,Garfield,MesaandMontrosecountiesforthetime periodof1992through2005showthatincidenceratesforallcancershavechangedlittle overthisperiod,althoughcancerratesinmalesdroppedslightlyforDelta,Garfield,and Montrosecountiesforthemostrecenttimeperiodforwhichdataareavailable(2003 2005).Thereisrelativelylittledifferenceamongthecountiesforeithermaleorfemale cancerincidenceratesacrossthedesignatedtimeperiods. Whencomparedtothestatecancerrates,thefollowingobservationscouldbemadefor GarfieldCounty: o Theincidencerateofallcancerscombinedwassignificantlyhigherthanthestaterate formalesfrom1992through2000andforfemalesfrom1992through1998. o Theincidencerateofallcancerscombinedwassignificantlylowerthanthestaterate forfemalesfrom1999through2000. o Theincidencerateofprostatecancerwassignificantlyhigherthanthestateratefrom 1992through2000. o Incidence rates for colorectal cancer, lung cancer, melanoma, bladder cancer, leukemias,andthyroidcancerinbothmalesandfemales,andforbreastandcervical cancerinfemales,didnotdiffersignificantlyfromthestateratesfortheperiod1992 through2005. Forcomparison,thefollowingobservationsweremadeforMesaCounty: o Theincidencerateofallcancerscombinedwassignificantlyhigherthanthestaterate formalesfrom1992through2005. o Theincidencerateofprostatecancerwassignificantlyhigherthanthestateratefrom 1999 through 2005, but was significantly lower than the state rate from 1992 through1998. o The incidence rate of lung cancer was significantly higher than the state rate from 1992through2000andfrom2003through2005forbothmalesandfemales.
xxxi
o Theincidencerateofmelanomainmaleswassignificantlyhigherthanthestaterate from1999through2000. o Theincidencerateofbladdercancerinfemaleswassignificantlyhigherthanthestate ratefrom2001through2002. o Theincidenceofthyroidcancerinfemaleswassignificantlylowerthanthestaterate from2003through2005. o Incidence rates for colorectal cancer, and leukemias in both males and females, and formelanoma,thyroidcancer,breastcancerandcervicalcancerinfemales,didnot differsignificantlyfromthestateratesfortheperiod1992through2005. SexuallyTransmittedDiseases(STDs) Of the conditions that are reportable to the state health department, other than cancer, a recenttrendinthefrequencyofSTDdiagnosesisworthnoting: Forthetimeperiod,2003through2007,therateofreportedcasesofChlamydiainboth GarfieldandMesaCountieshassteadilyincreased.Itisnotclear,atthistime,whether thisrateincreaseisduetoanactualincreaseindiseaseorisanartifactofchangesin screeningpatternsinthetwocounties.Overall,GarfieldCountysratesarelowerthan those for Mesa County and the state, but higher than those for Delta and Montrose Counties. Both Delta and Montrose Counties show an increase in reported cases for 20042005, but then the rates are flat (Delta) or decrease (Montrose) from 2005 through2007. Similarly,therateofreportedcasesofGonorrheainGarfieldandMesaCountiessteadily increasedfrom2003through2007.GarfieldCountysratesareagainlowerthanthose for both Mesa County and the state, but higher than those for Delta and Montrose Counties.DeltaCountysGonorrheacasesshowedthesamepatternasforChlamydia; Montrose Countys Gonorrhea cases increased slightly over the period 2004 through 2007. GarfieldCountyhadthehighestrateofreportedHIVcasesamongthefourcountiesfor theyears2000,2001,2003,and2005.Theserateswerehigherthanthestateratefor theyears2001and2003.(Therewerenocasesreportedin2002,2006,or2007.) GarfieldCountyhadthehighestrateofreportedAIDScasesamongthefourcountiesfor theyears2000,2003,2004,and2005.(Therewerenocasesreportedin2001,2002,or 2006.)
xxxii
SelfReportedHealthStatus:HouseholdSurvey.Atargetedhealthsurveywasadministeredto GarfieldCountyresidentsbytrainedinterviewers.Theinhomesurveyscapturedinformation about the general health and health risk factors of residents, as well as information about specific health conditions that were identified as priority concerns during focus group discussions.Thisinformationwasintendedtoprovideamoreobjectivemeasureofthehealth statusofcommunityresidents,andameansofconductingwithinCountycomparisons. Wecollecteddataon1,048individuals,representing~2%ofthehouseholdsinGarfieldCounty that have listed telephone numbers. This number included 49 interviews, conducted in Spanish,ofprimarilySpanishspeakinghouseholds(representing195individuals).Respondents wereaskedtoprovideinformationforeveryindividuallivinginthehousehold.Thus,wewere able to achieve a population sample that is representative of the gender, age, and ethnicity demographics of Garfield County, and approximated the occupational, education, and household income diversity that exists in the county. Respondent households represented eachofthezipcodeareaswithinGarfieldCounty,withalowof1.25%andahighof2.52%of householdswithlistedphonenumbersinzipcodes81601and81652,respectively,completing thesurvey.(Zipcodeareas81601and81623aretheleastaffectedbynaturalgasdrillingand production activities; zip code areas 81635, 81647, 81650, and 81652 are most affected by theseindustryactivities.)ThemajorityofthoserespondingtothesurveyhadlivedinGarfield Countyforgreaterthan5years. HouseholdSurveyOutcomesHighlights (Pleaseseethefullreportforadditionalmeasuresand comparisons): Greaterthan80%ofindividualsfromeveryzipcodeareainGarfieldCountyratedtheir currenthealthaseitherexcellentorgood,andlessthan10%ofindividualsineveryzip codeareafeltthattheircurrenthealthissomewhatworseormuchworsethanitwas oneyearago. Approximately12%ofindividuals,countywide,reportedthattheyhadsufferedanillness orinjuryduringthepastyearthathadaffectedtheirhealthforgreaterthan5days. Approximately8%ofindividuals,countywide,reportedsufferingfromdepression. 20%ofindividualsreportedsufferingfromavarietyofallergies,includinghayfever. 8% of individuals suffer from frequent headaches or migraines; a zip code comparison shows that the lowest frequency of headache sufferers live in zip code 81623 (4%), whilethehighestfrequencywasreportedfromzipcodes81635and81647(8.3%each). 3.8% of individuals report living with diabetes and its side effects such as kidney problems, loss of feeling or pain in hands and feet, and eye problems. There was no differenceamongthezipcodeareasforfrequencyofdiabetes.
xxxiii
24%ofindividualshavecoronarydisease,acategorythatincludesheartattackorheart surgery,highbloodpressure,strokeandangina.Therewasnodifferenceamongthezip codeareasforfrequencyofcoronarydisease. 10% or less of children born in any zip code area of Garfield County were reported to havedevelopedhealthordevelopmentalproblemswithin5yearsoftheirbirth. 5%ofindividuals,countywide,reportedhavingsomekindofcancerduringtheirlifetime. ~2Xasmanyindividualsresidinginzipcodearea81635reportedhavingcancer thanwasreportedforthecountyoverall.Itisimportanttonote,however,that the average age of the respondents from this zip code area was considerably olderthanfortheotherzipcodeareas. 53.3%ofthereportedcancerswerediagnosedinindividualswhowere55years orolder;TherewereNOcancersreportedinindividualsunderage25. Themostfrequentlyreportedcancerswerefemalebreastcancer(20.7%),non melanomaskincancers(26.4%),prostatecancer(15.1%),cervicalcancer(9.4%), andcoloncancer(7.5%).Malignantmelanomaandlymphomaeachaccounted for 3.8% of the reported cancers. Uterine, thyroid, liver, kidney, and bladder cancers, along with leukemia, glandular carcinoma, made up the remainder of thecancersreported(1.9%each).
Because respiratory complaints were expressed so frequently in interviews and focus groups, we asked a number of specific questions about respiratory conditions and contributingfactorssuchassmoking. 6.5% of individuals, countywide, reported having a diagnosis of asthma; the highestfrequencyofindividualswithasthmawasinzipcodearea81647(8.3%), while the lowest frequency of individuals with asthma was in zip code area 81652(4.7%).Bothzipcodeareashavesignificantnaturalgasindustryactivity. o SimilartowhatwasobservedwithotherrespiratoryconditionssuchasChronic ObstructivePulmonaryDisease(COPD),emphysemaandotherlungorbreathing problems, the zip code area having the highest frequency of these conditions wasamongthosemosthighlyimpactedbynaturalgasindustryactivity,butthe lowest or next to lowest frequency was also found among these impacted zip codeareas. o Age and smoking are factors that clearly influence the incidence of these conditions. 27% of county residents, overall, reported having smoked at least 100 cigarettes during their lifetime; 60% of these have quit smoking. 85% of respondentsreportedthatsmokingisNOTallowedwithinthehome.
xxxiv
OccupationandDisease Responses to questions about occupational history (current and longest job titles and industry affiliations) allowed correlations with diseases and symptoms reported by surveyrespondents. Itisimportanttonotethatthisstudywasnotdesignedasanoccupationalhealthstudy, andthus,noconclusionsmaybedrawnregardingoccupationalexposuresanddisease outcomes. The numbers of individuals within any occupational category that report havingaparticulardiseaseorconditionaretoolowforstatisticalsignificance.However, thefollowingobservationsmaybemade: o Individualswhoreportedthattheircurrentand/orlongestoccupationwasinthe professionalandrelatedservicesindustries(e.g.,healthcareproviders,attorneys, etc.), personal services occupations (e.g., housekeepers, hair stylists, etc.), construction industries or transportation (including truck drivers), communicationsandpublicutilitiesindustriesweremostlikelytohavereported having respiratory conditions; neurological symptoms such as dizziness, numbness,weakness;skinproblems;andfrequentheadaches/migraines. o Thoseindividualswhorefusedtoanswerquestionsabouttheiroccupationand/or industry affiliation were most likely to have reported having frequent headaches/migraines; neurological symptoms such as dizziness, numbness, weakness; anemia; seizures; skin problems; and cancer (but no bladder, kidney, liver,lymphomaorthyroidcancersorleukemia).
RelationshipsBetweenHealthandEnvironmentalExposures:HouseholdMemberConcern Aseriesofquestionsregardingperceptionsofriskrelatedtohomeandoutsideenvironmental exposures and their relationship to health outcomes were asked at the end of the survey. These questions were intended to serve as measures of concern and perceptions among a randomly selected population within Garfield County (as opposed to the more selfselected populationthatprovidedcommentsduringfocusgroups,interviews,andpublicmeetings),and to provide some measure of the potential bias with which survey respondents might have respondedtoquestionsabouttheirhealth. When asked whether or not they are concerned that their home drinking water source was related to any of their health problems, between 5.5 and 17% of individuals who liveinareaswithhighnaturalgasindustryactivity(zipcodeareas81635,81647,81650, and81652)respondedthattheyareconcerned,whileonly35%ofindividualswholive in the areas least impacted by natural gas industry activity (zip code areas 81601 and 81623)respondedinthesamemanner. When asked whether or not they are concerned that their health problems may be relatedtochemicalsinorneartheirhomes,between6and16%ofindividualswholive inareaswithhighnaturalgasindustryactivity(zipcodeareas81635,81647, 81650,and 81652)respondedthattheyareconcerned;between1.4and7%ofindividualswholive
xxxv
in the areas least impacted by natural gas industry activity (zip code areas 81601 and 81623)respondedinthesamemanner. Whenaskediftheyareconcernedthateitherenvironmentalorchemicalhazardsintheir neighborhoodsmayberelatedtohealthproblems, o Between24and38%ofindividualsresidinginzipcodeareas81635,81647,81650, and 81652 responded that they are not worried at all. Between 76 and 62% of individualsinthesezipcodeareasrespondedthattheyarealittleworried,very muchworried,ordontknow/notsure. o Between 56 and 43% of individuals residing in zip code areas 81601 and 81623 respondedthattheyarenotworriedatall.Between44and57%ofindividualsin these zip code areas responded that they are a little worried, very much worried,ordontknow/notsure. o Countywide,individualswhohaveahighschooleducationorlessareslightlyless worried about the relationship between their health and environmental or chemicalhazardsintheirneighborhoods.
When asked specifically whether they are concerned that natural gas industry activities
mayberelatedtohealthproblems, o Between69and92%ofindividualsresidinginzipcodeareas81601and81623 respondedthattheyarenotconcern.Between8and31%ofindividualsinthese zip code areas responded either that they are concerned or that they dont knoworarenotsure. o 90% of individuals residing in zip code areas 81601 and 81623 responded that theyarenotworriedatall.10%ofindividualsinthesezipcodeareasresponded that they either that they are concerned or that they dont know or are not sure. o There was essentially no difference related to education between individuals who responded that they are concerned about healthrelated impacts of the naturalgasindustryandthosewhoarenotconcernedorarenotsure.
xxxvi
CONCLUSIONSANDRECOMMENDATIONS
Conclusions. AtthepresenttimebasedonourdatasourcesthereisnotahealthcrisisinGarfield County, but there are some health trends that should be monitored. We cannot say conclusively that any of these health trends are directly related to the presence of naturalgasindustryactivitiesortootherfactors. Accident, injury and trauma hospitalization and emergency room visits have been increasing,particularlyforadults. Child(114years)deathsinGarfieldCountyfor20012005areconsiderablyhigherthan the states 2010 goal (35.5/100,000 versus 19/100,000, respectively) and higher than theratesforMesaandMontroseCounties. AlthoughcirculationdisordersandcardiacdiseaseratesarelowerinGarfieldCountythan in the other three counties studied, these conditions have been among the most commonreasonsforhospitalization,andratesforinpatient,outpatientandemergency roomvisitsappeartobeincreasing. Hospitalizationratesforbirthing,pregnancyandgynecologicaldisordersareconsistently higher for Garfield County than for the other counties studied, and emergency room visits for these disorders appear to be increasing, at least among Silt, Parachute/BattlementMesaandRifleresidents. Upper respiratory infections, bronchitis, asthma and otitis media rates in children, in particular, have generally been higher for Garfield County than in the other counties studied.Emergencyroomvisitsfortheseconditionshavebeenincreasing. MentalhealthrelatedemergencyroomvisitsforresidentsofSilt,Parachute/Battlement MesaandRifleincreasedbetween2005and2006. HospitalizationratesforseizureandheadacheinchildreninGarfieldCountyshowedan increasing trend between 2004 and 2007. Emergency room visits for the general category of nervous system disorders showed an increasing trend for residents of Parachute/Battlement Mesa and Silt for the 20042006 time period. However, it is important to note, that overall, rates for nervous system disorders in Garfield County areaslowasorlowerthanthosesameratesinDelta,MesaandMontroseCounties. Although the actual number of cases is relatively low, the frequency of diagnosed and reported sexually transmitted diseases has increased steadily in Garfield County since 2003.
xxxvii
Atthistime,therearenocancertrendsinGarfieldCountythatareofnotableconcern. However,becauseofthelagperiodbetweenexposureandcancerdevelopment,cancer ratesshouldbereviewedonaperiodicbasis. Riskmodelingindicatedthatthereareindustryfactorsthatcouldpresentapublichealth riskuseofbestpracticescanreducethatrisk. Recommendations Establish a medical monitoring system especially through primary care networks to identifyanychangesinbaselinedataortrendsand/oranomaliesinmedicalpractices. Conductathoroughstudyofairemissionsduringdrilling,includingenoughsitestocover therangeofdrillingapproaches. o Collect 24hour samples daily around the perimeter of the drill pad to achieve continuousmonitoringduringseveralcyclesofwellinstallation. o Monitormeteorologicalconditions. Identifythecomponentsofhydraulicfracturingfluids. o Wouldallowopenevaluationofdegreeofthreat. o Wouldimprovepublicacceptanceofnaturalgasoperations. Inspectsurfacesoilsatcompletionofdrillingoperations. o Minimizepossibleexposureoflandownerstoresidualsoilcontamination. o Sampleandanalyzeareassuspectedtobecontaminated. o Cleanupareasexceedingactionlevels. Usegreencompletionsandapplicablebestmanagementpractices,includinglocating drilling and production facility operations far enough from public buildings and residencestoreducetheriskofexposuretoairtoxics,suchasbenzene,toluene,and xylenes. Establishamonitoringprogramforprivatewells o Provides the most direct way to assess contamination of drinking water resources o Analyze for methane, benzene and other volatile organic compounds, and selected components of hydraulic fracturing fluids having the greatest potentialtoaffecthumanhealth
INTRODUCTION TO THE STUDY
TheCommunityHealthRiskAnalysisofOilandGasIndustryImpactsinGarfieldCountywas aresponsetoarequestforproposalsforprojectsthatwouldaddresspublicenvironmental
needsinGarfieldCounty.TheseprojectsweretobefundedbyafineleviedonEnCanaOiland Gas (USA), Inc. as a result of a violation of Colorado Oil and Gas Conservation Commission (COGCC) rules. All proposals were due by August 30, 2004. The Garfield County Energy Advisory Board (EAB) discussed the proposals at a meeting on September 2, 2004, at which they made a decision to forward a recommendation to the Garfield County Board of Commissioners to fund three of the proposals. On September 6, 2004, the Garfield County Board of County Commissioners selected three proposals for funding, one of which was the Community Health Risk Analysis of Oil and Gas Industry Impacts in Garfield County. The initial work on the project was begun by Dr. Teresa Coons (St. Marys Saccomanno Research Institute) and Dr. Russell Walker (Environmental Sciences and Technology Department, Mesa State College) shortly thereafter, although the final Scope of Work and contract were not signedbytheBoardofCommissionersuntilDecember31,2006. il and gas activity within Garfield County has generated public concern with regard to impactsonboththeenvironmentandpublichealth.Often,publicperceptionofpotential healthrisksbecomesanoverridingconcern.Thepublicmayfeelasthoughtheirhealthisatrisk and they may wish to obtain an understanding of possible threats to their health, as well as acknowledgement by the oil and gas industries of their concerns. Opening this dialogue throughdiscussionoftherealandperceivedimpactsofextractiveindustriesandcouldserveas afocalpointforcollaborativeresolutionstocommunityandindustryconcernsoverthebalance betweentherisksandbenefitsofnaturalresourceextraction. FOCUSOFTHESTUDY
Exposure Potential (risk assessment)
Health Assessment (current health status)
Determine the type and magnitude of potential human exposures to toxins
Relationship between exposure and health
Figure 1. Relationship between Exposure PotentialandHealthAssessment
RiskAssessment Characterizethesources NaturalGasOperations Modelshowingpotentialpathwaysformovementofcontaminantsfromsourcetohuman exposure Othersourcesofpotentiallytoxicmaterials(e.g.mines,milltailings,landfills,agricultural spraying)? Characterizethecontaminants Collectandinterpretdataoncontaminantsandtheirconcentrations Summarizetoxicologicalinformationoncontaminants RiskCharacterization Developprobabilitystatementsaboutrisktoindividualswithinthecommunity(current orfuturerisk) Exposurepotential(proximitytosource,dose,etc.) Probabilityofsufferingharmfromexposure(incorporatesfactorsrelatedtoage,health status,etc.) HealthAssessment PerceptionofHealth How do community members feel about their health and the health of others in the community? Whatdocommunitymembersfeelarethedeterminantsofhealthintheircommunity? MatchingPerceptionwithReality Statisticalhealthdataandselfreportedhealthinformation MergingRiskandHealthData PointintimepictureofthehealthstatusofresidentsofGarfieldCounty Informationaboutdeterminantsofhealthpicture Relationshipsbetweenexposureandhealthstatus Recommendations Datagapsthatshouldbefilled Possiblefuturemonitoringneeds Optionsformanagingrisk
STUDYLIMITATIONS Funding:studymayuncoverissuesorareasthatneedmoreindepthresearchthancan beperformedwithexistingfunding Lackofbaselinehealthdatawithwhichtodeterminetrendsorchanges Relativelynewpresenceoftheindustryintheregion:healthoutcomesmaynotyetbe apparent Both risk and causation involve probability statements: may never be able to say with certaintythataparticularhealthconditioniscausedbyanexposuretoapotentiallytoxic material. STUDYAREA TheprimaryfocusofthishealthandriskassessmentstudyisGarfieldCounty.Mesa,Montrose, andDeltaCountieswereselectedascomparisoncountiesforthehealthassessmentportionof the study. These three counties share a Western Slope of Colorado location with Garfield County(andhence,somecommonpopulationdemographicsandculture).However,theyall, currently,havefewerimpactsfromnaturalgasindustrydrillingandprocessingactivities.
EXTRACTIVEINDUSTRIESINCOLORADOANDGARFIELDCOUNTY Thisstudywasinitiated,inlargepart,becauseoflocalconcernsabouthealthrelatedimpactsof the natural gas industry. However, it is important to put those impacts into a context that includesnaturalgasactivityinotherpartsofColorado(includingthecomparisoncounties),as wellasimpactsfromotherextractiveindustriesthatarenoworhaveinthepast,operatedin GarfieldCounty.
Natural Gas. ThenaturalgasindustryinGarfieldCountyhasgrownrapidly,overarelatively shortperiodoftime,bothinscopeandintensity.In1988,atotalof19permitswereissuedby the COGCC for drilling in Garfield County (www.oilgas.state.co.us/staffreports). By 2002, whichaccordingtosomeobserversmarkstheobservablebeginningofthecurrentboom,362 permitshadbeenissued.In2007,2550drillingpermitswereissued.Thefollowingtablesand graphs were obtained from the Colorado Oil and Gas Conservation Commission (COGCC) website:www.oilgas.state.co.us/.
Table1.AnnualpermitstodrillissuedbytheColoradoOilandGasConservationCommission(www.oil gas.state.co.us/staffreports).Thenumberofdrillingpermitsissueddoesnotfullyreflecttheimpacts)of the natural gas industry in a region. A more accurate reflection of industry activity is the number of wellsactuallybeingdrilled(asreflectedbythenumberofoperatingdrillrigs)andthenumberofactively producingwells.
Garfield County
Mesa Montrose County County
Delta W e ld La Cheyenne County County Plata County County

1 0 424 702 302 156 203 3
1988 2001
19 353
1 27
0 3
2002 2003
362 657
30 27
2 4
7 4
760 757
104 162
3 3
2004 2005
796 1509
54 136
2 0
5 10
832 901
102 115
3 10
2006 2007
1845 1550
265 293
1 3
9 2
1418 1527
235 251
21 15
Figure2.
Figure3.
Coal Mining. Other current and historic mining activities in Garfield County include coal, uranium,andgravelmining.TheMcClaneCanyonundergroundcoalmineislocatedinthefar westernpartofGarfieldCounty.CoalminingalsotakesplaceinDeltaandMontroseCounties, andhastakenplaceinMesaCountyintherecentpast. Figure4.LocationofCoalMines,PowerPlants,RailroadsandCoalbearingRegionsinColorado,2006.
ColoradoGeologicalSurvey,InformationSeries75,ColoradoMineralandEnergyIndustryActivities
Table2.ColoradoCoalMineStatistics,2006. ColoradoDivisionofReclamation,MiningandSafety.
Uranium mining. Beginningintheearlypartof the twentieth century, the Rifle area in Garfield County was host to underground vanadium and uranium mining and two processing mills. In 1991, the Bureau of Land Management transferred 205 acres of land in Estes Gulch (approximatelysixmilesnorthofthecityofRifle) totheDepartmentofEnergytouseasadisposal site for the tailings that had been produced and stored at the Old and New Rifle Uranium Mills. The alluvial aquifer iscontaminated by seepage fromtheformermilltailingspilesatbothformer mill sites. Because of the large dilution by the river(atleastafactorof30,000),contaminantsin alluvial groundwater discharging to the river are quicklydilutedtobackgroundconcentrationsand nomillrelatedcontaminationhasbeendetectedinsamplesofColoradoRiverwatercollected at or downstream from the sites. (Rifle, Colorado, Processing Sites and Disposal Site. U.S. DepartmentofEnergyOfficeofLegacyManagement.)
ProjectRulison.NamedaftertheruralcommunityofRulison,Colorado,ProjectRulisonwasa 43kiloton underground nuclear detonation that took place on September 10, 1969, about 8 milesSEofthetownofGrandValley,Colorado(nowParachute,Colorado)inGarfieldCounty.It was part of the Operation Mandrel weapons test series as well as the Operation Plowshare projectwhichexploredpeacefulusesofnuclearexplosions.ThepeacefulaimofProjectRulison wastodetermineifnaturalgascouldbeeasilyliberatedfromundergroundregions.Thetest, wasperformedbytheAtomicEnergyCommissionandtwocorporatepartners,CERGeonuclear and the Austral Oil Company, at the bottom of an 8,426 foot deep shaft. The blast was marginally successful in causing the gas to collect in the cavity and fissures produced by the bomb;however,thegaswastooradioactivetobesoldcommercially.Anothernucleardevice gasstimulationtest,calledRioBlanco,wasperformednearbyin1973. ThesurfaceofthesitebegantobecleanedupbytheDepartmentofEnergyinthe1970s,and wascompletedin1998.AbufferzoneputinplacebythestateofColoradostillexistsaround thearea.AJanuary2005reportbytheDOEstatedthatradioactivitylevelswerenormalatthe surfaceandingroundwater. OTHERENVIRONMENTALISSUES Althoughacompletediscussionisbeyondthescopeofthisreport,itisimportanttonotethat thereareotherlocalandregionalactivitiesthatcananddoimpactairqualityormaycontribute tohumanexposurestotoxicmaterials.Amongtheseareagriculturalactivities;seasonalopen burningtoclearditchesandpreparefieldsforplanting,andsprayingforweedsandinsectpests takes place in many areas of Garfield County. Gravel mining along rivers to support local construction and traffic along the I70 corridor and local roads contributes significantly to
airbornepollutants.Seasonalforestfiresandweatherinversionpatternsalsocontributetoair qualityissuesthatmayimpactthehealthofGarfieldCountyresidents.
PART 1. RISK ASSESSMENT

ANOVERVIEWOFNATURALGASOPERATIONSINGARFIELDCOUNTY ost of the natural gas extracted in the Piceance Basin comes from the Williams Fork Formation of the Mesa Verde Group (Figure 6). The gas occurs in sandstone deposits (oftenreferredtobygeologistsassandbodies)whichareisolatedfromeachotherbynon gasbearing,impermeableshale.Thesandbodiesarerelativelysmall,averagingaboutninefeet thickfromtoptobottomandabout500feetacross.Becauseoftheirlowpermeability,ithas not been economically feasible to recover gas from the sand bodies using conventional methods.However,recentadvancesinwelldrillingandstimulationtechnologycoupledwitha strong national market for natural gas has made the Williams Fork Formation the focus of intensedevelopment.
SANDSTONE
Figure6.PiceanceBasinWilliamsForkFormation Wells are commonly drilled to depths between 4,000 and 9,000 feet below the land surface. Over the last few years, directional drilling has emerged as the dominant approach. This technology allows the drillers to manipulate the direction of the borehole and target sand bodies up to 2,500 feet horizontally away from the drill rig. Each sand body must be penetrated vertically, and each well is designed to access several vertically stacked bodies. Twelveormorewellsarecommonlyinstalledfromthesamewellpad,reducingthedisturbance associatedwithmultiplepads. A special mud is circulated through the borehole during drilling. The mud is pumped down through the center of the drill stem pipe into the borehole, where it flows back to the land
10
surface.Themudlubricatesthedrillbit,conveyscuttingsuptothesurface,andbyvirtueofits weightpreventsblowoutsofgasescapinguncontrollablyupwardthroughtheborehole.After returning to the surface, the mud is passed through a shaker to separate out rock fragments and is thenstored in reserve pits,which mayor may notbe lined. In recent years there has beenincreasinguseofclosedloopsystemsthatdonotrequireareservepit. Thewellitselfisconstructedaccordingtospecificrequirements.A16to20inchdiameterpipe or conductor casing is installed to a depth of 20 to 50 feet. This casing stabilizes the uppermostportionoftheboreholeandfacilitatesflowofthedrillingmud.Asurfacecasing about10inchesindiameterisinstalledthroughtheconductorcasing.Thiscasingisolatesthe natural gas, drilling mud, and other materials from shallow ground water so as not to contaminateasourceofdrinkingwater.Thesurfacecasingisrequiredtoextendfromtheland surface through at least 10% of the depth of the well, typically ending at a depth of 600 to 1,000 feet. The space between the casing and the borehole wall is filled with cement to preventthemigrationofnaturalgasordrillingfluidsupthroughthisspace,thuspreventingthe contaminationofgroundwaterthatisused(ormaybeusedinthefuture)asdrinkingwater. The production casing is a pipe about 5 inches in diameter that is installed through the surfacecasing all the way down to the sand bodies. This casing is also cemented in place to preventupwardmigrationofgasandfluids. Hydraulic fracturing (fracing) is a key step in the development of a viable well. The productioncasingisperforatedatlocationsthatwillallowaccesstothesandbodiespenetrated bythewell.Waterwithproprietaryadditivesispumpedthroughtheperforationsintothesand bodiesatapressureofabout10,000poundspersquareinch.Thepressureinducesfractures, which are propped open with sand or other solids that are part of the fracing fluid. These fractures provide a pathway for natural gas to flow out of the sand body into the well more easilythanitcouldifthesandbodywasleftintact.Thefracturescanextendupto1,500feet awayfromthewell.Itisimportanttonotethatpriortorecentyears,dieselfuelwasroutinely usedasthebaseforthefracingfluidratherthanwater. Afterfracing,wellsundergoacompletionphase.Thewellheadassemblyisinstalledandthe wellisplumbedtoapipeline.Anotablepartofcompletionisaflowbackperiodofupto30 days, in which gas is allowed to vent to the atmosphere in order to remove mud, sand, and other impurities from the well. Sometimes this gas is burned in a flare. In recent years, operatorsareincreasingtheiruseofgreencompletions,inwhichthenaturalgasiscaptured ratherthanvented. During the routine production phase, natural gas recovered from the well is piped to a gas processingplantforpurificationanddistribution.Liquidhydrocarbonsalsoflowoutofthewell. Thiscondensateisseparatedfromthegasstreamandcollectedinatankatthewellhead. The tank is periodically emptied into trucks and the condensate is transported to refineries. Dehydrators are also installed at the well head for removing water from the gas stream. WilliamsForkwellshaveanexpectedlifetimeofabout30years.
11
THERISKASSESSMENTPROCESS Conceptual model for pollutant migration and human exposure. A conceptual model is a usefulstartingpointforstudyinghumanexposuretopollutantsreleasedintotheenvironment. The conceptual model serves as a roadmap showing the routes by which a pollutant may comeintocontactwithpeople.Thisroadmapbecomesthebasisforevaluatingpossiblehealth threatsandidentifyingwaystomitigatethosethreats. The conceptual model breaks down the exposure process into a pollutant source, release mechanism,exposuremedium,exposureroute,andreceptor. Thesourcecanbeviewedaseitheradevice(suchasaninternalcombustionengine),a facility (such as an unlined pond for hydrocarboncontaminated water), or a process (suchasrefueling)fromwhichpollutantsarereleasedintotheenvironment. Areleasemechanismisthespecificmannerinwhichthepollutantescapes.Exhaustgases areventedoutoftheengine;hydrocarbonsevaporatefromthepond,andhydrocarbon contaminated water infiltrates into the soil below the unlined pond and possibly percolatesallthewaytogroundwater;fuelevaporatesintotheairduringrefuelingand infiltratesintothegroundifspillsoccur. Theexposuremediumistheenvironmentalmediumair,surfacewater,groundwater,or soilthat becomes polluted by the release and brings the pollutant into contact with people.Theengineexhaustcontaminatesair;thehydrocarbonsfromthepondwater contaminate air and possibly ground water; the hydrocarbons from refueling contaminateairand,ifspilled,thesurfacesoil. Theexposureroutedescribeshowapersonmaytakethepollutantintohisorherbody. Pollutantsinairareinhaled;pollutantsindrinkingwaterareingested;pollutantsinsoil areingestedinsmallamountsincidentaltoyardworkorspendingtimeoutsideondusty days;certainpollutantsinwaterorsoilmaybeabsorbedthroughtheskinwhenthesoil orwatercontactstheskin. Areceptorisapersonwhomakescontactwithapollutantthroughoneormoreofthe exposure routes. Receptors can be residents, workers, or visitors either at the site whereareleaseoccursoratnearbysitestowhichthepollutantsmigrate. Secondarysourcesmayoccurasaresultofaprimarysourceandrelease.Forexample, soil contaminated by a fuel spill becomes a secondary source of pollutant as the fuel hydrocarbonsundergoasecondreleasebyevaporatingfromthesoilintotheair. Onecanseethatlinkingasource,releasemechanism,exposuremedium,exposureroute,and receptor creates a roadmap by which a pollutant can travel from a particular source to a particular receptor. Exposure is actual contact of a pollutant with a receptor. If the journey between source and receptor is complete and exposure has occurred, we have an actual
12
exposurepathway.Inmanysituations,thejourneyisnotcompleteandthereisonlyapotential exposurepathway.Thesepotentialpathwaysmustbeevaluatedcarefullytoseeiftheyhavea significantlikelihoodofbecomingcompleteandcreatinganexposure.Oneconsiderationisthe properties of the pollutant of interest. Not all release mechanisms, exposure media, and exposure routes are relevant to each pollutant. For example, a pollutant with little or no tendency to evaporate will not occur as a gas in air, nor will a pollutant with little or no solubilityinwateroccurathighconcentrationsinwater. AgeneralconceptualmodelisshowninFigure7fornaturalgasoperationsinGarfieldCounty. Thereisnosingle,correctwaytoconstructsuchamodel.Thegeneralmodelshownisintended to be a representative rather than exhaustive roadmap of actual and potential exposures to pollutantsfromnaturalgasoperations. Drilling and completion operations can be viewed as one source. Releases from drilling and completion include venting of natural gas from the well and exhaust gases from engines associatedwiththedrillrigandsupportactivities.Thegasescontaminatetheonsiteairand canmigrateoffsite.Bothonsiteandoffsitereceptorscanbeexposed. Wellhead glycol dehydrators, condensate tanks, and condensate transfer operations may be viewedasanothersource.Hydrocarbonsindehydratorsandcondensatecanevaporateandbe vented to the atmosphere, where they can be inhaled by onsite or offsite receptors. Any hydrocarbon liquids that are leaked or spilled may evaporate as well. However, a portion of theseliquidswouldlikelyinfiltrateintothesoil,thenpossiblypercolatetothewatertableand contaminate ground water, which can migrate to neighboring properties where it may be capturedbyawell.Ifthiscontaminatedgroundwaterisusedfordrinking,thepollutantswill be ingested. If the water is used for showering, the hydrocarbons may evaporate into the shower air and be inhaled. Absorption of pollutants through the skin may also take place duringashower,ifanyofthechemicalspresenthaveasignificantabilitytobeabsorbedthat way. Anothersourceofinterestmaybeopen,unlinedpondsforproducedwater.Hydrocarbonswill evaporate from these ponds into the air, where they may be inhaled by onsite or offsite receptors. Hydrocarbons dissolved in water that infiltrates into the soil below the pond may undergothesamesequenceofeventsasdescribedaboveforhydrocarbonliquidsthatinfiltrate intosoilasaresultofspillsorleaks. There is considerable interest in any threat posed by hydraulic fracturing fluids. These fluids have the potential to escape from the well into a surrounding aquifer if the well casings are breached. They also have potential to escape from the target formation through induced fractures,naturallyoccurringfractures,oranyotherpreferentialflowpathsthatmayoccurin thesubsurface.Theresultingcontaminatedgroundwaterhasthepotentialtobeusedinthe samemannerasdescribedabove.
13
Someofthesepathwaysarehighlylikelytobecomplete,ifonlyonatransientbasis.Inhalation ofnaturalgasandofairpollutantsfromcombustionenginesisthebestexample.Thereislittle questionthatpollutantshavebeenemittedintotheair;thequestionishowmuchpollutantis presentatthelocationwhereareceptorbreathestheair.Otherpathwaysarelesslikelytobe complete.Thebestexampleisthepathwayforhydraulicfracturingfluids,whichcanrequire migrationupwardthroughthesubsurfaceforthousandsoffeet,arrivinginashallowaquiferat alocationincloseproximitytoawellusedfordrinkingwater.Thedetailsofpollutanttransport throughthepathwaysintheconceptualmodelarediscussedingreaterdetailinlatersections ofthisreport.
14
Figure7.
15
Risk assessment. The U.S. Environmental Protection Agency (EPA) developed a standard processforassessingrisktohumanhealthfrompollutantsintheenvironment.Theprocessis described in Risk Assessment Guidance for Superfund, Volume 1. Human Health Evaluation Manual(PartA)publishedbytheEPAin1989.Whiletailoredtositesbeinginvestigatedand cleanedupundertheComprehensiveEnvironmentalResponse,Compensation,andLiabilityAct (alsoknownasSuperfund),thebasicprinciplesandproceduresoftheguidanceareapplicable toanysituationinwhichpeopleareexposedtopollutants.Theoverallriskassessmentprocess isdividedintofourstepshazardidentification,exposureassessment,toxicityassessment,and riskcharacterization. HazardIdentification Thepurposeofthisstepistoidentifythosepollutantsthathavetheabilitytocauseharmto people,andsingleoutforfurtherconsiderationthosepollutantsthatarelikelytoresultinthe most significant exposures. Toxicity is an obvious key property, but the mobility and persistence of a pollutant are factors as well. Pollutants whose physical and chemical propertiesmakethemmorelikelytomigrateawayfromtheirsourceareconsideredtobemore ofathreattooffsitereceptors.Forexample,ahighwatersolubilityorastrongtendencyto evaporatemakeapollutantmoremobile.Lessmobilepollutantscanbemoreofathreattoon site receptors because the pollutants can remain near their source in high concentrations. Persistence refers to how long a pollutant remains unchanged in the environment. Many pollutants have the ability to be biodegraded (broken down into simpler, usually less toxic substancesbynaturallyoccurringbacteria),hydrolyzed(brokendownbyreactionwithwater), orphotodegraded(brokendownbyexposuretolight).Thesepollutantswillbelessofathreat becausetheyarelesspersistent.Pollutantdistributionisconsideredaswell.Ifmorethanone pollutantispresent,thosethatoccurinthemostsamplesandatthehighestconcentrationsare givenahigherpriority.Thosepollutantshavingthegreatesttoxicity,mobility,persistence,and distributionarecarriedforwardforfurtherstudy. ExposureAssessment Exposureassessmentisthesecondstepinriskassessment.Thegoalofthisstepistoquantify theexposureexperiencedbyareceptortoeachofthepollutantsofconcern.Theoutcomeof exposure assessment is a number for pollutant intake expressed as milligrams of pollutant absorbedperkilogramofbodyweightperday.Theequationsusedinexposureassessmentare showninAppendixA.Muchoftheessenceoftheequationsiseasytounderstand. Intuition suggests that the more contaminated the exposure medium, the greater the healththreatwillbe.Thedegreeofcontaminationisexpressedasconcentrationthe massofaspecificpollutantinagivenamountofthemedium.Forexample,pollutant concentrationinairistypicallyexpressedasmicrogramsofpollutantpercubicmeterof air.Inwater,concentrationistypicallymilligramsofpollutantperliterofwater,andin soil,milligramsofpollutantperkilogramofsoil. Intuition also suggests that the greater the contact rate with the polluted medium, the greaterthethreatwillbe.Inotherwords,thegreaterthevolumeofcontaminatedair
16
inhaled, or the greater the volume of contaminated water consumed, the greater the threatwillbe. Theentireamountofpollutanttowhichapersonisexposedmaynotbeabsorbedinto thebody.Forexample,whenweinhale,notallofthepollutantgasmoleculesdrawn intoourlungswillpassthroughlungcellmembranesintothebloodstream.Somewill, but others will exit the lungs during the next exhalation. The degree of actual absorptionisdifficulttoknowforanyparticularpollutantandanyparticularindividual. It is customary to make the protective assumption that all of the pollutant contacting the receptor is absorbed. If such an assumption turns out to be wrong, then we will haveerredonthesideofbeingmoreprotectiveratherthanless. Intuitionsuggeststhatthelongerthedurationofexposure,thegreaterthehealththreat willbe.Durationcanbeviewedasthenumberofdaysofexposureperyearcombined withthenumberofyearsthatapersonspendsatthesitewhereexposuretakesplace. In general,theprotectiveassumptionismadethattheexposureoccursfor24hours eachday. Thebodymassofthereceptorplaysarole.Forexample,imaginethatamouseisexposedto onemilligramofapollutant,andthatthisamounthasadefiniteadverseeffect.Exposingan elephanttoonlyonemilligramofthepollutantmaywellhavenoeffectatall.Thelargerthe bodymassofthereceptor,thelessthepollutantintakeisonaperkilogramofbodymassbasis. Averagingtimeisthefinalfactor.Theaveragingtimedependsonwhattypeofhealtheffect thepollutantofinterestcauses.Foranoncancereffect,theaveragingtimeistakentobethe durationoftheexposure.Forcancer,thepollutantintakeisaveragedoveralifetime,whichis generallytakentobe70years.Thebasisforthisapproachistheassumptionthatahighintake receivedforashorttimeperiodhasthesamecancereffectasalowintakereceivedforalong timeperiod. Toxicityassessment This step in the risk assessment process brings into consideration the known or estimated toxicity of the pollutants. Toxicity is primarily determined by observing what effects occur whentestanimals(suchasmiceandrats)areexposedtoapollutantinacontrolledlaboratory setting. The use of animals rather than people is a fundamental limitation in toxicity testing becauseofpossibledifferencesbetweentheresponseinhumanscomparedwiththeresponse inatestspecies.However,whensimilareffectsareobservedacrossnumeroustestspeciesand for exposure by both oral ingestion and inhalation, the relevance ofthese studies to humans becomesmoreconvincing. Toxicitytestsareconductedbydividingthetestanimalsintogroups,witheachgroupreceiving adifferentexposuretothepollutant.Attheendofaspecificperiodoftime(anywherefrom four days to a lifetime), each animal is examined for adverse effects. Not every animal in a groupwillshowaneffectbecauseindividualresponsevarieswithavarietyoffactors,suchas
17
age,sex,healthstatus,genetics,priorexposurestothatpollutant,andcoexposurestoother toxic substances. We generally find that the percentage of animals in each group that experience an adverse effect is greater for greater exposures. A notable exception to this generalization occurs for pollutants that exhibit endocrine disrupting effects. These effects sometimesappeartooccuratlowerexposuresbutnotathigherexposures. PanelsoftoxicologyexpertsareconvenedbytheEPAtoevaluatetheresultsandreliabilityof laboratory toxicity studies. When available, information from epidemiological and clinical studies of humans is also considered. For noncancer effects, the panels determine a referenceconcentrationorreferencedoseforthepollutant.Thesereferencevaluesarea minimumdailyexposurelevelthatishighlyunlikelytoproduceanadverseeffect,notonlyin thegeneralpopulationbutforsensitivesubpopulationsaswell.Separatereferencevaluesare developedforacute(oneto14day),intermediate(15to365day),subchronic(2weeksto7 years), and chronic (7 to 70year) exposures, and for exposures by oral ingestion and by inhalation.Thisapproachreflectstheexperimentalobservationthatnoncancereffectsshowa thresholdexposurelevelbelowthatthreshold,noadverseeffectsareobserved,whileabove that threshold effects occur with increasing frequency as the exposure increases beyond the threshold. The smaller the reference concentration or reference dose, the more toxic the pollutantis. Similar testing and evaluation is conducted for cancer as the adverse effect. However, a pollutants cancer potency is expressed as a slope factor rather than a reference concentrationordose.Thegreatertheslopefactor,thegreatertheprobabilitythatareceptor will develop cancer for a given exposure. Inherent in the slope factor is a protective assumption by the EPA that exposure to any amount of a carcinogen will increase an individualschanceofdevelopingcancer.Inotherwords,unlikenoncancereffects,canceris notconsideredtohaveathresholdexposurelevelbelowwhichnocancerwilloccur. The EPA classifies chemicals according to their carcinogenicity using an approach based on weightofevidence.Ifitistheopinionofexperttoxicologiststhatepidemiologicstudiesof humans establish a cause and effect relationship between exposure to the chemical and the occurrence of cancer, the chemical is classified as a Category A human carcinogen. If the evidence from human studies is judged to be limited or inadequate, but there is sufficient evidence of carcinogenicity in animals, the chemical is classified as a Category B probable human carcinogen. Category C, possible human carcinogens, is for chemicals with even less evidenceofcarcinogenicitythanCategoryBchemicals.CategoryDchemicalsareunclassifiable duetoinadequatestudy.ClassEchemicalsarethoseforwhichevidencesuggeststheyarenot carcinogenic. The primary source of toxicity information for risk assessment is the EPA Integrated Risk Information System (IRIS). This extensive data base provides reference doses, reference concentrations, and slope factors for over 500 substances along with an explanation of how these quantities were determined. A discussion is also provided regarding the level of
18
confidenceinthetoxicityvaluesandsupportingstudies.ThecompleteIRISdatabasecanbe accessedatwww.epa.gov/iris. Riskcharacterization This is the final step in the risk assessment process, where toxicity assessment and exposure assessmentareintegratedintoanexpressionofthedegreeofhealththreat. To evaluate the threat from noncancer effects, the intake determined in the exposure assessmentiscomparedwiththereferencedoseorreferenceconcentrationforthepollutant. The duration of the exposure is taken into account. For example, the intake from a chronic exposure is compared with a reference value derived from chronic exposure studies. The referencevalueisconsideredtobesafebyawidemargin,sothatiftheintakeislessthanthe referencevalue,itishighlyunlikelythatevenasensitivesubpopulationwillexperienceadverse healtheffects.Intakesthatexceedthereferencevaluesareconsideredtobeunacceptableand usually trigger actions that will reduce the exposure. An alternate way of making the comparison is to divide the intake by the reference value to obtain a hazard quotient. A hazard quotient greater than one corresponds to an unacceptable exposure, based on the possibleoccurrenceofnoncancerhealtheffects. To evaluate the threat of cancer, we perform a calculation to determine cancer risk for an individualexperiencingtheexposureofconcern.Theequationforthiscalculationisshownin AppendixA.Cancerriskisoftenexpressedasaprobability,suchasa1in10,000chanceof developing cancer from this exposure. This probability statement canbe interpreted in two waysasthechancethatanindividualexperiencingtheexposurewilldevelopcancerfromthis exposure, or that one person out of every 10,000 people experiencing this exposure will developcancerfromthisspecificexposure.Theriskobtainedfromthisprocedureisconsidered to be an upper bound estimate; in other words, the true risk is likely to be less than the calculated value. The EPA uses a cancer risk of 1 in 1,000,000 (one in one million) as a regulatorygoal,meaningthatregulatoryprogramsaregenerallydesignedtotrytoreducerisk tothislevel.Whenitisnotfeasibletomeetthisregulatorygoal,theEPAconsiderscancerrisks lowerthan1in10,000tobeacceptable. Itisimportanttonotethatcanceroccursatapoorlycharacterizedbackgroundleveldueto natural causes beyond our control. Ionizing radiation is one such cause; we are constantly exposedtoalowlevelofradiationfromtheearthscrustandouterspace.Naturallyoccurring chemicalcarcinogens(suchascertainpolynucleararomatichydrocarbons,whichcanoccurin sootfromnaturalfires)areanothercause.RandommutationsinDNAarealsoafactor.The cancerriskcalculatedinariskassessmentisanadditionalriskontopofthebackgroundrisk. Itisoftenthecasethatareceptorisexposedtomorethanonepollutantatthesametime.As afirstapproximationofthethreatfromacombinedexposure,thehazardquotientscalculated for noncancer effectsareadded together toobtain a hazard index, whichis interpreted in thesamemannerasahazardquotientanindexgreaterthanoneisconsideredunacceptable. Similarly,thetotalcancerriskfrommultiplecarcinogenicpollutantsisobtainedbyaddingthe
19
risks calculated for the individual pollutants, with total risk values greater than 1 in 10,000 consideredtobeunacceptable.Amorerefinedapproachtomultiplepollutantsistosegregate the pollutants according to their major effect, summing only those pollutants with similar effects.Thisapproachrequiresadvancedknowledgeoftoxicologyandmustbedonewithcare soastoavoidunderestimatinghealththreat. A report on the results of a risk assessment must always include a discussion of the uncertaintiesinherentintheresults,ofwhichtherecanbemany.Amongtheseuncertainties are the degree to which actual exposure parameters (pollutant concentration, contact with polluted media, duration of exposure, body mass of receptor) match those used in the risk calculationsanduncertaintiescausedbythelimitationsoftoxicitytesting. The Garfield County Air Toxics Inhalation Screening Level Human Health Risk Assessment (available at http://www.garfieldcounty.com/index.aspx?page=1098) demonstrates the applicabilityofthisriskassessmentprocesstonaturalgasoperationsinGarfieldCounty.This study, conducted by the Disease Control and Environmental Epidemiology Division of the Colorado Department of Public Health and Environment, focuses on data collected in the GarfieldCountyAmbientAirQualityMonitoringReport(availableatthesamesite). ThepresentstudycomplementstheGarfieldCountyAirToxicsstudy.Ratherthanrelyingsolely onmonitoringdatathatpresentsonlyapartialpictureofairquality,thisstudyusespollutant transportmodelingasawayofaugmentingourunderstandingofpossiblehumanexposuresto hydrocarbonsinairfromnaturalgasoperations.Riskcalculationsarethenbasedontheresults of modeling rather than on actual air monitoring data. The present study also includes a qualitativeexaminationofpossiblethreatsfrompollutantsinwaterandsoil. HEALTHEFFECTSOFSELECTEDHYDROCARBONS Benzene, toluene, and m,pxylenes are hydrocarbons that are prominent environmental pollutantsassociatedwithnaturalgasoperations.Theirhealtheffectsaresummarizedbelow basedoninformationshowninIRIS. Benzene.Benzeneisconsideredtobeaknownhumancarcinogen(CategoryAintheweight ofevidencesystem)bothbyoralingestionandinhalationbasedonreliablehumanevidenceas wellassupportinganimalstudies.Thereisclearcausationbetweenexposuretobenzeneand the occurrence of leukemia. Hematologic neoplasms, aplastic anemia, and Hodgkins lymphomaareothercancerrelatedeffectsassociatedwithbenzene.Anincreasedriskoforal andnasalcanceraswellasliver,stomach,lung,ovary,andmammaryglandcancerisassociated withbenzenethroughanimalstudies.Thenoncancereffectsofbenzeneincludeneurotoxicity and depression of bone marrow function, resulting in blood disorders (decreased counts of specificbloodcells,suchaserythrocytes,leukocytes,andthrombocytes)andimpairmentofthe immunesystem.
20
Toluene. The available study results for toluene are inadequate to determine the ability of toluene to cause cancer (Category D in the weightofevidence system). Noncancer effects observedinhumanstudiesincludemanyofaneurologicalnature,suchasimpairedcolorvision, impaired hearing, changes in nerve conduction, headache, dizziness and possible respiratory irritation. m,pXylenes. There are no adequate human data on the carcinogenicity of m,pxylenes and animal studies are inconclusive (Category D in the weightofevidence system). Noncancer effects include mild neurological impairment (such as reduced motor coordination) based on humanandanimalstudies,andpossibledevelopmentaleffectsinanimals(suchascleftpalate, observedinmousestudies.) AIRPOLLUTIONEFFECTS Airpollutanttransport.Ofthethreeexposurepathwaysair,water,andsoilairislikelyto be the most significant pathway related to natural gas operations in Garfield County. In general, people experience relatively little contact with soil outside of their own property, hencetheirexposuretosoilislimited.Manypeopledonotconsumeagreatdealofwaterfrom their home tap, and thus have a limited exposure to ground water. However, everyone breathestheairandhasconsiderableopportunitytobeexposedtoairpollutants. Whether a specific pollutant contributes to the air pathway when released into the environmentdependsonthepollutantstendencytoevaporate.Substancesthataregasesat normal temperatures and pressures (i.e., ambient conditions) will be found in the air when releasedtotheenvironment.Thelighterhydrocarbons,suchasmethane,ethane,andpropane behaveinthismanner.Substancesthatareliquidsatambientconditionsmayalsooccurinthe air.Iftheliquidhasalowboilingtemperatureandhighvaporpressure,atleastaportionof thatliquidreadilyevaporatesintotheatmosphere.Hydrocarbonsthataresomewhatheavier, suchasbenzene,toluene,andm,pxylenesaregoodexamplesofliquidsthatreadilyevaporate. The chemicals known as volatile organic compounds (VOCs) have a strong tendency to evaporateandweexpectthemtooccurasairpollutants.Thisholdstrueregardlessofwhether theVOCsarereleasedintotheenvironmentashydrocarbonliquidsordissolvedinwater. Onceapollutantenterstheatmosphere,itdoesnotremainatthereleaselocationforlong.If thereisanywind,thepollutantiscarriedwiththewindinaprocesscalledadvection.Asthe pollutantmovesdownwind,theturbulentmotionofthewinddispersesthepollutantfromside toside(perpendiculartothewinddirection)andupanddown.Theregionofcontaminatedair (the plume) takes on the approximate shape of a cone, with the point of the cone at the location of the release. At any given distance from the release location, the highest concentration of pollutant is on the centerline of the cone. The concentration along the centerline decreases with increasing distance from the release location. Higher wind speeds
21
spreadthepollutantoveralargervolumeofair,resultinginsmallerpollutantconcentrations thanwouldbeobservedforlowerwindspeeds. This description of air pollutant transport is an idealized picture. Real conditions result in a more complex shape for the contaminated region of air, and a more complex pattern of pollutantconcentrationswithinthatshape.Variabilityinwinddirectionresultsindispersalof thepollutantoveragreatervolumeofair.Thepresenceofhills,valleys,trees,andbuildings influences both the horizontal and vertical movement of air and air pollutants. Atmospheric conditionsareanimportantfactoraswell.Whentemperatureinversionsoccur,pollutantsare trappedinadistinctlayerofairwhosethicknessistypicallyanywherefromafewtensoffeetto hundredsoffeet.Thedevelopmentoftheconeshapedregionofcontaminatedairislimited whentheupperedgeoftheconereachesthetopoftheinversionlayerofair. Air pollutant concentrations determined by sampling and analysis. As described in the precedingsectiononriskassessment,themagnitudeofthehealththreatposedbyapollutant depends on the concentration of the pollutant that a person is exposed to. The best way to obtainpollutantconcentrationsforriskassessmentistocollectandanalyzesamplesthatare representative of the range of concentrations and conditions that occur while an exposure is takingplace.Intheabsenceofsuchsamples,mathematicalmodelsofpollutanttransportcan beusedtopredictpollutantconcentrationsthatoccurforspecificconditions.Modelingisthe primaryapproachusedinthepresentstudy. The Garfield County Ambient Air Quality Monitoring Study, June 2005May 2007 focused on particulatematterlessthan10micronsindiameter(PM10)andVOCs.MeasurementsofPM10 occurredeverythreedaysfortwoyearsatsevensitesinGarfieldCounty,includingurbanareas, ruralareas,andbothurbanandruralareasclosetonaturalgasoperations.Asummaryofthe resultsisshownbelow(Table3).AveragePM10concentrationsfoundateachlocationranged from 9.2 to 27.0 micrograms per cubic meter. The maximum concentrations found at each locationrangedfrom28to92microgramspercubicmeter. Table3.SummaryofPM10MonitoringResults Overall Overall Possible Percent Arithmetic 24Hour Number Number Data Average Maximum ofSample ofSamples Recovery 3 Site Days Recovered (%) (g/m ) (g/m3) GlenwoodCourthouse 14.4 36 243 235 97% NewCastleLibrary 22.8 92 243 238 98% SiltBell 10.7 34 243 237 98% SiltDaley 9.2 28 243 231 95% SiltCox 13.6 62 243 223 92% RifleHenryBldg. 27.9 72 243 240 99% Parachute 27.9 76 243 234 96% FromGarfieldCountyAmbientAirQualityMonitoringStudy,June2005May2007.
22
SamplesofairforanalysisofVOCswerecollectedovera24hourperiodon232occasionsover thecourseoftheAmbientAirQualityMonitoringStudyfromthesamesamplinglocationsused forPM10.Anadditionaltwentysevengrabsampleswerecollectedovera15secondperiod fromsevenadditionalruralsitesclosetonaturalgasoperationsonoccasionswhenodorswere reportedbythepublic.Twoadditionalgrabsampleswerecollectedintheimmediatevicinityof acondensatetankascondensatewasbeingtransferredfromthewellpadtanktoatanktruck. OfthefortythreeVOCscoveredbytheEPAapprovedmethodsusedforanalysis,fifteenwere detectedinatleastonesample.OnlysixoftheseVOCsweredetectedinmorethan20%ofthe totalnumberofsamples:acetone,vinylacetate,2butanone(MEK),benzene,toluene,andm,p xylenes.AsummaryoftheresultsisshowninTable4. Of the VOCs shown in Table 4, benzene, toluene, and m,pxylenes are expected to have the strongest association with natural gas. They are known to occur in unpurified natural gas, albeitinconcentrationswellunder1%.Theirassociationwithnaturalgasissupportedbythe highfrequencyofoccurrence(93%)observedforthecomplaintdrivengrabsamplesforeachof thesethreeVOCs,aswellasthehighconcentrationsfoundinsamplesofcondensatevapors. Itisdifficulttoobtainairsamplesthatarerepresentativeofbenzene,toluene,andm,pxylenes at locations of interest, such as a property line or residence near a natural gas well. The difficulty lies in the inherent variability of pollutant concentrations at the location of interest because of variations over time in emissions from the well along with variations in meteorologicalfactorsthataffectpollutanttransportthroughtheair. Duringthelifeofawell,emissionsofnaturalgascomponentsintotheairarehighlyvariable. Few,ifany,emissionsoccurfromtheboreholeuntilitisextendedintonaturalgasproducing zones.Then,duringaperiodofwellconditioningcalledflowback,naturalgasmaybevented fromthewellintotheair.Oncethewellgoesintoitsroutineproductionphase,thenaturalgas is piped to a local facility for processing and injection into regional and transcontinental pipelines.However,afractionoftheheavier,nonmethanegasesemergingfromthewellare capturedatthewellheadascondensateanddivertedtoanonsitestoragetank.About10% ofcondensateisbenzene,toluene,andm,pxylenes,andthuscondensatetankemissionsarea sourceofthesecompoundsinair.
23
Table4.SummaryStatisticsforDetectedVOCs Detectedcompounds CAS# 74873 67641 75694 75092 108054 78933 67663 71432 79016 108883 591786 127184 100414 13677761 2 100425 95476 106467 Overall24HrSamples (232samples,14sites) OverallGrabSamples (27samples) % Detect s 3.7% 77.8% 7.4% 0.0% 14.8% 70.4% 3.7% 92.6% 0.0% 92.6% 14.8% 0.0% 63.0% 92.6% 0.0% 81.5% 0.0%
% Avg Max Min Avg Max Min Compound Detect g/m g/m g/m g/m g/m g/m s Chloromethane 0.0% 1.5 15.0 0.7 Acetone 18.5 80.0 3.6 81.9% 26.0 81.0 3.7 Trichlorofluoromethane 1.0 26.0 0.7 0.4% 1.5 15.0 0.7 Methylenechloride 1.0 8.4 0.7 1.7% VinylAcetate 2.1 15.0 0.7 23.3% 2.5 15.0 0.7 2Butanone(MEK) 2.2 12.0 0.7 55.2% 3.0 15.0 0.8 Chloroform 0.0% 1.5 15.0 0.7 Benzene 2.2 49.0 0.8 39.2% 28.2 180.0 0.8 Trichloroethene 0.9 2.7 0.7 0.4% Toluene 7.4 130.0 0.8 89.7% 91.4 540.0 0.8 2Hexanone 1.0 4.4 0.7 3.0% 1.7 15.0 0.7 Tetrachloroethene 0.9 2.3 0.7 0.9% Ethylbenzene 1.0 3.4 0.7 3.4% 8.3 96.0 0.8 m,pXylenes Styrene oXylene 1,4Dichlorobenzene 3.9 0.9 1.1 1.1 24.0 0.8 6.0 4.3 12.0 0.7 0.7 0.7 64.2% 106.6 1500.00.8 0.9% 10.3% 3.4% 18.1 260.0 0.8
Detectedcompounds CAS# 71432 108883 100414 Compound Benzene Toluene Ethylbenzene
CondensateLoadoutGrabs (2samples)
13677761 m,pXylenes 2 95476 oXylene
% VentOutlet 50distance Detect g/m g/m s 100.0 590000 360 % 100.0 770000 480 % 100.0 37000 29 % 100.0 250000 200 % 100.0 49000 43 %
NOTE:1/2oftheMinimumReportingLevelwassubstitutedfornondetectvaluesforstatisticalcalculations. FromGarfieldCountyAmbientAirQualityMonitoringStudy,June2005May2007,November2007.
24
Meteorological conditions also change with time. Wind speed undergoes considerable variation,withgreaterspeedsdispersingpollutantsmoreeffectivelythanlesserspeeds,leading tolowerconcentrationsatthelocationofinterest.Winddirectionatmostsitesvariesthrough all the compass directions, so that pollutants are transported toward the location of interest onlypartofthetime.Nighttimetemperatureinversionsandmultidaywinterinversionstrap pollutantsinarelativelythinlayerofstableairfromthelandsurfaceupward,withthetopof theinversionlayerjustafewtensoffeetabovethesurfaceearlyintheevening(leadingtohigh pollutant concentrations) but rising to several hundred feet before sunrise (leading to lower pollutant concentrations, providing the inversion does not persist for an extended period of time).Incontrast,onaclear,sunnyafternooninmidsummer,pollutantsmaybedispersedby convectioncurrentsthroughoutalargervolumeofair,fromthegroundtoelevationsofseveral thousandfeet(creatinglowerpollutantconcentrations).Giventhecomplexityandvariabilityof realconditions,airatalocationofinterestwouldneedtobecharacterizedatleasthourlyfora year or more to develop an accurate knowledge of the range and distribution of pollutant concentrations. Samplescollectedovera24hourperiodprovideasnapshotoftheaverageconcentrationfor thattimeperiod;samplescollectedoverjusta15secondperiodprovidemoreofasnapshot of transient conditions. Unless samples are collected continuously (e.g., a 24hour sample every day), neither sampling approach provides a reliable indication of concentrations over longerperiods.AsdiscussedinSection1,weareinterestedinconcentrationsrepresentativeof longerperiodsinordertoevaluatethedegreeofhealththreatfromexposuretoapollutant. Thevariabilityinemissions,windspeed,andairstabilityalongwiththerelativelysmallnumber of samples collected at natural gasaffected locations during the Garfield County Ambient Air QualityMonitoringStudysuggeststhatwecannotdrawconclusionsaboutnaturalgasrelated pollutant concentrations with the desired degree of confidence. Do the results of the study reflecthighendpollutantconcentrations?Oraverages?Orconditionssomewhereelseonthe spectrum between maximum and minimum concentrations? Without knowing, we are handicapped in our ability to make a reliable assessment of the degree of health threat experienced by Garfield County residents. However, we can use mathematical modeling to estimateorpredictpollutantconcentrationsand,ineffect,generatemissingdata,givingusa betterideaoftherangeanddistributionofconcentrations.Thatsaid,onemustbearinmind that mathematical models have limitations. Modeling inevitably requires that a number of simplifyingassumptionsbemadeinordertomakethecalculationsfeasible.Theseassumptions do limit the ability of the model to accurately reflect real conditions, but as long as the limitationsofmodelingareunderstoodandtakenintoaccountwheninterpretingtheresults, modelingcanprovideausefulballparklevelofaccuracythathelpstoilluminatethedegree ofthreatandprovidesguidancetodecisionmakers. Air pollutant concentrations determined by modeling. The Gaussian plume model is the simplestapproachtopredictingpollutanttransportfromasingle,welldefinedpointsource of air pollutant emissions (such as a well or condensate tank). A Gaussian plume is the qualitative picture described earlierthe coneshaped region of contaminated air extending
25
downwind from the pollutant release point. The result of Gaussian plume modeling is an estimate of pollutant concentration at any specific location of interest downwind of the pollutant source. The model is based on userspecified values for pollutant emission rate, positionofthelocationofinterestrelativetothesource,windspeed,andatmosphericstability (Air Pollution Modeling, Paolo Zannetti, Van Nostrand Reinhold, 1990). The mathematical equationsusedinthismodelareshowninAppendixB. TheGaussianplumemodelwasusedtoestimateconcentrationsofbenzene,toluene,andm,p xylenesinairformeteorologicalconditionscharacteristicofGarfieldCounty. Wind speeds were based on data collected during the Garfield County Ambient Air Quality MonitoringStudyforfivesites:SiltBell,SiltCox,SiltDaley,Parachute,andNewcastleLibrary. Data from all five sites were combined to generate an average proportion of time over the courseofatypicalyearthateachwindspeedrangeoccurred,withseparateaveragesforday andnight(AppendixC). The wind direction data from the Ambient Air Quality Monitoring Study were examined to determine the maximum proportion of time in which the wind blew in any given direction (AppendixD). One of the factors that determines air stability is insolation the intensity of sunlight. Assumingaclearsky,stronginsolationoccurswhentheanglebetweenthehorizonand the sun is greater than sixty degrees; moderate insolation occurs for angles between thirtyfiveandsixtydegrees;andweakinsolationoccursforanglesbetweenfifteenand thirtyfivedegrees.Sunangletableswereconsultedtodeterminetheproportionofthe yearinwhichthesunwasineachofthesethreeangleranges(AppendixE). Airstabilityisalsodeterminedbythedegreeofcloudiness.Nocloudcoverinformation was available for Garfield County. The best available information was a monthby monthbreakdownoftheaveragenumberofdaysclassifiedasclear,partlycloudy,and cloudydevelopedbytheNationalWeatherServiceforGrandJunction.Thesedatawere convertedtotheproportionofayearineachcloudcovercategory(AppendixF). The Gaussian plume model requires that meteorological conditions be characterized as falling into one of six air stability classes. The information on wind speed, insolation, and cloudiness described above was combined so as to determine the proportion of timeduringatypicalyearthatcouldbeclassifiedasfallingintoeachofthesesixstability classes(AppendixG). Generic pollutant concentrations were calculated for each stability class for distances ranging from 25 meters to 4,800 meters directly downwind of the source for a convenient,arbitraryemissionrateof1grampersecond.Theconcentrationresultsfor each stability class were combined into a single set of results by weighting each class according to the proportion of a typical year in which that stability class occurs
26
(Appendix H). The result was a set of concentrations for various distances downwind from the pollutant source. These generic concentrations, which are characteristic of meteorological conditions over a typical year, were used as the basis for calculating specific concentrations for each of the pollutants and emission scenarios described below. FivespecificemissionscenarioswereexaminedusingtheGaussianplumemodel: Flowbackisastageintheconditioningofanewnaturalgaswellpriortoproduction.In thetraditionalflowbackprocess,gasisallowedtoventfromthewelltotheatmosphere for as long as thirty days. Typical gas emission rates and composition were used to calculaterepresentativeemissionratesspecificallyforbenzene(0.07gramspersecond), toluene (0.07 grams per second), and m,pxylenes (0.07 grams per second) (Appendix Ia). Inrecentyears,morecompaniesareturningtogreencompletions,inwhichthenatural gasventedduringflowbackiscapturedandprocessedforsale.Atypicalcapturerate and gas composition were used to calculate representative emission rates under this scenario for benzene (0.0049 grams per second), toluene (0.0049 grams per second), andm,pxylenes(0.0049gramspersecond)(AppendixIa). Glycoldehydrationunitsareusedatthewellheadduringroutineproductiontoremove water from the gas stream before injecting it into a pipeline for transport to a local processingfacility.Atypicalemissionrateandcompositionfordehydrationemissions was used to calculate representative emission rates for benzene (0.0036 grams per second), toluene (0.01 grams per second), and m,pxylenes (0.01 grams per second) (AppendixIb). During routine production, heavier hydrocarbons are captured at the wellhead and separated from the gas stream as condensate. The condensates are stored at the wellhead in condensate tanks, from which they are periodically collected and transportedtorefineries.Emissionsfromcondensatetanksvarywiththerateatwhich condensate is produced by the well, which varies from one well to another as well as during the lifetime of the well. InWeld County, Colorado, condensate tankemissions arelimitedto20tonsperyearoftotalvolatileorganichydrocarbons(VOCs)aspartof theefforttoreducetheformationofozoneinFrontRangeair.Asapointofreference, this emission rate was used along with typical condensate emission composition to calculaterepresentativeemissionratesforbenzene(0.0081gramspersecond),toluene (0.0415gramspersecond),andm,pxylenes(0.0415gramspersecond)(AppendixIc). Condensate tanks can be outfitted with combustors to destroy VOC emissions. These combustorscantypicallyoperatewithadestructionefficiencyof98%,leadingtolower emission rates for benzene (0.00016 grams per second), toluene (0.00083 grams per second),andm,pxylenes(0.00083gramspersecond)(AppendixIc)
27
TheGaussianplumemodelwasusedtocalculateairconcentrationsforbenzene,toluene,and m,pxylenes at multiple distances ranging from 25 meters (82 feet) to 4,800 meters (approximately 3 miles) for each of the five emission scenarios. The results of the model calculationsareshowninTables5A5C. Of the five scenarios, flow back with no gas recovery produces the highest concentrations of eachofthesepollutants.Inthisemissionscenario,benzeneconcentrationsrangedfrom358 microgramspercubicmeteratalocation25metersdownwindofthesourceto0.1micrograms per cubic meterat a distance 4,800 meters downwind. Benzene concentrationsreported for actualairsamplesintheGarfieldCountyAmbientAirQualityMonitoringStudyhadamaximum of49microgramspercubicmeterfor24hoursamplesand180microgramspercubicmeterfor 15secondgrabsamples.Theseconcentrationspredictedbythemodelaresimilartotherange ofconcentrationsfoundthroughsamplingandanalysis. In this same emission scenario, toluene and m,pxylenes were calculated to range from 215 micrograms per cubic meter at a location 25 meters downwind to 0.1 micrograms per cubic meter at a location 4,800 meters downwind. Toluene concentrations reported in actual air samples collected for the Ambient Air Quality Monitoring Study were as high as 130 microgramspercubicmeterinthe24hoursamplesand540microgramspercubicmeterinthe 15second samples, although the concentrations averaged only 7.4 and 91 micrograms per cubic meter, respectively. It is noteworthy that the 15second maximum value exceeds the maximum calculated using the model. m,pXylenes concentrations were as high as 24 microgramspercubicmeterinthe24hoursamplesand1,500microgramspercubicmeterin the15secondsamples,withaveragesof3.9and107microgramspercubicmeter,respectively. Again,themaximum15secondvalueexceedstherangecalculatedusingthemodel. For any specific location, time, and conditions, actual concentrations may vary from those predicted by the model for several reasons. The model is based on wind speeds, wind directions, cloudiness, sun angles, and atmospheric stability averaged over the course of a specificyear.However,conditionsatanyspecifictimecanbeconsiderablydifferentthanthe average.Forexample,pollutantconcentrationsinaircanbeespeciallyhighinearlyeveningas a nighttime inversion begins to develop. Emissions are trapped in a layer of air that, for a limited period of time, may be no thicker than a few tens of meters above the land surface, confiningthepollutantstoasmallervolumeofairleadingtohighconcentrations.Topography intheformofhillsandvalleyscansometimeschannelairflowandalterpollutantdispersion.
30
Table 5A. Benzene Concentrations in Air Resulting from Selected Natural Gas Development Operations (Concentrations in micrograms per cubic meter) Condensate emissions with 98% removal of 20 tons per year total VOCs 0.5 0.2 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Downwind distance (meters) 25 50 75 100 125 150 175 200 250 300 350 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 3200 4800
Flow back, no recovery 214.5 96.5 58.5 39.4 28.3 21.5 16.9 13.7 9.6 7.2 5.6 4.4 3.1 2.3 1.7 1.4 1.1 1.0 0.8 0.7 0.6 0.6 0.5 0.4 0.2 0.1
Flow back, 93% recovery 15.0 6.8 4.1 2.8 2.0 1.5 1.2 1.0 0.7 0.5 0.4 0.3 0.2 0.2 0.1 0.1 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Wellhead glycol dehydration 11.0 5.0 3.0 2.0 1.5 1.1 0.9 0.7 0.5 0.4 0.3 0.2 0.2 0.1 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Condensate emissions at 20 tons per year total VOCs 24.8 11.2 6.8 4.6 3.3 2.5 2.0 1.6 1.1 0.8 0.6 0.5 0.4 0.3 0.2 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.0 0.0
31
Table 5B. Toluene Concentrations in Air Resulting from Selected Natural Gas Development Operations (Concentrations in micrograms per cubic meter) Condensate emissions with 98% removal of 20 tons per year total VOCs 2.5 1.1 0.7 0.5 0.3 0.3 0.2 0.2 0.1 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
32
CommunityHealthRiskAnalysisofOiland GasIndustryImpactsinGarfieldCounty
Table 5C. m,p- Xylenes Concentrations in Air Resulting from Selected Natural Gas Development Operations (Concentrations in micrograms per cubic meter) Condensate emissions with 98% removal of 20 tons per year total VOCs 2.5 1.1 0.7 0.5 0.3 0.3 0.2 0.2 0.1 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Exposure scenarios. Three standard exposure scenarios for noncancer effects were considered:acuteexposureslastingfromonetofourteendays;intermediateexposureslasting fromfifteento364days;andchronicexposureslastingfromsevenyearstoalifetime.Three exposure scenarios tailored to natural gas operations in Garfield County were used in the evaluationofcancerrisk: Lifetime exposure (seventy years) to ongoing natural gas developmentA person lives for seventy years in one or more locations where he or she is always downwind of natural gas operations. Because ongoing development of new wells appears to be unlikely for such an extended time in Garfield County, this is unlikely to occur for exposures due to flow back. However, this scenariomay be more realistic for exposure to condensate tankemissions and glycoldehydrationunitemissions,whichmaycontinueaslongasthewellisinproduction.The lifetimeexposurescenariowillresultinthegreatestcancerriskandisincludedprimarilyasa possibleworstcase. Twoyear exposure to natural gas developmentA person lives for two years in a location downwind of natural gas operations. This scenario includes situations in which multiple well padsaredevelopedinproximitytothisperson,withthetotaldurationofdevelopmentbeing twoyears. Oneyear exposure to natural gas developmentA person lives for one year in a location downwindofawellpadthatisdevelopedoverthecourseofthatyear. TheDraftGarfieldCountyAirToxicsInhalationScreeningLevelHumanHealthRiskAssessment identified two sites, Cox and Daley, as representative of rural background locations representative of areas unaffected by natural gas operations. The exposure point concentrations for benzene identified in Table A7 of that study are 1.9 micrograms per cubic meterfortheCoxsiteand0fortheDaleysite.Anaverageof1.0microgramspercubicmeter was thus used as a background concentration applicable to all three of the above exposure scenariosforcancerrisk.Inadditiontotheseventyyear,twoyear,andoneyearexposureto benzene from natural gas operations, a seventyyear exposure to the background concentrationofbenzenewasincluded. Each of these three exposure scenarios also includes a lifetime exposure to background concentrations of benzene in air reported for rural areas in the Garfield County Ambient Air QualityMonitoringStudy. Threatofnoncancereffectsfromairpollutants.ThethreatofnoncancereffectsfromPM10is belowanylevelofconcern.TheEPAhasestablishedaNationalAmbientAirQualityStandard fortwentyfourhouraverageconcentrationsofPM10thatisprotectiveofnotonlythegeneral population but sensitive subpopulations as well. As discussed in Section 2.2 of the Garfield CountyAmbientAirQualityMonitoringStudy,therewerenoexceedancesofthisstandard.
34
The threat of noncancer effects from exposure to benzene, toluene, and m,pxylenes was evaluatedusingtheriskassessmentproceduredescribedinSection1.Pollutantconcentrations weregeneratedusingtheGaussianplumemodel,asdescribedabove.Exposurestobenzene, toluene, and m,pxylenes may produce noncancer health effects if the exposures are high enough. Reference concentrations are pollutant concentrations in air that represent thresholds below which health effects are not expected to occur. Reference concentrations have been developed for acute exposures (one to fourteen days), intermediate exposures (fifteen to 364 days), and chronic exposures (seven years to a lifetime). Table 6 shows referenceconcentrationsusedinthisstudy.
Table 6. Reference Concentrations (in micrograms per cubic meter) Acute Benzene Toluene m,p-xylenes 30 3,766 9,000 Intermediate 20 none available 3,037 Chronic 30 5,000 100
FromDraftGarfieldCountyAirToxicsInhalationScreeningLevelHumanHealthRiskAssessment, December2007.
Tables7AEshowthefollowingsituationsinwhichoneormorereferenceconcentrationsare exceeded: For flow back with no gas recovery, the acute reference concentration for benzene is exceededfordistancesupto125metersdownwind. For flow back with no gas recovery, the intermediate reference concentration for benzeneisexceededfordistancesupto150metersdownwind. For flow back with no gas recovery, the chronic reference concentration for benzene is exceededfordistancesupto125metersdownwind. Forflowbackwithnogasrecovery,thechronicreferenceconcentrationform,pxylenes isexceededfordistancesuptotwentyfivemetersdownwind For emissions of VOCs at twenty tons per year from condensate tanks, the chronic reference concentration for m,pxylenes is exceeded for distances up to twentyfive metersdownwind. Flowbackwithnogasrecoveryistheonlyoperationthatexceedsreferenceconcentrationsat distancesbeyondtwentyfivemeters.Ofthethreepollutantswhoseemissionsweremodeled, benzeneistheonlyonethatexceedsreferenceconcentrationsatdistancesbeyondtwentyfive meters. These resultssuggest that emission of benzene during uncontrolled flow back is the situation that presents the greatest threat of noncancer effects. These effects may occur in
35
peoplewhospendonedayormorewithinadistance125metersdownwindofthenaturalgas well. ThethreatofnoncancereffectswasalsoexaminedintheGarfieldCountyAirToxicsInhalation Screening Level Human Health Risk Assessment. That study was based on pollutant concentrations generated by sampling and analysis using a reasonable estimate of the concentration likely to be contacted over time (i.e., the 95% upper confidence level on the arithmeticmean). One of the fourteen sitesevaluated(Brock)exceeded the acute reference concentrationforbenzene.Thatsiteandoneother(NewcastleLibrary)hadanexcessivelevel ofintermediateexposure,mostlyduetobenzene. Cancer risk from air pollutants. Of the three pollutants that are the focus of this study, benzeneistheonlycarcinogen.AccordingtotheU.S.EnvironmentalProtectionAgency(EPA) IntegratedRiskInformationSystem,benzenehasaslopefactorthatrangesfrom0.000077up to0.00027permicrogramperkilogramperday.Thehigherofthetwoslopefactorswasused inthisstudy. Tables5AEshowthebenzenecancerrisksforeachoftheaboveexposurescenariosappliedto eachofthefiveemissionscenarios.TheEPAadvocatesasagoalthatcancerriskforaspecific exposurebenomorethan1chancein1million.However,theEPAfurtherspecifiesthatcancer riskforaparticularexposurebeconsideredacceptablewhenitisbetween1chancein1million and 1 chance in ten thousand. Of the risks shown in Tables 5AE, the following exceed the upperendoftheEPAacceptablerange: For flow back with no gas recovery, the seventyyear exposure exceeds the acceptable range fordistancesupto200metersdownwindofthewell. For flow back with 93% recovery of gas, the seventyyear exposure exceeds the acceptable rangefordistancesuptotwentyfivemetersdownwindofthewell. ForVOCemissionsoftwentytonsperyearfromcondensatetanks,theseventyyearexposure exceedstheacceptablerangefordistancesuptotwentyfivemetersdownwindofthewell. Benzeneemissionsduringuncontrolledflowbackpresentthegreatestriskofcancer.However, the risk of cancer exceeds the EPA acceptable range only for a seventyyear exposure. An exposureofthatdurationappearsunlikely. CancerriskwasalsoexaminedintheDraftGarfieldCountyAirToxicsInhalationScreeningLevel HumanHealthRiskAssessment.ThreesitesBrock,Parachute,andGlenwoodSpringswere determined to have cancer risk exceeding the upper end of the EPA acceptable range for a seventyyear exposure. Brock, identified as a site affected by natural gas operations, had a cancer risk slightly over the upper end of the range due entirely to benzene. Parachute was identifiedasurban,withtheusualurbansourcesofVOCs,aswellasbeingdirectlyaffectedby natural gas operations. Cancer risk in Parachute was from cumulative exposure to
36
Table 7A. Benzene Cancer Risk Resulting from Flow Back with No Gas Recovery (Benzene emission rate of 0.07 grams per second) Rural background cancer risk (chances per million) 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8
Concentration (micrograms per cubic meter) 214.5 96.5 58.5 39.4 28.3 21.5 16.9 13.7 9.6 7.2 5.6 4.4 3.1 2.3 1.7 1.4 1.1 1.0 0.8 0.7 0.6 0.6 0.5 0.4 0.2 0.1
70-Year exposure, total benzene cancer risk (chances per million) 1680.6 760.2 463.8 314.7 228.8 175.3 139.7 114.7 82.7 63.6 51.1 42.5 31.7 25.3 21.3 18.6 16.7 15.3 14.2 13.3 12.7 12.1 11.7 11.3 9.0 8.5
37
Table 7B. Benzene Cancer Risk Resulting from Flow Back with 93% Recovery of Natural Gas (Benzene emission rate of 0.0049 grams per second) Rural background cancer risk (chances per million) 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8
38
Table 7C. Benzene Cancer Risk Resulting from Operation of Wellhead Glycol Dehydration Units (Benzene emission rate of 0.0036 grams per second) Rural background cancer risk (chances per million) 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8
39
Table 7D. Benzene Cancer Risk Resulting from Condensate Tanks Emitting 20 Tons Per Year of Volatile Organic Hydrocarbons (Benzene emission rate of 0.0081 grams per second) Rural background cancer risk (chances per million) 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8
40
Table 7E. Benzene Cancer Risk Resulting from Condensate Tanks with 98% Removal of Emissions of 20 Tons per Year of Volatile Organic Hydrocarbons (Benzene emission rate of 0.0016 grams per second) Rural background cancer risk (chances per million) 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8 7.8
Benzene, 1,4dichlorobenzene, and trichloroethene. Cancer risk in Glenwood Springs, identifiedashavingonlyurbancharacteristics,wasfromcumulativeexposuretobenzene,1,4 dichlorobenzene,andmethylenechloride. Uncertaintiesinnoncancerthreatandcancerrisk.Thereareanumberofuncertaintiesinthe noncancer threat and cancer risks determined using pollutant concentrations based on the Gaussianplumemodel.Someoftheseuncertaintieshavetheeffectofreducingthreatandrisk. Actual pollutant emission rates may be less than those used in the model calculations. Thereissomevariationinemissionratesfromonewellpadtoanother.Theremaybe variationfromonecompanytoanotheraswell,ifdifferentoperatingpracticesareused. Sitespecific meteorological conditions and topography may differ from the average conditions upon which the model is based. If sitespecific conditions and topography enhancethedispersionofpollutants,therewillbelessexposuretoresidentsatthatsite andlesslikelihoodofhealtheffects. CancerriskwouldbelessifthelesspotentofthetwoslopefactorsprovidedintheEPA IntegratedRiskInformationSystemwasused. Otheruncertaintieshavetheeffectofincreasingthreatandrisk. Actualpollutantemissionratesmaybegreaterthanthoseusedinthemodelcalculations, forthesamereasonsidentifiedabovethatratesmaybeless. Similarly, if sitespecific meteorological conditions and topography limit pollutant dispersal, the likelihood of health effects would be increased. This situation is most likely to apply to acute exposures, which may occur from transient or shortterm conditions,suchasamultidaywinterinversion. The presence of other naturalgas related sources of these pollutants may increase the likelihoodofhealtheffects.Benzene,toluene,andm,pxylenesarecomponentsoffuel andexhaustfromvehiclesorenginesthatareusedonwellpads.Thisemissionsource wasnotincludedinthemodel. If multiple well pads are in development at the same time in the same vicinity, their effectswillbeadditiveforanysitethatreceivespollutantemissionsfromeachsite. Thepresenceofothercarcinogensinoutdoorair,notincludedinthemodel,willincrease cancerrisk.Forexample,1,4dichlorobenzenewasoccasionallyobservedintheGarfield CountyAmbientAirQualityMonitoringStudy.Thisconsiderationappliesaswelltothe presenceofotherairpollutantshavingnoncancereffects.
42
Thepresenceofotherpollutantsinindoorairmayincreasethelikelihoodofnoncancer effectsandcancer.Combustiondevices(suchasgasandwoodstoves),tobaccosmoke, paints,furnishings,andbuildingmaterialsallemitpollutants. Discussion.Giventheaboveuncertainties,theuseofmodelingasasubstituteforsamplingand analysis is no panacea. However, modeling can provide pollutant concentrations that can be usedtoilluminateanapproximateunderstandingofthelikelihoodofhealtheffects. Theresultspresentedabovesuggestthatthesingleoperationcreatingthegreatestthreatfrom exposure to benzene, toluene, and m,pxylenes is flow back with no recovery of natural gas. Until recently it was routine to conduct flow back in this manner. However, at this time companiesaremakingincreasinguseofgreencompletions,inwhichover90%ofthenatural gasisrecoveredduringflowback.Thispracticeisaneffectivewaytoreducethesinglemost significant exposure to benzene, toluene, and m,pxylenes. For cases in which the green completionapproachisnotused,asetbackrequirementisdesirable. WATERPOLLUTIONEFFECTS Pollutanttransportingroundwater.Ausefulstartingpointistoconsiderthegeneralbehavior of pollutants in ground water. Contamination in ground water has several possible origins. Liquidsthatarespilled,leaked,ordumpedontothelandsurfacemayinfiltrateintothesoiland percolate downward to ground water. These can be organic liquids, such as gasoline or condensate,orwaterbasedliquidscontainingdissolvedpollutants.Liquidsmayleakintothe subsurface soil from pits, ponds, underground tanks, and pipelines. These liquids can also percolatedownwardtogroundwater.Solidmaterialsthatarespilled,leaked,ordumpedonto thelandsurfacemayundergothesameprocess,providedtheyarewatersoluble,bydissolving intorainwaterthatinfiltratesintothesoil. Asanyliquidmovesthroughthesubsurface,itmayleavebehindathinfilmofitselfonthesoil androcksurfaces.Thiseffectgraduallydecreasestheamountofliquidthatcontinuestoflow downward, and releases of small volumes of liquid are less likely to reach the ground water thanreleasesoflargervolumes.Anothercomplicatingfactoristhetendencyofmostpollutant molecules to undergo sorption to the rock and soil surfaces through physical and chemical interactions.Thesorbedpollutantbecomesaresiduethatisleftbehindastherestoftheliquid continues to percolate. This residue can act as a secondary source of contamination. When clean rain water percolates through the subsurface, pollutant molecules in the residue can desorborredissolve(leach)intothiswaterandbecarrieddownwardtothegroundwater. When polluted water reaches the water table, it mixes with the ground water and creates a slugorplumeofcontaminatedgroundwater.Whenahydrocarbonliquidreachesground water, a large portion of the hydrocarbon liquid will float on the water table. Over time, hydrocarbonmoleculesfromthisfloatinghydrocarbonlayerwilldissolveintothegroundwater, creatingaslugorplumeofcontamination.
43
Ground water typically flows slowly through an aquifer. The contaminant plume is carried alongwiththemovinggroundwaterbyadvection.Thevelocityofthegroundwaterdepends on the hydraulic gradient of the ground water the hydraulic conductivity of the aquifer. (Hydraulicgradientcanbepicturedastheslopeofthewatertable,withgroundwaterflowing fromanareaofhighwatertabletoanareaoflowwatertable.Hydraulicconductivityrefersto howreadilythewatercanmovethroughtheaquifer.Highvaluesofgradientandconductivity leadtoahighervelocityofgroundwaterflow.)Thecontaminantplumeitselfusuallymovesat a slower velocity than the ground water due to the sorption process described above. Any particular dissolved pollutant molecule will sorb at some random time. While the pollutant moleculeissorbed,itisstationary.Whenitdesorbsatsomelaterrandomtime,thepollutant becomesmobileandresumesadvection.Overall,anygivenpollutantmoleculewillspendpart ofitstimestationaryratherthanallofitstimemobile,andthuswillmoveataslowervelocity thanthegroundwateritself,whichismobileallofthetime. Initially, the plume is small and the pollutant concentrations within the plume are relatively high. As the plume moves, it spreads out in all directions by hydrodynamic dispersion, and pollutant concentrations within the plume decrease over time because the pollutant mass is distributed over a larger volume of water. Many organic pollutants, including most of the smallerhydrocarbons,aresusceptibletobiodegradation.Naturallyoccurringbacteriacanuse thepollutantsasasourceoffood(thatis,asasourceofcarbonforgrowthandenergyfor metabolism). The biodegradation process breaks down the hydrocarbon molecules into simplerhydrocarbonsortocarbondioxideandwater.Therateofbiodegradationdependson thechemicalstructureofthepollutantmolecule,thetypeofbacteria,theamountofdissolved oxygen present in the ground water, and the availability of nutrients for the bacteria. Biodegradationcanbeavaluableassetifitworksatafastenoughrate.Takentogether,the effectsofdispersionandbiodegradationareoftenreferredtoasnaturalattenuation. ThegroundwaterpathwayinGarfieldCounty.Naturalgasoperationshavethepotentialto create ground water contamination. One potential source of contamination is an improperly constructed gas well, which can allow the escape of natural gas, drilling fluids, or produced water into the surrounding geologic formation at the breach. Another potential source is seepagefromunlinedpitsusedtostorewaterthatmaybecontaminatedwithhydrocarbonsor chemicals used in different stages of the well installation and completion process. Hydraulic fracturing creates fractures having the potential to act as conduits for the flow of water contaminatedwithhydrocarbonsandsubstancesusedinthefracturingprocess.Surfacespills or leaks of materials such as condensate can allow hydrocarbons to eventually reach ground water, as described on the preceding page. From the standpoint of evaluating the threat to human health, the key question (as explained in the earlier section on risk assessment) is whetherornotagivenexposurerouteiscomplete. Rural residents of Garfield County have expressed concern about possible effects on ground waterusedfordrinkingandotherdomesticpurposes.Withtheexceptionofthewellknown contamination problem near West Divide Creek caused by faulty well construction, available
44
datafromsamplingandanalysisdonotshowthepresenceofnaturalgasrelatedpollutantsin groundwater.GroundwatercharacterizationrequiredbyGarfieldCountyfornewsubdivisions has not revealed any contamination problems (J. Rada, Garfield County Health Department). Similarly,examinationofthegroundwaterdatabasemaintainedbytheColoradoOilandGas Commission revealed no contamination outside the vicinity of the problem in West Divide Creek. Can ground water be contaminated even though no contamination has been revealed by resultsfromsamplingandanalysis?Giventhatthereisnoexhaustivegroundwatermonitoring program in place, the possibility exists that undetected contaminant plumes may be present. Rural water wells could be an important source of water quality data. However, landowners mustbearthecostofanalyzingtheirwater,andmostlandownerslikelydonothavetheirwater analyzedbecauseofcost. Howlikelyisitthataplumewillaffectawaterwell?Thelikelihoodofacontaminantplumein thedrinkingwateraquiferreachingaprivatewellusedbyruralresidentsisstronglydependent on location and releasespecific factors. Most ground water used in rural areas of Garfield Countyisfrom20toabout250feetbelowthelandsurface(D.Dennison,GarfieldCountyOil& GasLiaison).Pollutantreleasesthatoccurwithinthiszonedohavethepotentialtoreachwater wells.Whetherthecontaminantplumedoessodependsfirstonthedirectionofgroundwater flowawellmustbeinthepathoftheplumeinordertobeaffected.Theverticallocationof thereleasemustalsobeamatchforthepointwherethewelldrawswater.Apollutantrelease atagivendepthcanresultinaplumethatmaynottravelverticallytoadepththatwouldallow ittoenterthewell.Evenifawellisinthepathoftheplume,thedistancebetweentherelease locationandthewellisalsoakeyfactor.Thegreaterthedistancethattheplumetravels,the more the plume spreads out by hydrodynamic dispersion, reducing the maximum pollutant concentrationsfoundintheplume.Biodegradationcancausefurtherreductionsinpollutant concentrations within the plume to the extent that the pollutant may no longer be a threat. Therefore the degree of threat posed by a release to ground water depends strongly on the locationofthereleaserelativetonearbywaterwells,intermsofbothdirectionanddistance. There is public concern about the possibility of hydraulic fracturing leading to ground water contamination.Partofthisconcernisderivedfromtheunknowncompositionofthematerials pumped through the gas well into the gasbearing formation during the hydraulic fracturing process; the detailed composition of these materials is treated as proprietary information by thenaturalgasindustry.Concernalsostemsfromthepossibilitythatthenewfracturesmay promotethemovementofnaturalgasandhydraulicfracturingfluidstowardwaterwells. Natural gas wells in Garfield County range in depth from about 2,000 feet below the land surfacetoasmuchas18,000feetbelowthelandsurface(2006PhaseIHydrogeologicStudyby URS, Inc.), with most typically falling in the range from 4,000 to 9,000 feet (D. Dennison, Garfield County Oil & Gas Liaison). Thus, considerable transport must take place to bring hydraulicfracturingchemicalsfromsuchgreatdepthsuptothesurficialaquiferusedbyrural residents.Giventhatthegasesandfluidsinthesedeepformationsareundergreatpressure,
45
there is a tendency for upward migration of pollutants towards lower pressure zones at shallowerdepths.Therateofmigrationcanbeveryslowbecauseofthelowpermeabilityof formationsbetweenthegasbearingzonesandthesurficialaquifers.However,fracturesand faults can provide preferential flow paths that offer less resistance to pollutant transport. EvidenceofdeepgroundwatermixingwithshallowgroundwaterintheMammCreekgasfield in southeastern Garfield County was presented in the 2006 Phase I Hydrogeologic Study performed by URS, Inc., demonstrating that the potential for pollutant transport from deep zones to shallow ground water is real. Due to the complexity and heterogeneity of the subsurfaceacrossGarfieldCounty,itisdifficulttopredictwithanyconfidencewhenandwhere thiskindofmigrationmayoccur. A risk assessment for ingestion of contaminated ground water was not performed. Without data showing contaminant concentrations in water currently used or potentially useable for drinking,thereisnobasisforcalculatingrisk.Similarly,availableinformationisinsufficientto permit the use of modeling to predict contaminant concentrations. However, the maximum contaminantlevelsdescribedbelowinthediscussionofsurfacewaterarealsoausefulguide forgroundwaterusedfordrinking.Eventhoughtheselevelsarelegallyenforceableonlyfor public water systems, residents who monitor their private well may use these levels as a guidelineforunacceptablecontaminantconcentrations. The surface water pathway in Garfield County. There are several ways in which natural gas operationsmaycontaminatesurfacewaters(i.e.,rivers,streams,lakes,ponds,andwetlands). Hydrocarbons and other materials involved in natural gas operations can enter the water directlythroughspills,suchasthosethatsometimesoccurintransportationaccidents.Spillsor leaksthatoccursomedistanceawayfromsurfacewatermaystillenterthewaterasaresultof overlandflowacrossthelandsurface.Spilledorleakedmaterialsthatdontflowallthewayto surfacewaterontheirownmaybewashedintothewaterbyrainorsnowmeltflowingacross the land surface. Some bodies of surface water are fed by seepage from ground water. In these situations, ground water contaminant plumes described in the preceding section may also affect surface water. In contrast, some surface waters act as a source of recharge for ground water. In these situations, contaminated surface water can lead to a plume of contaminatedgroundwater. TheColoradoRiverandcertaintributariesareusedasasourceofdrinkingwaterbythepublic watersystemsoperatedbytheCityofRifleandtheTownsofParachuteandSilt. PublicwatersystemsareregulatedbytheEPAundertheSafeDrinkingWaterActandarethus subject to the National Primary Drinking Water Regulations, which can be viewed at http://www.epa.gov/safewater/contaminants/index.html. These regulations set maximum contaminantlevelsforavarietyofmicroorganismsandchemicalsinwaterthatissuppliedto thepublic.Fourofthemostprominentpetroleumhydrocarbonshavebeenassignedmaximum contaminantlevels:benzeneat0.005milligramsperliter,tolueneat1milligramperliter,ethyl benzeneat0.7milligramsperliter,andtotalxylenesat10milligramsperliter.Barium,whichis a component of drilling muds, has a maximum contaminant level of 2 milligrams per liter.
46
Public water systems are required to monitor the water for these contaminants on a regular basisand,ifviolationsoccur,providenoticetothepublicandtakecorrectiveaction. TherequirementsoftheSafeDrinkingWaterActprovideGarfieldCountyresidentssuppliedby publicwatersystemswithsignificantprotectionfromdrinkingwatercontaminantsassociated withnaturalgasoperations.Inorderforthisregulatorysystemtobeeffective,itisimperative that the public water systems meet all the requirements for monitoring. Should benzene, toluene, ethyl benzene, xylenes, or barium be discovered at unusually high concentrations through this monitoring, the public water systems should consider additional monitoring for othercontaminantsthatareassociatedwithnaturalgasoperations. SOIL Soilcanbecomecontaminatedbyspills,leaks,oroverflowsontothelandsurface.Liquids(both hydrocarbonsandwaterbasedliquids)canflowdownslopeacrossthelandsurface,spreading pollutantsacrossalargerareaofsurfacesoil.Someportionoftheliquidswillinfiltrateintothe soil,spreadingpollutantsfromthesurfacedownward.Sorptionofpollutantstosoilresultsina residue that persists until rain and snowmelt have leached the pollutants off the soil and carriedthemfartherawayacrossthelandsurfaceorfartherdownintothesubsurface.Water solublesolidmaterialsthatarespilledorleakedontothelandsurfacemayundergothesame processbydissolvingintorainandsnowmeltthatinfiltratesintothesoil. The primary manner in which people are exposed to pollutants in soil is through ingestion. Whileitiscertainlynotthecasethatadultsintentionallyeatdirt,asmallamountofincidental ingestion is plausible. For example, soil on hands from gardening or other outdoor activities can be transferred into a persons mouth. Young children are more likely to ingest soil as a result of their handtomouth habits. The EPA recommends making the assumption that an adultingestsanaverageofonetenthofonegram(lessthanfouronethousandthsofanounce) ofsoilperday(RiskAssessmentGuidanceforSuperfund,Volume1.HumanHealthEvaluation Manual).Forchildren,therecommendationistwotenthsofagramofsoilperday.Thereisa possibilityfordermalabsorptionofpollutantsfromsoilonhands.However,thisislimitedto therelativelysmallnumberofpollutantsthathavethisability. Thereislittletonoinformationavailableonpollutantsinsoilfromnaturalgasoperations(R. Chesson, Colorado Oil and Gas Conservation Commission). This precludes the possibility of doingariskassessmentforanactual,pollutedsite. The Colorado Oil and Gas Conservation Commission requires that soil contaminated with greaterthan1,000milligramsoftotalpetroleumhydrocarbonsperkilogramofsoilbetreated or removed and properly disposed of. Soil with this level of contamination resulting from a condensate spill was assessed for risk as a way of illuminating what kind of hydrocarbon concentrations would present a significant threat. Only benzene, toluene, and m,pxylenes wereincludedbecausetoxicityvaluesforotherhydrocarbonsincondensatearenotavailable.
47
The EPArecommended values for a reasonablemaximum exposure were used in the calculation,whichisshowninAppendixJ.Thehazardquotientsare0.011,0.0031,and0.0014 respectivelyforbenzene,toluene,andm,pxylenes.Thesumofthehazardquotientsis0.016. All of these numbers are well under one, which is the threshold for unacceptable exposure leading to noncancer effects. The cancer risk due to benzene is one chance in one million, whichmeetstheEPAregulatorygoalforacceptablecancerrisk.That1,000milligramsoftotal petroleum hydrocarbons per kilogram of soil presents an acceptable health threat is not surprisingthelevelwasdeliberatelysetatasafevalue. RECOMMENDATIONS Asdescribedinthisreport,therearenumerousgapsanduncertaintiesinourunderstandingof pollution from natural gas operations in Garfield County. The motivation for the following recommendations is to reduce these uncertainties and fill data gaps. Better information is neededinordertomakeacompleteandaccurateevaluationofthreatstohumanhealth. ConductathoroughstudyofairemissionsduringdrillingAtleastthreewellpadsshould be selected for study, with different companies, drilling approaches, and operations represented. A scenario believed to be a worst case from an air emission standpoint should be among the three. Daily 24hour grab samples of ambient air should be collected at the boundary of the well pad at each of sixteen equallyspaced compass directions. Samples should be analyzed for volatile organic compounds, carbon monoxide, and ozone. The duration of the study at each well pad should cover the drilling and completion of at least three wells. Onsite measurements of wind speed, wind direction, and temperature should be made at 15minute intervals in order to understand elements of the basic meteorology of the site. The results of this study shouldprovideamuchimprovedpictureofairpollutantconcentrationsrepresentative ofdrillingoperationsingeneral. Identify the components of hydraulic fracturing fluids and other materials used in well installationThere is considerable public concern about the chemical composition of thesematerials.Theindustrymaintainsthathydraulicfracturingfluidsarenotathreat but has offered only incomplete evidence in the form of material safety data sheets, which are not required to include the identity of all components. Disclosure of the fracingfluidcompositionwouldallowmonitoringofgroundwaterforfluidcomponents, evaluation of the potential health threat, and the potential for improved public acceptance of the industry. Similarly, the chemical composition of used drilling muds andproducedwatershouldbedeterminedandrevealed. EstablishamonitoringprogramforprotectionoftheprivatewellusersAsdescribedin thesectiononwater,GarfieldCountypresentsadifficultsituationtoassessforground water contamination. The subsurface is complex, and the behavior of any specific contaminantplumemaymakecontaminationdifficulttolocate.Itmaynotbepossible
48
toimplementagroundwatermonitoringnetworkatafeasiblecostthatwouldprovide ahighdegreeofconfidencefordetectingcontamination. However,theprincipalgoalistodetectanycontaminationthatmaybeaffectingprivate wells,whereexposuretocontaminantswouldactuallytakeplace.Themostdirectway toaccomplishthisgoalistocollectquarterlyorsemiannualsamplesfromprivatewells. Themonitoringprogramshouldbefunded,organized,andoperatedbyeitherGarfield County or the Colorado Department of Public Health and Environment. Well water should be analyzed for methane, benzene and other volatile organic compounds, and selected components of hydraulic fracturing fluids that have the greatest potential to affecthumanhealth. InspectionofsurfacesoilsatwellpadsafterdrillingactivitiesarecompleteColoradoOil and Gas Conservation Commission staff should inspect well pads for contamination of surface soil and collect samples of any areas where contamination is suspected based on observation of discolored soil or odor. Inspections should occur frequently and in particularshouldbetimedsoastoallowobservationofemptypitspriortobackfilling when drilling operations area complete. Samples should be analyzed for petroleum hydrocarbonsandanymetals(suchasbarium)thatareusedoroccurindrillingfluids. Soils exceeding the spill cleanup criterion for total petroleum hydrocarbons or typical action levels for metals should be remediated. This practice should ensure that landownerexposuretosoilcontaminantsisminimized. Use green completions and applicable best management practicesThe industry can accomplish a significant reduction in pollutant emissions by continuing to adopt more environmentallyfriendlypractices.Asdescribedinthisreport,thecaptureofgasduring flowbackshouldeliminatethemostseriouspollutantemissions.
49
PART 2. HEALTH STUDY
PREMISESOFTHESTUDY ebeganthisstudywiththepremisethatwewouldusetheWorldHealthOrganizations (WHO) definition of human health; i.e., that health is broader and more inclusive than physicalwellness.Health,bytheWHOdefinition,isinclusiveofthesocialcultural,spiritual andpoliticaleconomicconditionsofthecommunityanditsmembers.Thus,bothquantitative andqualitativedatacollectionincludedinformationonsocioeconomicandbehavioralimpacts andoutcomes.Anindepthstudyofthesocioeconomicimpactsofgrowthtakingplaceonthe WesternSlopeofColoradowasrecentlypublishedbytheMesaStateCollegeNaturalResource andLandPolicyInstitute(September2007,Rediferetal.).Itisbeyondthescopeofthecurrent reporttodescribeorsignificantlyaddtotheinformationcontainedintheSocioeconomicsof growthreport.
Figure 8. World Health Organization Definition of Human

Social-Cultural
Physical
HUMAN HEALTH
Spiritual
Political-Economic
This study represents a snapshot in time of the health of Garfield County residents. This snapshotprovidesabaselineassessment,againstwhich,futurechangesinhealthparameters maybemeasured.Trenddataareprovidedinthisreport,wherepossible. Thisisadescriptivestudy.Statisticalanalysesarepresentedwhenandwhereappropriate. Whereas, it is possible, in many cases to discuss correlations between health status and residence,itisnotpossibletomakedefinitiveconclusionsaboutcausation,giventhedatathat areavailableatthistime.
50
METHODOLOGY TheHealthStudyhasthreemajorcomponents: QualitativeData:Communityperceptionsandconcernsregardinghealthandwellbeing QuantitativeData:Healthoutcomesdata SelfreportedHealthStatus:Householdsurvey Theoverarchingquestionsbeingaskedinthisstudyarethefollowing; 1)IsthehealthofGarfieldCountyresidentsdifferentthanthehealthofresidentsofDelta, MesaorMontroseCounties? 2)IsthehealthofresidentsofareasofGarfieldCountythatareheavilyimpactedbythe naturalgasindustrydifferentfromthehealthofresidentsoflessimpactedareasofGarfield County? Description of study area and comparison counties. The three comparison counties, Delta,Mesa,andMontrose,wereselectedforthefollowingreasons: All four counties (including Garfield County) are located on the Western Slope of Colorado,andthus,sharesimilarsocialandpoliticalcultures. All four counties have experienced energy and mining activities, although to different degrees(seeBackgroundsection).Asaresultofthenaturalcyclesoftheseindustries, all four counties have experienced similar economic cycles of growth and recession (boomandbust),andtheaccompanyingenvironmentalandsocialimpacts. Figure9.
51
Thesomewhatoverlappinghealthcarenetworksandserviceareasallowedustoacquire
comparativehealthdataforresidentsof the four counties. Demographic data for Garfield County and the three comparison counties were largely obtainedfromtheColoradoDepartmentofLocalAffairs,StateDemographyOffice.Thesedata areavailableonlineathttp://dola.colorado.gov.
Table 8. County Profiles: 2000 & 2006 Data*
Average Household Size
County Delta 2000 2006 Garfield 2000 2006 Mesa 2000 2006 Montrose 2000 2006
July Population
Total Households
Births
Deaths
Birth/Death Ratio
Similarity Ranking
28,009 30,676
11,058 12,225
2 2
296 347
330 303
0.90 1.14
44,267 53,020
16,229 19,587
3 3
733 893
242 273
3.03 3.27
--
117,656 135,468
45,823 53,416
2 2
1,485 1,853
1,109 1,268
1.34 1.46
33,666 38,903
13,043 15,261
3 3
435 503
296 361
1.47 1.39
http://dola.colorado.gov/demog_webapps/profile_county; accessed 4/10/2008; Division of Local Government, StateDemographyOffice *Countyprofilevariables Julypopulation:July1ofeachcalendaryear,preparedbytheDemographicSectionincooperationwith theU.S.BureauoftheCensus. Total Households: Households (total occupied housing units are estimated from total housing units, householdpopulationandpersonsperhousehold. Average Household Size: Computed by dividing the household population (at July 1) by the number of households.
52
Births:ThesedataarereceivedfromtheDepartmentofHealthonafiscalyear(July1toJune30)basis. Deaths:ThesedataarereceivedfromtheDepartmentofHealthonafiscalyear(July1toJune30)basis. Birth/DeathRatio:GarfieldCountyhad>3Xthenumberofbirthstodeathsinboth2000and2006.This wasincontrasttotheotherthreecounties,allofwhichhad<1.5Xthenumberofbirthsasdeaths. Similarity Ranking: A ranksum algorithm (JD Boise, Jr. et al. Rad Res 167:711726, 2007) was used to identifywhichofthethreecountiesismostsimilartoGarfieldCountydemographically.Rankvaluesfor9 socioeconomicvariables(populationin2000,populationin2006,numberofhouseholds,householdsize, birth/deathratio,%changeinpopulationfrom2000to2006,medianhouseholdincome,populationage groupings, and percentage of children living below poverty) were determined, based on similarity to GarfieldCounty.Rankvaluesforallvariablesweresummed,withthelowestsumrepresentingthemost similaritytoGarfieldCounty.
Some of the socioeconomic variable comparisons are illustrated in the graphs that follow. All of the data for these graphs were obtained from the Department of Local Affairs, State Demography Office website(seeabove). Garfield,Montrose,andDeltaCountypopulationsaremostsimilar,whileMesaCountyspopulationis 2.5toalmost4.5timesthepopulationoftheothercounties.

160,000
53
140,000
120,000
100,000 Delta Garfield Mesa Montrose
80,000
60,000
40,000
20,000
0 April 2000 USCB Count July 2001 SDO Est. July 2002 SDO Est. July 2003 SDO Est. July 2004 SDO Est. July 2005 SDO Est. July 2006 SDO Est.
USCB:UnitedStatesCensusBureau SDO:StateDemographersOffice
Figure10.PopulationEstimatesbyCounty,20002006. Allfourcountieshaveundergonedecreases,andsubsequentincreases,inpopulationoverthe pastdecade.Currently,allfourcountiesareexperiencingpopulationgrowth. Figure11.AverageAnnualPercentChangeinPopulation,20002006.

6
Percent change
Delta Garfield Mesa Montrose COLORADO
0 00-01 01-02 02-03 03-04 04-05 05-06
Year
54
COLORADO
Montrose
Mesa
Garfield
Delta
10
15
20
25
30
35
Percentage of the Total Population (Year 2000 Census Data)
Under 15
15 to 24
25 to 44
45 to 64
65+
Figure12.PopulationAgeasaPercentageofTotalPopulation ThelargestpopulationbyageinGarfieldCountyisthe25to44yearoldagegroup.Garfield County tends to have a younger population, overall, than the comparison counties. The age demographicoftheGarfieldCountyspopulationismoresimilartothatofthestate,overall. Figure13.RaceandEthnicityasaPercentageofTotalPopulation COLORADO
Montrose
Mesa
Garfield
Delta
10
20
30
40
50
60
70
80
90
Percentage of the Total Population (Year 2000 Census Data) White Hispanic* Black** AI/AN*** Asian**** Other Two or more races
55
Greater than 80% of the individuals who live in Garfield County are nonHispanic white, with Hispanic individuals making up <15% of the residents. Garfield County is most similar to MontroseCounty,withrespecttotheraceandethnicitymakeupofthepopulation. Thefollowingmapsillustrateotherdemographiccomparisonsamongthefourcounties:
(http://dola.colorado.gov/demog_webapps/profile_county;accessed4/10/2008;DivisionofLocal Government,StateDemographyOffice).
Figure14.MedianHouseholdIncomebyCounty,2000Census
Figure15.MedianHouseholdIncomebyCensusTract,2000Census
56
AccordingtoU.S.CensusBureaudatafor2000,medianhouseholdincomeinGarfieldCountyis higherthaninanyofthecomparisoncounties.However,thecensustractdataindicatethatthe western part of Garfield County, where more of the natural gas activity is located, has lower median household incomes than the eastern half of the county. Census tract data for Mesa Countyshowshighermedianhouseholdincomesforthesouthernhalfofthecounty.Thereis less discrepancy for median household incomes among census tracts for Delta and Montrose Counties. The2000U.S.CensusshowedthatfewerpeoplewerelivingbelowpovertyinGarfieldCounty than in Delta, Mesa or Montrose Counties. This is generally true, also for the percentage of childrenlivinginpoverty,althoughpercentagesdifferbycensustractwithinthecounties(see below). Figure16.PercentofPersonsBelowPovertybyCounty,2000Census
57
Figure17.PercentofPersonsBelowPovertybyCensusTract,2000Census
Figure18.PercentofChildrenBelowPovertybyCensusTract,2000Census
58
PerceptionsandConcerns(QualitativeDataCollection). DataCollectionProcess Perceptions about community health and priority health concerns were collected through a process of public meetings, focus groups, and individual interviews that took place over a periodofapproximatelytwoyears.Aseparateprocessofcollectingcomplaintsandconcerns occurredattheGarfieldCountyHealthDepartment,wheretheEnvironmentalHealthManager kept a running phone log. Information in that log, as it relates to this studys objectives, is includedinthedataanalysis. CommunityMeetings Two public meetings were held early in the data collection process; one each in Rifle (at the Fairgrounds)andinBattlementMesa(attheCommunityCenter).TheRiflemeetingwasheld duringthedayonaSaturday,whiletheBattlementMesameetingwasheldintheeveningona weeknight. The objective was to give community members a choice of venue and time of day/dayoftheweektoallowforgreaterparticipation.Bothmeetingswereadvertisedthrough several venues; in the local newspapers, through flyers, at the Garfield County Health Department,andthroughannouncementsatGrandValleyCitizensAlliance(GVCA)meetings. A total of 36 individuals (including Garfield County Health Department staff) attended the meetings. The format for the meetings was an initial presentation by Drs. Teresa Coons and RussellWalker,describingthehumanhealthassessmentstudy,followedbyanopportunityfor attendeestoaskquestionsandstateconcerns. KeyInformantInterviews Interviews (either inperson or by telephone) were conducted by study staff with individuals who requested the opportunity to tell their stories. All of these interviews were voluntary andselfselected.Individualslearnedoftheopportunitytobeinterviewedandsharepersonal experiences and concerns through newspaper and radio stories about the study, wordof mouth,andasaresultofattendingcommunityorGVCAmeetings. FocusGroups Six focus groups were held, with a total of more than 60 individuals participating. The focus groups were held at different times and locations, over a time period of approximately one year. Individuals attended by invitation. Participants were not compensated, nor did they receive any incentive for their participation. Focus group discussions were recorded and transcribedforanalysis.Thefollowinggroupsofcommunitycitizenswereeachrepresentedby afocusgroup: Electedofficials(municipalandcountyelectedofficialsandadministrators) GrandValleyCitizensAlliance(localconcernedcitizensorganization) Elderlyresidents(BattlementMesaAssistedLivingfacilityresidents)
59
HealthcareProviders(ThefocusgroupwasheldatGrandRiverMedicalCenterinRifleand wasadvertisedtoallmedicalstaffandalliedhealthcareproviders.Participationinthisgroup waspoor;wefollowedupwithasurveyoflocalproviders.) GarfieldCountyHumanServicesCouncil(Onehourofaregularmeetingwasdevotedtothe studyfocusgroup.Awiderangeofcountyserviceagencies,bothprofitandnonprofit,were represented.) Garfield County School Districts (Separate groups were held for School District Re2, comprising Rifle, Silt and New Castle, and School District Re16, comprising Parachute and Battlement Mesa. Invitations to participate were coordinated through the districts administrativeoffices.) Focus group participants were instructed to think of health broadly, i.e., not just physical health, but also psychosocial health and wellbeing. All focus groups were guided discussionsconductedbyDr.TeresaCoons,usingacommonsetofdiscussionquestions. FocusGroupQuestions 1.HowwouldyouratethehealthofpeopleinGarfieldCountyand/oryourcommunityrelative tothehealthofpeopleinotherColoradocommunitiesorothercommunitiesinwhichyouve lived?Note:Iamsimplyaskingforyourperception,basedonyourownexperience. 2.Whatdoyouconsidertobethepriorityhealthrelatedissuesinyourcommunity?Note:I am using the term, health, in the context of not only specific physical ailments, but also the broaderconceptofwellbeing. 3. Who is most affected by health issues in your community? Are there particular age or demographicgroupsthataremoreatriskorhavemorehealthproblems? 4. To what environmental or societal factors (if any) would you attribute any of the priority healthissuesthatyouveidentifiedinyourcommunity? 5.Whathealthissuesconcernyouthemost?(Questionaskedofeachparticipantindividually.) Note: Healthcare providers were also asked to identify the health conditions that they saw mostfrequentlyintheirpractices. ComplaintPhoneLogs The phone log maintained by the Garfield County Environmental Health Manager documents telephonecallsreceivedfromearly2003through8/2007.Informationinthelogincludesdate ofcall,typeofcomplaintorconcern,regionorlocation(ifapplicable),callernameandaddress, complaint or concern narrative and resolution of the call. The list of callers overlapped, to some extent, with the list of individuals who asked to be interviewed, and some of the individuals were repeat callers. The health concerns expressed in the phone log reflect the concernsdocumentedinthepreviouslydescribeddatacollectionprocess.
60
QuantitativeHealthData.Healthoutcomesdatawerecollectedfromanumberofsources forGarfieldCountyandcomparisoncounties.Thesearchforhealthoutcomesdatawasdriven bytwoobjectives:1)tocompleteasnapshotintimepictureofthehealthofGarfieldCounty residentsincomparisontothehealthofresidentsinthecomparisoncounties,and2)toobtain statistical data that could be used to respond to the concerns voiced by Garfield County residentsduringthequalitativedatacollectionprocess.Thus,totheextentthatthedatawere available, we collected statistical information on the prevalence of conditions such as cancer andasthmaandthepredominantcausesofmortalityandmorbidityinGarfieldCounty. The following data were obtained from the Colorado Department of Public Health and Environment (datasets represent the most current data available at the time this report was written): Deathstatistics:ratesandleadingcausesofdeath(19902006) Birthdefects:typesandrates(20002006) Adolescenthealthmeasures(2007) Reportableconditions(19982006) WestNilevirus(20022007) Cancerstatistics(19922005) BehavioralRiskFactorStudySurvey(BRFSS)data(20002005) o Generalhealthstatus(physical&mental) o Diabetes,asthma o Smoking,weight o Healthinsurance Injuryhospitalizationanddeath:causesandrates(20012003) Hospitalandoutpatientdatawereobtainedfromthefollowingsources: Colorado Hospital Association, DRGbased hospital discharge data (2000 through 1st quarter2006) Emergencyroomdata o Grand River Medical Center (located in Rifle, CO and serving western Garfield County)
61
o ValleyViewHospital(locatedinGlenwoodSprings,CO;primarilyservingeastern GarfieldCounty) o RockyMountainHealthPlans*(RMHP)memberdatafor4counties o St.MarysHospitalCareFlightdata RMHP hospital inpatient, outpatient and ambulatory member data 4 county comparison ByDRGcategory Focusonrespiratoryconditions
*RMHP is a nonprofit, Coloradobased health insurance company that provides healthcare coverage for a significantnumberofWesternColoradoresidents.RMHPprovidesafullrangeofhealthplans,includingMedicare supplementandMedicaidcoverage.
Selfreported Health Status: Household Survey. A targeted health survey was administered to Garfield County residents by trained interviewers. The inhome surveys capturedinformationaboutthegeneralhealthandhealthriskfactorsofresidents,aswellas information about specific health conditions that were identified as priority concerns during focusgroupdiscussions.Thisinformationwasintendedtoprovideamoreobjectivemeasureof the health status of community residents, and a means of conducting within County comparisons. A randomly selected sample of households from throughout Garfield County was surveyed. Thesamplingprotocolwasdesignedtoprovideazipcodebasedsamplingpoolof24%ofthe households in Garfield County that have listed telephone numbers (4% of households in zip codeareasthatarehighlyimpactedbythepresenceofnaturalgasindustryoperationsand2% of households in the zip code areas that have little or no natural gas activity); a sample that would give a 95% confidence level(5% error rate) for the study results. Randomly selected phonenumbers(andtherespectivemailingaddresses)fromeachGarfieldCountyzipcodearea wereobtainedfrominfoUSA(infoUsa,Inc.,5711South86thCircle,P.O.Box27347,Omaha,NE 68127), a marketing research firm that specializes in healthrelated community surveys. In order to achieve the desired number of completed household interviews, we requested 15 phone numbers for each number that we hoped to complete. Drawbacks to this survey approach include the unwillingness of many people to respond to telephone calls from unknownindividuals(eitherbecausetheydontwanttobebotheredorbecausetheyareafraid thatthecallisfromatelephonesolicitor),andthefactthatmanyindividualsarehardtoreach at home during convenient hours. We hoped to avoid the problem of individuals being unwillingtospeakwithusonthetelephonebyprecedingourinitialphonecallwithaletterto thehouseholdthatexplainedthenatureofthesurveyandrequestedreturninformationonthe
62
bestday(s)andtime(s)ofdaytocall.Apostagepaidreturncardwasincludedintheletterfor thispurpose. Because of the difficulty of reaching people by phone (see above), and a relatively low initial response rate, we also developed a mail survey. We split the sample of randomly selected phone numbers/addresses for each zip code area, and mailed a written survey to half of the selected respondents. Because we used the same randomly selected sample to create a mailing list, we have no reason to suspect that there is any identifiable significant difference betweenthegroupthatreceivedphonecallsfromaninterviewerandthegroupthatreceiveda surveyinthemail.Byusingthetwomethodsbothofwhichcanhavelowresponserateswe attemptedtomaximizeourabilitytoreacharepresentativesampleofhouseholdsinGarfield County.Thewrittensurveycontainedquestionsthatwereessentiallythesameasinthephone survey,modifiedonlytoallowtherespondenttodocumenttheirresponsesonpaper.(Acopy ofthesurveydocumentsareprovidedinAppendixT.) Telephone surveys were conducted by a limited number of trained interviewers, using an ACCESS(2002)databasethatalloweddirectdataentryduringthetelephoneconversation.All samplehouseholdinformationandsurveydocumentation(e.g.,attemptedphonecontactsand outcomes,reasonsforrefusingtoparticipateinthesurvey)andwrittensurveyresponseswere enteredintothesameACCESSdatabase. Toassurethatweachievedarepresentativecrosssectionofthepopulationdemographic,adult respondentswereaskedtocompletethesurveyquestionsforallmembersoftheirhousehold, regardless of age or family relationship. The telephone survey required, on average, 20 minutes to complete depending on the number of household members and the range of reportedhealthconditions. Inaddition,toassurethatwedidnotexcludeprimarilySpanishspeakingindividualsfromthe surveyed population, the survey document was translated into Spanish and then back translatedbyaprofessorofSpanishlanguage,who,inparticular,testedfortheappropriateness ofdialectandidiomforthisgeographicregion.Astelephonecontactwasmadewithpotential surveyrespondents,interviewersdocumentedhouseholdswithinwhichtheprimarylanguage wasSpanish.ThosehouseholdswererecontactedbyanativeSpanishspeakingpublichealth nurse, who was an employee of the Garfield County Health Department. Standardized interviews were conducted in Spanish, results were recorded in English on survey forms, and thedatawereenteredintothesurveydatabase. ACCESS data tables were downloaded into EpiInfo Version 3.4.1, Centers for Disease Control andPreventionforanalysis.(PleaseseeAppendixUfordataanalysisprotocolsandoutcomes.)
63
RESULTS Perceptions and Concerns. (Qualitativedatafromfocusgroups,interviews,publicmeetings andcomplaintreports) KeyInformantInterviews Atotalof23interviewswereconductedbetweenOctober2005andOctober2007;16(69.6%) wereconductedinperson,7(30.4%)wereconductedbytelephone. Table9.IntervieweeDemographics
Distance of Residence from Natural Gas Extraction or Production Activities
Gender
Residence
Residence Time in the County Range: 9 months to 20 years
How Did Individual Hear About Study?
Female: 12 (52.2%) Male: 11 (47.8%)
Rifle: 5 (21.7%) Parachute: 10 (42.6%) Silt: 8 (34.8%)
Range: 150 feet to 6.5 miles
Meeting (public meeting or GVCA* meeting): 18 (78.3%) Newspaper article: 4 (17.4%) Word of Mouth: 1 (4.3%)
Mean: 9.8 years Median: 9.5 years
Mean: 0.93 miles
Median: 0.25 miles
* GVCA: Grand Valley Citizens Alliance
KeyWordsandCommonThemesAnalysis Transcriptions from the key informant interviews and focus group discussions were reviewed forkeywordsandthemes,whichwerethengroupedintothematiccategories.Thecomments andconcernsof119individualsarerepresentedbelow. PhysicalHealthIssues/Concerns.(Note:Thefollowingconcernsmayormaynothavebeen associatedwithenvironmentalexposures.) Increaseinorexacerbationsofallergiesandasthma o Relatedconcerns:coughing,wheezing,otherrespiratorycomplaints Generalizedchemicalsensitivities Fibromyalgia/chronicpain o Relatedconcern:chronicfatigue,lethargy Chroniccolds o Relatedconcerns:concernaboutcompromisedimmunesystems
64
Headaches, dizziness, burning/itching eyes, nausea/vomiting, sinus problems most oftenattributedtoodors Burning/itchingskin Mentalhealthissuessuchasstress,depression,anger,inabilitytosleep Cancer (adrenal cancer, brain tumors, unknown/presumed cancers or fear of developingcancer) Lossofvoiceorspeechproblems Trauma/workrelatedinjuries Agerelatedillnesses Diabetes Obesity Perceptions that preexisting health conditions have been exacerbated; people feelingworsethaninthepast Social/CommunityIssuesandConcerns Increaseinchildandspousalabuse;childneglect;stressedfamilyrelationships Alcoholabuse(especiallyamonghighschoolstudents) Drugabuse(especiallymethamphetamineuse) Highsuiciderate Increaseinsexuallytransmitteddiseasesrelatedtoincreaseintemporaryworkers Lackofhealthinsurance o Relatedconcerns:lackofdentalcareforchildrenandpreventivecare Accesstohealthcareandmentalhealthservices Numberoflowincomefamilies Growthissues:Availabilityofhousingandcommunityservices,increaseinlowincome families,culturalclash(longtimeresidents,industryworkers),traffic,publicsafety EnvironmentalConcerns Noise Odors Dust Toxicchemicalsinwaterandair Wildlife/DomesticAnimals Reportedchangesinherdanimalreproductivepatterns/illnesses Decreasesinbird,insect,anddeerpopulations
65
Garfield County Health Risk Study - Community Interviews/Focus Groups
200 180 160 140 120

Total Comments 100
80 60 40 20 0 Environment Health Societal / Health

Comment Categories
Societal
Wildlife / Lifestock
Figure19.CommunityInterviews/FocusGroupComments GarfieldCountyEnvironmentalHealthPhoneLogComplaints For the documented time period, the log contains 372 calls. The largest number of complaints/concernswasrelatedtoodors(93calls).Othercomplaints/concernsincludedthe following: Roadconditions(31calls) Groundwaterconcerns(25) Noise(25) Safety(24) Dust(23) Spills(23) Propertydamage(18) Visualappearances(14) Permitting(14) Surfacewaterconcerns(13) Specifichealthconcerns(13) Traffic(10) Trespassing(9) Surfaceuseagreements(9) Flaring(7) Generalenvironmentalconcerns(7) Industryworkerspeeding(5) Stormwater(3)
66
Fire(2) Pitwater(productionwater)containment(2) Smoke(1) Trash(1)
Quantitative Health Data. FinaldatacollectionforthisreporttookplaceinJanuary2008, unlessotherwisenoted.Thus,thesedataarethemostcurrentlyavailableatthatpointintime. VitalStatistics Tables 8 through 11 present vital statistics data and trends for 1996 and 20022006 by county.Table12providesselectedbirthstatisticsbyCountyfor2006,specifically. Accordingtotheannualbirthratedatafortheyears1996and20022006,birthratesfor allcountieshavebeenessentiallystableduringthisperiod.TheexceptionisMontrose County,forwhichtherewasadropinthebirthratebetween1996and2002,afterwhich the birth rate stabilized between 2002 and 2006. Garfield County had the highest averageannualbirthrateduringthisperiod. The percentage of teen mothers has been stable over the study period, with the exception of Delta County, which saw a drop in the percentage of mothers who are between the ages of 10 and 17 years between 1996 and 2002. Based on the average annual rate for the period 20022006, Garfield County had the lowest rate of teen mothersamongthecomparisoncounties. Therateoflowbirthweight(<2,500grams)babiesamongthefourcountiesissimilar. MesaCountyhadthelowestrateofmotherswhodidnotreceiveprenatalcareduringthe firsttrimesteroftheirpregnancies.Theratesfortheotherthreecountiesweresimilar. The highest rate of mothers who had less than 12 years of formal education was in GarfieldCounty. GarfieldCountysrateofneonatalandinfantdeathswassimilartothatofMesaCounty, andhigherthanthatofDeltaandMontroseCounties. The ageadjusted, total death rate for Garfield County was comparable to the age adjusted,totaldeathrateforDeltaCounty,andlowerthantheageadjusted,totaldeath rates for Mesa or Montrose Counties. The graph in the section below provides a comparisonofcrude,totaldeathratesforthefourcountiesovertheperiodof1990 2006.GarfieldCountyhadthelowestcrudedeathrateforthepast2decades.
67
Table 10. Vital statistics by County 1996, 2002-2006 1996 Total population Total Births Number Birth Rate Age of Mother 10-17 Number Percent of Total Births Low Weight Births (<2,500 grams) Number Percent of Total Births Prenatal Care Later than 1st Trimester Number Percent of Total Births Mother's Education <12 years Number Percent of Total Births 89 28.6 84 26.8 74 22.5 88 28.8 94 29.7 74 22.7 23 7.4 24 7.6 28 8.4 31 9.9 18 5.7 19 5.7 312 12.1 317 10.9 332 11.2 25,718 2002 29,194 2003 29,666
Delta County 2002-2006 2004 30,073 2005 30,264 2006 30,672 Annual Average 29,974
350 11.6
338 11.2
344 11.2
336 11.2
19 5.4
14 4.1
11 3.2
16 4.8
37 10.6
26 7.7
21 6.1
27 8.1
85 26.6
97 30.5
103 32.2
91 28.3
95 28
77 23.4
75 22.6
81 24.7
68
Infant Deaths Number Rate Neonatal Deaths Number Rate Total Deaths Number AgeAdjustedRate 308 798.3 316 766.1 338 801.8 278 652.6 331 759.5 323 747.6 745.2 317 * * * * 3 9 * * 0 0 * * * * * * * * 5 15.1 3 8.6 0 0 * * *
Source: Health Statistics Section, Colorado Department of Public Health and Environment. *indicates one or two events Birth rates are per 1,000 total population. Infant and neonatal mortality rates are per 1,000 live births. Infant death is death in the first year of life. Neonatal death is death occurring prior to the 28th day of life. Total death rates are per 100,000 population. Age-adjusted rates are adjusted to the 2000 U.S. population using the direct method applied to 10-year age groups.
69
Table 11. Vital statistics by County 1996, 2002-2006
Garfield County 2002-2006 Annual Average 49,740
1996 Total population Total Births Number Birth Rate Age of Mother 10-17 Number Percent of Total Births Low Weight Births (<2,500 grams) Number Percent of Total Births Prenatal Care Trimester Later than 1st 37 6.1 22 3.6 604 16 37,678
2002 47,439
2003 48,395
2004 49,319
2005 50,669
2006 52,878
855 18
871 18
803 16.3
816 16.1
945 17.9
858 17.2
32 3.7
30 3.4
25 3.1
24 2.9
31 3.3
28 3.3
63 7.4
70 8
65 8.1
57 7
69 7.3
65 7.6
Number Percent of Total Births Mother's Education <12 years Number
157 26.2
326 38.2
217 24.9
197 24.6
266 32.7
346 36.7
270 31.6
144
281
311
276
270
273
282
70
CommunityHealthRiskAnalysisofOilandGasIndustryImpactsinGarfieldCounty Percent of Total Births Infant Deaths Number Rate Neonatal Deaths Number Rate Total Deaths Number Age-Adjusted Rate 229 815.8 279 819 310 904.3 271 759.9 268 729.9 289 729.4 283 786.6 5 8.3 3 3.5 3 3.4 6 7.5 3 3.7 * * 3 4 6 9.9 9 10.5 5 5.7 8 10 3 3.7 3 3.2 6 6.5 24 33 35.8 34.6 33.3 29.2 33.1
Table 12. Vital statistics by County 1996, 2002-2006 Mesa County
71
2002-2006 Annual Average 128,299
1996 Total population Total Births Number Birth Rate Age of Mother 10-17 Number Percent of Total Births Low Weight Births (<2,500 grams) Number Percent of Total Births Prenatal Care Trimester Later than 1st 102 7.5 99 7.2 1,366 12.7 107,536
2002 122,467
2003 125,077
2004 127,816
2005 130,662
2006 135,473
1,553 12.7
1,662 13.3
1,711 13.4
1,819 13.9
1,880 13.9
1,725 13.4
66 4.2
75 4.5
69 4
88 4.8
87 4.6
77 4.5
123 7.9
126 7.6
146 8.5
141 7.8
138 7.3
135 7.8
Number Percent of Total Births Mother's Education <12 years Number Percent of Total Births
195 14.3
242 15.6
257 15.5
210 12.3
255 14
292 15.6
251 14.6
295 21.6
295 19.1
370 22.4
276 22.1
272 20.6
415 22.2
366 21.3
72
Infant Deaths Number Rate Neonatal Deaths Number Rate Total Deaths Number Age-Adjusted Rate 1,024 952.2 1,220 996.2 1,232 985 1,162 909.1 1,268 970.4 1,268 936 1,230 958.7 12 8.8 13 8.4 7 4.2 10 5.8 3 1.6 7 3.7 8 4.6 16 11.7 16 10.3 10 6 12 7 6 3.3 9 4.8 11 6.1
Table 13. Vital statistics by County 1996, 2002-2006 Montrose County
73
2002-2006 Annual Average 37,056
1996 Total population Total Births Number Birth Rate Age of Mother 10-17 Number Percent of Total Births Low Weight Births (<2,500 grams) Number Percent of Total Births Prenatal Care Trimester Later than 1st 41 8.5 30 6.2 484 15.9 30,482
2002 35,433
2003 36,122
2004 36,932
2005 37,883
2006 38,908
440 12.4
458 12.7
496 13.4
521 13.8
493 12.7
482 13
19 4.3
23 5
22 4.4
15 2.9
19 3.9
20 4.1
31 7
32 7
46 9.3
42 8.1
32 6.5
37 7.6
Number Percent of Total Births Mother's Education <12 years Number Percent of Total Births
146 30.4
142 32.4
122 27.2
176 35.9
222 42.9
198 40.5
172 36.1
143 29.9
133 31.1
133 30.4
138 29.4
131 26.1
120 26.3
131 28.6
74
Infant Deaths Number Rate Neonatal Deaths Number Rate Total Deaths Number Age-Adjusted Rate 304 997.3 373 1053 336 930.2 339 917.9 330 871.1 365 938.1 349 940.1 * * 0 0 3 6.6 * * 0 0 3 6.1 * * 3 6.2 * * 3 6.6 * * * * 3 6.1 * *
75
Figure20.FirstTrimesterPrenatalCarebyCounty(Percent) HealthyPeople2010Goal:Increasethepercentofinfantsborntopregnantwomenreceivingprenatalcarebeginninginthefirst trimesterto90.0percent. Countyratesarefortheyear2005.Thestaterateis80.1percent.
76
Figure21.LowBirthWeightbyCounty(Percent) HealthyPeople2010Goal:Reducethelowbirthweightratetoanincidenceofnomorethan5.0percentofallbirths. Countyratesare20032005threeyearaverages.Thestaterateis9.1percent.
77
Figure22.VeryLowBirthWeightbyCounty(Percent) HealthyPeople2010Goal:Reducetheverylowbirthweightratetoanincidenceofnomorethan0.9percentofallbirths. Countyratesare20032005threeyearaverages.Thestaterateis1.3percent.
78
Figure23.NeonatalMortalitybyCounty(Rateper1,000) HealthyPeople2010Goal:Reducetheneonatalmortalityratetonomorethan2.9deathsper1,000births. Countyratesare20012005fiveyearaverages,exceptforBroomfieldwhichis20022005only. Thestaterateis4.4.
79
Figure24.InfantMortalitybyCounty(Rateper1,000births) HealthyPeople2010Goal:Reducetheinfantmortalityratetonomorethan4.5deathsper1,000births. Countyratesare20012005fiveyearaverages,exceptforBroomfieldwhichis20022005only. Thestaterateis6.1.
Table 14. Selected birth characteristics by County, 2006
Low Birth Age of Mother (%) 1017 3.2 3.3 4.6 3.9 1819 9.3 7.5 8.2 8.3 2029 66.3 59 61.9 59 Weight (%) <2,500 grams 6.1 7.3 7.3 6.5
Very Low Birth Weight (%) <1,500 grams * 1 0.9 2
County Delta Garfield Mesa Montrose
Total Live Births 344 945 1,880 493
30+ 21.2 30.2 25.2 28.8
*indicates one or two events
Education of mother (%) Smoking during Pregnancy (%) 11.7 13.8 17.5 8.9
Prenatal Care (%) No care 1.9 1.8 0.9 0.8
Pre-term births (%)
County Delta Garfield Mesa Montrose
after 1st trimester 32.2 36.7 15.6 40.5
<37 weeks 7.8 8.5 7.6 9.1
<12 years 21.8 28.9 22.1 24.3
81
Figure25.CrudeTotalDeathRates Delta,Garfield,Mesa,MontroseCounties
1400
1200
1000
Rate per 100,000
800
600
400 Delta Garfield 200 Mesa Montrose 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 Year 1999 2000 2001 2002 2003 2004 2005 2006
Average,annual,ageadjusteddeathratesbycauseofdeathareshowninFigure26belowforthefour counties.Cardiovasculardiseaseisthebiggestcauseofdeathinallfourcounties,followedbyother heartrelated disease and cancer. The rates among the four counties are similar. Please see the CauseofdeathtableinAppendixKforthenumbersofactualdeathsbycountyandcause. Figure26.AverageAnnualAgeadjustedDeathRatesbyCause,20032006
Cause of Death
Work-Related Injury Homicide and Legal Intervention Suicide Chronic Lower Respiratory Diseases Other Unintentional Injuries Motor Vehicle Total Unintentional Injuries Breast Cancer Lung Cancer Malignant Neoplasms Cerebrovascular Disease Heart Disease Cardiovascular Disease Montrose County Mesa County Garfield County Delta County
50
100
150 Rate per 100,000
200
250
300
82
Table 15 provides information on the 2006 agespecific death rates for the four counties. The age distributionofdeathsissimilar. Table15.AgespecificdeathRatesbyCounty2006
Delta County AgeSpecific Garfield County AgeSpecific Death Rate * 1.4 3.0 41.6
Age 1-14 15-24 25-64 65+ All Ages
Deaths * 4 58 259
Population 5,321 4,159 15,157 5,708
Death Rate * 1 3.8 45.4
Deaths * 10 88 187
Population 11,290 7,031 29,181 4,497
323
30,672
10.5
289
52,878
5.5
Mesa County AgeSpecific
Montrose County AgeSpecific Death Rate 0 1.1 3.1 50
Age 1-14 15-24 25-64 65+ All Ages
Deaths 7 19 263 970
Population 24,507 21,127 68,064 19,954
Death Rate 0.3 0.9 3.9 48.6
Deaths 0 6 60 296
Population 7,636 5,342 19,480 5,920
1,268
135,473
9.4
36
38,908
9.4
*indicates one or two events
Ageadjustedratesareadjustedtothe2000U.S.populationusingthedirectmethodappliedto10year agegroups.Source:HealthStatisticsSection,ColoradoDepartmentofPublicHealthandEnvironment.
83
The figures on this page portray death rates by age group and gender for Garfield County residents overtheperiod1990through2006. Figure27.DeathRatesforGarfieldCountyMalesbyAgeGroups
8000
.
7000 6000
Rate per 100,000
5000 Age 18-25 4000 Age 26-45 Age 46-65 Age 66-99 3000
2000
1000
n<3 for Age 1825 for years 1994, 2000 and 2005 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Years
Figure28.DeathRatesforGarfieldCountyFemalesbyAgeGroups
7000
6000
5000
Rate per 100,000
4000
Age 18-25 Age 26-45 Age 46-65
3000
Age 66-99
2000
1000
0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
n<3 for Age 1825 all years and Age 26-45 for years 1990 and 1993
Years
AppendixLprovidesstatisticalinformationoncauseofdeathbyrace/ethnicityfor2006.Onlydatafor all races, nonHispanic Whites and Hispanic Whites are shown in this table. Individuals with other racialorethnicbackgroundsarearelativelysmallpercentageofthepopulationofthefourcomparison counties,andthus,makeacorrespondinglysmallcontributiontotheoveralldeathstatistics.Generally speaking,thedistributionofcauseofdeathforHispanicWhitesiseitherlowerthanorsimilartothatof
nonHispanic Whites in all counties. One notable exception is fact that unintentional injuries accountedforahigherpercentageoftotaldeathsforHispanicWhitesthanfornonHispanicWhitesin GarfieldCounty. The figure below shows the leading causes of diseaserelated death by county for 2006.
Figure29.LeadingCausesofDeathbyCounty,2006
250 223.6 200 198.1 176.7 163.4 150 162.7 148.6 138.4 135.9
Age-Adjusted Rate
100 67 65.9 43.7 35.4 63.1 60.3 47.2 47.3 48.3 47.6 46.2 31
Delta Garfield Mesa Montrose
50
0
se rt D is ea tN eo pl a sm s is e
42.1 20.7 18.1 12
42.3 37.3 28 21.2
15.3 12.4
10.8
20.6 19.6 10.9
16.6 15.8
12.5
14.4 10.6
12.3
as es Pn U eu ni m nt on en ia C tio er na eb lI ro nj va ur sc ie s ul ar C hr D A on is lz ea ic N he ep se Li im s hr ve er iti rD 's s, is D N is ea ep ea se hr se ot an ic d Sy C irr nd ho ro si m s e, N ep hr D os ia is be te s M el lit us O th er Pa Su D is rk ic ea in id so se e O s n' th of s er D R is D es is ea pi ea se ra se to s ry of Sy C irc st em ul at or y Sy st em an d
ea
na n
al ig
es pi ra R
flu en za
to
ry
Causes of Death
All Causes - Age-adjusted rate Delta 768.8 Garfield 724.8 Mesa 842.4 Montrose 730.2
w er C hr on i c Lo
In
85
BirthDefects Birth defects data by county for the years 20002005 was obtained from Colorado Responds to Children with Special Needs (CRCSN), Colorados birth defects monitoring and prevention program. ThisorganizationislocatedwithintheColoradoDepartmentofPublicHealthandEnvironment.Tobe includedinCRCSN,achildmustbeaColoradoresidentdiagnosedprenatallytothreeyearsofagewith an eligible condition. Eligible conditions include medical conditions, medical risk factors for developmentaldelay,oroneoftwoenvironmental(i.e.,maternal)riskfactorsfordelay.Childrenwho meet these criteria are identified, using information from hospitals, vital records (birth, death, and fetal death certificates), from the Newborn Screening Program, the Newborn Hearing Screening Program,laboratories,physicians,andspecialtyclinics(e.g.,geneticordevelopmentalclinics) Table 16 shows the number of live births identified in Garfield County by year of birth for the time period20002005. Table16.NumberofResidentLive BirthsinGarfieldCounty:20002005 YEAR 2000 2001 2002 2003 2004 2005 Total LIVEBIRTHS 786 782 855 871 803 816 4,913
Table17showsthatGarfieldCountysoverallratesformajorcongenitalanomalies(birthdefects)are no higher than those observed for a comparison area of all counties in Colorado excluding Garfield County.
86
TABLE17.RATESOFBIRTHDEFECTS20002005 DIAGNOSISCATEGORY
MAJORCONGENITALANOMALIES CENTRALNERVOUSSYSTEM ANENCEPHALUSANDSIMLARANOMALIES ENCEPALOCELE MICROCEPHALUS CONGENITALHYDROCEPHALUSWITHOUT SPINABIFIDA EYE MAJOREYE MICROPHTHALMOS CONGENITALCATARACTANDLENS ANOMALIES CARDIOVASCULAR MAJORCARDIOVASCULAR TRANSPOSITIONOFGREATVESSELS COMMONVENTRICLE VENTRICULARSEPTALDEFECT OSTIUMSECUNDUMTYPEATRIALSEPTAL DEFECT ENDOCARDIALCUSHIONDEFECTS CONGENTIALSTENOSISOFAROTICVALVE HYPOPLASTICLEFTHEARTSYNDROME PULMONARYVALVESTENOSISAND ATRESIA
GarfieldCounty
Count Rate 261 21 1 1 5 8 7 5 1 1 103 95 3 1 28 45 1 3 1 4
Coloradoexcluding GarfieldCounty
Lower Upper Lower Upper Count Rate Limit Limit Limit Limit 531.2 470.2 597.7 21,445 532.6 525.7 539.6 42.7 2.0 2.0 10.2 16.3 14.2 10.2 2.0 2.0 26.5 0.1 0.1 3.3 7.0 5.7 3.3 0.1 0.1 171.4 156.7 1.3 0.1 37.9 66.9 0.1 1.3 0.1 2.2 65.3 11.3 11.3 23.7 32.1 29.3 23.7 11.3 11.3 253.7 235.9 17.8 11.3 82.3 122.4 11.3 17.8 11.3 20.8 1,067 18 27 226 281 599 463 44 130 7,231 6,803 147 34 1,670 2,846 154 152 110 430 26.5 0.4 0.7 5.6 7.0 14.9 11.5 1.1 3.2 179.6 169.0 3.7 0.8 41.5 70.7 3.8 3.8 2.7 10.7 24.9 0.3 0.4 4.9 6.2 13.7 10.5 0.8 2.7 175.5 165.0 3.1 0.6 39.5 68.1 3.2 3.2 2.2 9.7 28.1 0.7 1.0 6.4 7.8 16.1 12.6 1.5 3.8 183.7 173.0 4.3 1.2 43.5 73.3 4.5 4.4 3.3 11.7
209.6 193.4 6.1 2.0 57.0 91.6 2.0 6.1 2.0 8.1
87
TABLE17(CONTINUED).RATESOFBIRTHDEFECTS20002005 DIAGNOSISCATEGORY
COARCTATIONOFAORTA ANOMALIESOFPULMONARYARTERY RESPIRATORY AGENESIS,HYPOPLASIA,DYSPLASIAOF LUNG OROFACIAL CLEFTPALATEWITHOUTCLEFTLIP CLEFTLIPWITH/WITHOUTCLEFTPALATE GASTROINTESTINAL MAJORGASTROINTESTINAL TRACHEOESOPHAGEALFISTULA, ESOPHAGEALATRESIAANDSTENOSIS CONGENITALHYPERTROPHICPYLORIC STENOSIS ATRESIA,STENOSISOFTHELARGE INTESTINE,RECTUM,ANALCANAL HIRSCHSPRUNGDISEASEANDOTHER COLONDISORDERS GENITOURINARY MAJORGENITOURINARY RENALAGENESISANDDYSGENESIS ARM/HANDLIMBREDUCTION LEG/FOOTLIMBREDUCTION
GarfieldCounty
Count 10 16 17 5 12 7 6 18 15 1 Rate 20.4 32.6 34.6 10.2 24.4 14.2 12.2 36.6 30.5 2.0 Lower Limit 9.8 18.6 20.2 3.3 12.6 5.7 4.5 21.7 17.1 0.1 Upper Limit 37.4 52.8 55.3 23.7 42.6 29.3 26.6 57.8 50.3 11.3
Count 377 1,029 1,231 180 769 360 450 2,055 1,810 182 Rate 9.4 25.6 30.6 4.5 19.1 8.9 11.2 51.0 45.0 4.5 Lower Upper Limit Limit 8.4 24.0 28.9 3.8 17.8 8.0 10.2 48.9 42.9 3.9 10.4 27.2 32.3 5.2 20.5 9.9 12.3 53.3 47.1 5.2
8.1
2.2
20.8
665
16.5
15.3
17.8
6.1
1.3
17.8
211
5.2
4.6
6.0
2 97 97 5 1 1
4.1 197.4 197.4 10.2 2.0 2.0
0.5 160.4 160.4 3.3 0.1 0.1
14.7 240.3 240.3 23.7 11.3 11.3
89 7,389 7382 162 123 59
2.2 183.5 183.3 4.0 3.1 1.5
1.8 179.4 179.2 3.4 2.5 1.1
2.7 187.7 187.5 4.7 3.6 1.9
88
Table17(continued).RatesofBirthDefects20002005 DIAGNOSISCATEGORY
REDUCTIONDEFORMITY ANOMALIESOFDIAPHRAGM(INC DIAPHRAGMATICHERNIA) ANOMLIESOFABDOMINALWALL 15 CHROMOSOMAL 3 TRISOMY21(DOWNSYNDROME) TRISOMY13(INCLUDINGPATAU SYNDROME) TRISOMY18(INCLUDINGEDWARDS SYNDROME) 1 2 6.1 2.0 4.1 1.3 0.1 0.5 17.8 11.3 14.7 529 35 40 13.1 0.9 1.0 12.0 0.6 0.7 14.3 1.2 1.4 30.5 17.1 50.3 1,131 28.1 26.5 29.8
GarfieldCounty
Count 2 2 2 Rate 4.1 4.1 4.1 Lower Limit 0.5 0.5 0.5 Upper Limit 14.7 14.7 14.7
Count 172 136 238 Rate 4.3 3.4 5.9 Lower Upper Limit Limit 3.7 2.8 5.2 5.0 4.0 6.7
Ratesareper10,000livebirths/Lowerandupperlimitsare95%confidencelimitsbasedonabinomial distribution.Denominator:studyarean=4,913livebirths,ComparisonArean=402,643 Table18showsthemajorcongenitalanomalyratesbyyearofbirthinGarfieldCounty.Thesedatado notindicatethattherateofmajorcongenitalanomalieswaseitherincreasingordecreasingoverthis timeperiod(CochranArmitageTrendTestp=0.68). Table18.MajorCongenitalAnomalyRatesGarfieldCounty:20002005 Yearofbirth NumberofMajor Numberof MajorCongenital Anomalies Births AnomalyRate 786 2000 36 458.0 2001 2002 2003 2004 2005 Total 46 53 37 38 51 261 782 855 871 803 816 4,913 588.2 619.9 424.8 473.2 625.0 531.2
Ratesareper10,000livebirths
ChildandAdolescentHealth The following table and maps provide information on how Garfield County ranks relative to the comparisoncountiesandtheStateofColoradoonselectedmeasuresofchildandadolescenthealth. We chose to look at some of these measures because of concerns expressed during focus group discussions with Garfield County human services providers, school district personnel, healthcare providers,andcitizensabouttheseimportantdeterminantsofhealth.(PleaseseetheQualitativeData section,above.) ThefollowingmeasuresareportrayedinTable19: Teen Fertility. This measure is the rate of births per 1,000 teens ages 1517 for 20032005. The numeratoristhetotalnumberofbirthstoteensin2003,2004,and2005.Thedenominatoristhesum oftheestimatedpopulationoffemalesages15through17forthesameyears.Theresultismultiplied by 1,000 to obtain the average rate for the 3year period. Data are calculated by the county of residenceoftheteen.DSindicatesdatathataresuppressedforconfidentialitywhenthenumeratoror denominatorhasavalueof1or2.Thenational,HealthyPeople2010goalreferstopregnancies,not births. The Colorado 2010 goal is 21 births per 1,000 teens. Data source: Numerator Health StatisticsandVitalRecords,ColoradoDepartmentofPublicHealthandEnvironment.Denominator DemographySection,ColoradoDepartmentofLocalAffairs. Teen Injury Hospitalizations. This measure is the rate of hospitalization for injuries per 100,000 teensages1519fortheperiod20002005.Thenumeratoristhesumofallhospitalizationsforinjuries ofteensages1519in2000,2001,2002,2003,2004,and2005.Thedenominatoristhesumofthe estimated population of teens ages 15 through 19 for the same years. The result is multiplied by 100,000toyieldtheaveragerateforthe6yearperiod.Ratescanbeaffectedbysmallnumbers,and cautionshouldbeusedwiththesedata.WhilethereisnoHealthyPersons2010goalforthismeasure, the Colorado 2010 goal is a rate of 450.0. Data Source: Injury Epidemiology, Prevention Services Division,ColoradoDepartmentofPublicHealthandEnvironment. Teen Suicide. This measure is the rate of suicide deaths per 100,000 teens ages 1519 for 2001 2005.Thenumeratoristhetotalnumberofteensuicidedeathsin2001,2002,2003,2004,and2005. Thedenominatoristhesumoftheestimatedpopulationofteensages15through19forthesesame years. The result is multiplied by 100,000 to obtain the average rate for the fiveyear period. DS indicatesthatdataaresuppressedforconfidentialitywhenthenumeratorordenominatorhasavalue of 1 or 2. Data Source: Numerator Health Statistics and Vital Records, Colorado Department of PublicHealthandEnvironment.DenominatorDemographySection,ColoradoDepartmentofLocal Affairs. TeenMotorVehicleDeaths.Thismeasureistherateofdeathsper100,000teensages1519that werecausedbymotorvehiclecrashesfor20032005.Thenumeratoristhetotalnumberofdeathsto teensin2003,2004,and2005.Theresultismultipliedby100,000toobtaintheaveragerateforthe3 year period. DS indicates that data are suppressed for confidentiality when the numerator or denominator has a value of 1 or 2. Data Source: Numerator Health Statistics and Vital Records,
90
Colorado Department of Public Health and Environment. Denominator Demography Section, ColoradoDepartmentofLocalAffairs. Child Abuse. This measure is the rate of maltreatment (including physical abuse, sexual abuse, emotionalabuseand/orneglect)ofchildrenyoungerthan18incalendaryear2005.Thevalueshown reflects the number of unique substantiated cases per 1,000 children less than 18 years of age in a county. Data source: Numerator Colorado Central Registry, Colorado Department of Human Services.DenominatorDemographySection,ColoradoDepartmentofLocalAffairs. ChildDeaths.Thismeasureistherateofchilddeathsper100,000childrenages114for20012005. Thenumeratoristhesumofallchilddeathsamongchildrenages1through14in2001,2002,2003, 2004 and 2005. The result is multiplied by 100,000 to obtain the average fiveyear rate for child deaths.WhilethereisnoHealthyPeople2010goalforthismeasure,theColoradogoalis19.0.Data Source:HealthStatisticsandVitalRecords,ColoradoDepartmentofPublicHealthandEnvironment. DenominatorDemographySection,ColoradoDepartmentofLocalAffairs. Child Motor Vehicle Deaths. This measure is the rate of deaths per 100,000 children ages 014 caused by motor vehicle crashes for 20032005. The numerator is the sum of all deaths among childrenages014in2003,2004,and2005.Thedenominatoristhesumoftheestimatedpopulation of children ages 0 through 14 for the same years. The result is multiplied by 100,000 to obtain the average3yearrateforchildmotorvehicledeaths.DSindicatesdataaresuppressedforconfidentiality whenthenumeratorordenominatorhasavalueof1or2.TheHealthyPeople2010goalisforages, notspecificallyforages014.DataSource:HealthStatisticsandVitalRecords,ColoradoDepartment of Public Health and Environment. Denominator Demography Section, Colorado Department of LocalAffairs. Child Injury Hospitalizations. This measure is the rate of hospitalization for injuries per 100,000 children ages 014 for 20002005. The numerator is the sum of all hospitalizations for injuries of children ages 014 in 2000, 2001, 2003, 2003, 2004 and 2005. The denominator is the sum of the estimated population ages 0 through 14 for the same years. The result is multiplied by 100,000 to obtain the average 6year rate for child injury hospitalizations. Rates can be affected by small numbers,andcautionshouldbeusedwiththesedata.WhilethereisnoHealthyPeople2010goalfor thismeasure,theColorado1020goalisarateof150.0.DataSource:InjuryEpidemiology,Prevention ServicesDivision,ColoradoDepartmentofPublicHealthandEnvironment. ChildrenwithSocialEmotionalDifficulties.Thismeasureistheestimatedpercentofchildrenages1 14whohavedifficultieswithemotions,concentration,behavior,orbeingabletogetalongwithother people.ThedatacombinetwoyearsofresultsfromtheColoradoChildHealthSurveyandarespecific to children ages 1 through 14 in Colorado in 20042005. Numerators and denominators are not provided.DataforGarfieldandMesaCountiesarePlanningandManagementRegiondata.Nodata are available for Delta and Montrose Counties. Data Source: Child Health Survey, Health Statistics Section,ColoradoDepartmentofPublicHealthandEnvironment.
91
Dental Care Received. This measure is the estimated percent of children ages 114 who received neededdentalcareinthe12monthsbeforethesurvey.Thedatacombinetwoyearsofresultsfrom the Colorado Child Health Survey and are specific to children age 1 through 14 living in Colorado in 20042005.Numeratorsanddenominatorsarenotprovided.PlanningandManagementRegiondata are used for Garfield and Mesa Counties. No data are available for Delta and Montrose Counties. There are no Healthy People 2010 or Colorado goals for this measure. Data Source: Child Health Survey,HealthStatisticsSection,ColoradoDepartmentofPublicHealthandEnvironment. HealthInsuranceCoverage.Thismeasureistheestimatedpercentofchildrenages114whohad anykindofhealthcarecoverage,includinghealthinsurance,prepaidplans,orgovernmentplans.The datacombinetwoyearsofresultsfromtheColoradoChildHealthSurveyandarespecifictochildren ages1through14livinginColoradoin20042005.PlanningandManagementRegiondataareshown forGarfieldandMesaCounties.NodataareavailableforDeltaandMontroseCounties.Numerators anddenominatorsarenotprovided.DataSource:ChildHealthSurvey,ColoradoDepartmentofPublic HealthandEnvironment. MedicalCareReceived.Thismeasureistheestimatedpercentofchildrenages114whoreceived allthemedicalcaretheyneededinthe12monthsbeforethesurvey.Thedatacombinetwoyearsof resultsfromtheColoradoChildHealthSurveyandarespecifictochildrenages1through14livingin Colorado in 20042005. Planning and Management Region data are shown for Garfield and Mesa Counties.NodataareavailableforDeltaandMontroseCounties.Numeratorsanddenominatorsare not provided. There are no Healthy People 2010 or Colorado 2010 goals for this measure. Data Source:ChildHealthSurvey,ColoradoDepartmentofPublicHealthandEnvironment. PersonalDoctororNurse.Thismeasureistheestimatedpercentofchildrenages114whohavea personal doctor or nurse. The data combine two years of results from the Colorado Child Health Survey and are specific to children ages 1 through 14 living in Colorado in 20042005. Planning and ManagementRegiondataareshownforGarfieldandMesaCounties.NodataareavailableforDelta and Montrose Counties. Numerators and denominators are not provided. There are no Healthy People 2010 or Colorado 2010 goals for this measure. Data Source: Child Health Survey, Colorado DepartmentofPublicHealthandEnvironment. Overweight.Thismeasureisthepercentageofchildren,ages214,whosBMI(bodymassindex)is atorabove85percentofnormalweightforheight.Thedatacombinetwoyearsofresultsfromthe ColoradoChildHealthSurveyandarespecifictochildrenages1through14livinginColoradoin2004 2005.PlanningandManagementRegiondataareshownforGarfieldandMesaCounties.Nodataare available for Delta and Montrose Counties. Numerators and denominators are not provided. Data Source: Child Health Survey, Health Statistics Section, Colorado Department of Public Health and Environment.
92
Table19.ChildandAdolescentHealthbyCounty Measure TeenFertility(1519) TeenInjuryHospitalizations(1519) TeenSuicide TeenMotorVehicleDeaths(1519) ChildAbuse ChildDeaths(114) ChildMotorVehicleDeaths(014) InjuryHospitalizations(014) ChildrenwithSocialEmotionalDifficulties DentalCareReceived HealthInsuranceCoverage MedicalCareReceived PersonalDoctororNurse Overweight Numerator 52 95 4 DS 46 9 N/A 93 N/A N/A N/A N/A N/A N/A
DeltaCounty Denominator 1,921 12,476 10,752 6,510 6,933 26,204 N/A 33,264 N/A N/A N/A N/A N/A N/A CountyMeasure 27.1 761.5 37.2 DS 6.6 34.3 N/A 279.6 N/A N/A N/A N/A N/A N/A Numerator 77 134 0 3 78 18 N/A 121 N/A N/A N/A N/A N/A N/A
GarfieldCounty Denominator 3,258 21,175 18,185 11,059 13,833 50,698 N/A 64,738 N/A N/A N/A N/A N/A N/A CountyMeasure 23.6 632.8 12.1 27.1 5.6 35.5 91.9 186.9 34.5 79.5 79.5 98.4 76.8 16.3
MesaCounty
MontroseCounty
Colorado
People 2010
93
Goal Measure TeenFertility(1519) TeenInjuryHospitalizations(1519) TeenSuicide TeenMotorVehicleDeaths(1519) ChildAbuse ChildDeaths(114) ChildMotorVehicleDeaths(014) InjuryHospitalizations(014) Children with Difficulties Social Emotional N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 34.5 79.5 79.5 98.4 76.8 16.3 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 28.9 87.9 87.9 98.5 77.6 14.5 N/A 100 100 N/A N/A 5
Numerator Denominator County Measure Numerator Denominator County Measure
224 464 7 7 412 26 N/A 364
8,284 58,924 50,268 30,520 31,026 115,227 N/A 146,958
27 787.5 13.9 22.9 13.3 22.6 91.9 247.7
59 81 DS 4 76 7 N/A 73
2,430 16,021 13,661 8,168 9,630 36,012 N/A 45,499
24.4 505.6 DS 49 7.9 19.4 N/A 160.4
24.4 573.3 12.1 26.2 9.2 20 91.6 198.3
43 N/A 5 9.2 10.3 N/A N/A N/A
DentalCareReceived HealthInsuranceCoverage MedicalCareReceived PersonalDoctororNurse Overweight
94
Figure30.TeenFertility(Ages1517)
Colorado FY2010 Goal: Reduce fertility rate among females age 15-17 to 21.0 births per 1,000 females.
County rates are for 2003-2005 three-year averages. The state rate is 40.9.
95
Figure31.TeenInjuryHospitalizations Colorado FY2010 Goal: Reduce the rate of injury hospitalization among children age 15-19 to 450.0 per 100,000.
County rates are 2000-2005 six-year averages, except for Broomfield which is averaged over the three-year period 2004-2005. The Colorado rate is 573.3.
96
Figure32.TeenMotorVehicleDeaths(Ages1519) Healthy People 2010 Goal: Reduce the rate of deaths among adolescents age 15-19 caused by motor vehicle crashes to no more than 9.2 deaths per 100,000 teens.
County rates are 2004-2005 three-year averages. The state rate is 26.2.
97
Figure33.TeenSuicide(Ages1519) Healthy People 2010 Goal: Reduce suicides to no more than 5.0 per 100,000 youth aged 15-19.
County rates are 2001-2005 five-year averages, except for Broomfield which is 2002-2005 only. The state rate is 12.1.
98
Figure34.ChildAbuse(Numberper1,000)
Healthy People 2010 Goal: Reduce the incidence of maltreatment of children younger than 18 (including physical abuse, sexual abuse, emotional abuse, and/or neglect) to 10.3 or below. County rates are for calendar year 2005. The state rate is 9.2 per 1,000.
99
Figure35.ChildDeaths(Ages114)
Number per 100,000 Colorado 2010 Goal: Reduce the death rate of children to no more than 19.0 deaths per 100,000 children aged 1-14. County rates are 2001-2005 five-year averages, except for Broomfield which is 2002-2005 only. The state rate is 20.0.
InjuryRelatedHospitalizationsandDeaths The Colorado Department of Public Health and Environment tracks injuryrelated hospitalizations. Although there are many ways by which these data may be sorted and reported(e.g.,byage,gender,mechanismofinjury,etc.),forthepurposesofthisreport,we havechosentoreviewonlyacoupleofcategoriesbycounty:injuryhospitalizationsbyintent, and injury hospitalizations by injury type. Tables 20a and 20b and the accompanying graphs show the relationship of Garfield County to the state and comparison counties in these categories. Table20A.InjuryHospitalizationsbyCounty,20012003 TotalIntentionCategories AgeAdjustedRatesper100,000 Delta Garfield Mesa Montrose Colorado Total Unintentional Intentional Hospitalizations Injury Injury 812.6* 654.9 746.1* 607.4** 694.0 721.2* 574.4 579.2 514.8** 591.6 61.9** 57.3** 153.6* 71.6 81.5 Figure36. *significantly higher than Coloradorate **significantly lower than Coloradorate
Injury Hospitalizations with Intention Categories

1000 800 600 400 200 0 Delta Garfield Mesa Unintentional Injury Montrose Colorado
Total Hospitalizations
Intentional Injury
101
Table20B.InjuryHospitalizationsbyCounty,20012003 TotalInjuryType AgeAdjustedRatesper100,000 Motor Vehicle Traffic Fall Poisoning Delta 114.4 309.7 20.8 Garfield 83.2 318.4 28.2 Mesa 93.5 292.8** 46.5* Montrose 98.9 252** 15.8** Colorado 96.5 340.6 26.7 *significantlyhigherthanColoradorate **significantlylowerthanColoradorate
Suicide/ Attempt 50.8 41.9** 126.2* 64.3 54.7
Assault/Legal Intervention 12.4** 15.4** 28.0 7.3** 27.1
Figure37.
Injury Hospitalizations by Injury Type
350 300 250 200 150 100 50 0 Delta Garfield Mesa Montrose Colorado Motor Vehicle Traffic Fall Poisoning Suicide/ Attempt Assault/Legal Intervention
Similarly,theCDPHErecordsandtracksinjuryrelateddeaths.Forthepurposesofthisreport, county and state comparisons are shown only for injury deaths by intention and injury deaths by injury type. This information is presented in Tables 21a and 21b and the accompanyinggraphs.
102
Table 21A. Injury Deaths by County, 2001 2003 TotalIntentionCategories AgeAdjustedRatesper100,000 Total Injury Unintentional Intentional Deaths Injury Injury Delta 76.9 47.4 28.1 Garfield 82* 56.1* 22.0 Mesa 65.6 38.2 21.8 Montrose 82.3 57.2* 24.0 Colorado 63.9 41.6 20.1 Figure38.
*significantlyhigherthanColoradorate **significantlylowerthanColoradorate
Injury Deaths with Intention Categories

100 80 60 40 20 0 Delta Garfield Total Injury Deaths Mesa Unintentional Injury Montrose Colorado
Intentional Injury
103
Table 21B. Injury Deaths by County, 2001 2003 TotalInjuryType AgeAdjustedRatesper100,000 Motor Vehicle Traffic Fall Poisoning Delta 21.6 3.2** 8.2 Garfield 22.3 6.8 5.1 Mesa 17.5 7.1 1.8** Montrose 26.0 3.4** 3.1 Colorado 15.9 8.4 6.6 *significantlyhigherthanColoradorate **significantlylowerthanColoradorate Figure39.
Injury Deaths by Injury Type
30 25 20 15 10 5 0 Delta Garfield Mesa Montrose Colorado Motor Vehicle Traffic Fall Poisoning Suicide Homicide Firearm Related
Firearm Related 19.4 15.7 13.2 16.8 11.3
Suicide 24.6 16.4 19.7 21.0 16.0
Homicide 3.6 5.6 2.2** 3.0 4.1
104
BehavorialRiskFactorStudySurvey(BRFSS)Data TheannualCentersforDiseaseControlandPreventionBehavioralRiskFactorStudy(arandomsample,telephonebasedsurveyofhealth risk factors and behaviors) provides information on the health of a representative sample of adults. Table 22 provides the results of selected20042005surveyquestionsbycounty.
Table22.20042005 BRFSS DeltaCounty GarfieldCounty MesaCounty MontroseCounty Topic % 95%CI* % 95%CI* % 95%CI* % 95%CI* Diagnosedwithdiabetes 2.8 0.05.8 2.6 0.05.1 5.4 3.07.8 6.4 1.611.2 Currentsmoker 23.9 14.133.7 18.7 9.627.7 20.7 15.226.3 21.8 12.531.1 Currently have health insurance 67.6 56.878.4 79.2 20.887.6 80.1 74.385.9 71.3 60.382.3 Everhadasthma 17.2 7.926.5 16 8.123.9 10.6 7.014.2 15 5.224.7 Overweight 60.7 49.272.2 41.8 31.152.5 57.3 50.863.8 56.3 44.867.8 BMI**25.029.9 Obese 14.1 5.422.7 12.6 4.920.2 20.9 15.526.3 21.3 12.230.3 BMI**>30 *ConfidenceInterval **BodyMassindex(BMI)isdefinedasweightinkilogramsdividedbyheightinmeterssquared Percentsareweightedtothetotalpopulation.
Colorado % 95%CI* 4.6 3.45.0 19.9 18.920.9 84.4 17.2 60.7 14.1 83.485.3 7.926.5 49.272.2 5.422.7
Accordingtotheresultsofthissurvey,fewerGarfieldCountyresidentshavediabetes,smokeorareoverweightorobesethaninthestate, asawhole,orinthecomparisoncounties.ThepercentageofadultsinGarfieldCountywhohaveasthmaissimilartothestate,asawhole, andtotwoofthethreecomparisoncounties.
105
CancerStatistics Cancer statistics for Colorado and the four comparison counties were obtained from the Colorado Department of Public Health and Environment, Colorado Central Cancer Registry, Prevention Services Division (www.cdphe.state.co.us). Mr. Jack Finch, (Statistical Analyst III) provided the most recently available data (through 2005) for the most frequently observed cancers in Colorado, as well as for several other cancers that are of interest because of their potential link to exposure to benzene or to radionuclides. For cancers that have sufficiently high rates of occurrence to provide meaningful trend data, trends in incidence and mortality ratesareprovidedforthefourcomparisoncountiesfortheyears1992through2005.(Please see Appendix M for data tables.) Table 23 (below) shows comparisons of selected cancer incidenceratestothestateincidenceratesforthesametimeperiods.Blankcellsindicatethat therewasnosignificantdifferencefromthestaterate. Table 24 provides the same information for mortality rates. It is important to note that mortality rates more so than incidence rates may reflect differences in socioeconomic status and/or healthcare access (i.e., access to stateoftheart treatment), rather than differencesinriskfactorsfordevelopingcancer.Thelattermayincludeanindividualsgenetic background, lifestyle choices, and/or exposures to a wide range of environmental toxins (chemical, biological, or radiological). One of the difficulties of observing relatively recent trendsincancerincidencestatisticsandtryingtodrawconclusionsaboutpossiblechangesin risk factors, is that, with respect to carcinogenic exposures, these trends reflect events that happened 10 20 years ago or are cumulative over a lifetime. Generally speaking, the appearance of clinical cancer has a lag time of up to two decades following initiation of carcinogenesis. (Childhood cancers and some rare cancers are exceptions.) Thus, with the exceptionofchangesincancerscreeningpracticesthatoftenartifactuallyinflatecancerrates, shortterm trends may not reflect changes in the potential for exposure to carcinogenic materials. Incidence and mortality trends among the comparison counties have also been portrayed graphically(seebelow).Incidenceratesforallcancershavechangedlittleovertime,although cancerratesinmalesdroppedslightlyforDelta,Garfield,andMontrosecountiesforthemost recenttimeperiod.Thereisrelativelylittledifferenceamongthecountiesforeithermaleor femalecancerincidenceratesacrossthedesignatedtimeperiods.Inadditiontothecounty specific cancer statistics, the information on Colorado and National incidence rates and risk factors was obtained from the Cancer in Colorado 19982003 report (www.cdphe.state.co.us/pp/cccr/19982003/index.html) and Cancer Facts and Figures 2008 (AmericanCancerSociety,www.cancer.org). Selectedcancers. ProstatecancerAlthoughGarfieldCountysaverageannualincidenceratesfortheperiodof 19921998 were higher than those of Delta, Mesa and Montrose Counties, there has been a steadydownwardtrendinprostatecancerratesinGarfieldCounty.Forthemostrecenttime period, prostate cancer rates were lower than those for the other three counties. It is importanttonotethatincidenceratesareareflectionofdetectionrates.Changesinrates maybeanartifactofscreeningpractices(which,inturn,reflectstandardsofpractice,health insurancecoverage,andpublicacceptance).
106
FemaleBreastcancerAsimilardownwardtrendinfemalebreastcancerincidenceisseenfor GarfieldCounty.Incidenceratesamongthecomparisoncountiesarenotsignificantlydifferent fromeachotherorfromthestaterate. ColorectalcancerColorectalcancerratesinmalesforGarfieldCountyhavedecreasedsince 1992,andarenowsimilartothoseinMontroseCountyandlowerthanthoseinDeltaandMesa Counties and in the state, as a whole. Female colorectal cancer incidence has been highly variable,andiscurrentlysimilartotheratesinDeltaandMontroseCountiesandinthestate,as awhole(whichareallhigherthanthemostrecentincidencerateforMesaCounty). Lung cancer There has been a significant drop in the male lung cancer incidence rates in Garfield County over the most recent tracking period (20032005). Garfield County has the lowestincidencerateamongthecomparisoncountiesandthestateofColoradoforthisperiod. FemalelungcancerratesinGarfieldCountyhaveessentiallyremainedstablesince1992.The mostrecentratesaresimilartothestaterates,arehigherthanthoseforDeltaCounty,andare lowerthanthoseforMesaandMontroseCounties. Bladder cancer Bladder cancer incidence rates have been highly variable in Garfield County forbothmalesandfemales.Allofthecomparisoncountiessawaspikeinmalebladdercancer incidenceforthe20012002period(truealsoforfemalebladdercancer,withtheexceptionof DeltaCounty,forwhichnofemalebladdercancerswerereportedduringthisperiod).Themost recent tracking period showed decreases in the incidence rates for all counties, except MontroseCounty(continuedupwardtrendinmalebladdercancer)andDeltaCounty(increase infemalebladdercancer).Staterateshaveremainedstableforbothmaleandfemalebladder canceroverthe19922005timeperiod. MelanomaMelanomaratesforGarfieldCountydonotdiffersignificantlyfromthoseinthe comparison counties or the state as whole. Male melanoma rates have tended to decrease since1999,whilefemalerateshavebeenmorevariable,althoughrelativelylow. Leukemias The incidence rates for all male leukemias in Garfield County were lower than those in the comparison counties and the state during 20032005 (the only period for which countyspecificdatawereavailable).FemaleleukemiaincidenceratesforGarfieldCountywere essentiallythesameasforthestate,asawhole,slightlylowerthanforMontroseCounty,and higherthanthoseforMesaandDeltaCounties. Thyroid cancer In all cases, incidence rates for male thyroid cancer were lower than for thyroid cancer in females, during 20032005, the only period for which countyspecific data were available. In Garfield County, male thyroid cancer rates were lower than in Montrose County and the state, but higher than in Mesa and Delta Counties (No thyroid cancer was recordedformalesinDeltaCounty).ThyroidcancerratesforfemalesinGarfieldCountywere onlyhigherthantheratesforfemaleresidentsofMesaCounty. CervicalcancerGarfieldCountyhadthelowestincidenceratesforcervicalcanceramongthe comparisoncountiesandthestate,asawhole.
107
TABLE23.COMPARISONOFCOUNTYRATESWITHSTATERATESINCIDENCE Male Delta Garfield Prostate Lung Colo rectal Melan oma Bladder Leuke mias Thyroid All Cancers Female Breast Lung Colo rectal Melan oma Thyroid Leuke mias Bladder
1992 1998 1999 2000 2001 2002 2003 2005 1992 1998 1999 2000 2001 2002 2003 2005
Mesa
1992 1998 1999 2000 2001 2002 2003 2005
Montrose
1992 1998 1999 2000 2001 2002 2003 2005
*** ***
*** ***
***
Cervix *** *** All Cancers Rate is significantly higher than Colorado rate. Rate is significantly lower than Colorado rate. ***Indicates fewer than 3 events in this category.
108
TABLE24.COMPARISONOFCOUNTYRATESWITHSTATERATESMORTALITY Male Delta Garfield Prostate Lung Colo rectal Melan oma Bladder Leuke mias Thyroid All Cancers Female Breast Lung Colo rectal Melan oma Thyroid Leuke mias Bladder Cervix All Cancers
1992 1998 1999 2000 2001 2002 2003 2005 1992 1998 1999 2000 2001 2002 2003 2005
Mesa
1992 1998 1999 2000 2001 2002 2003 2005
Montrose
1992 1998 1999 2000 2001 2002 2003 2005
*** *** ***
***
***
*** *** ***
*** *** *** *** *** ***
*** *** *** *** *** *** *** *** ***
***
*** ***
*** *** *** *** ***
*** *** *** *** ***
*** *** *** ***
***
***
*** ***
*** *** ***
*** *** ***
*** *** *** ***
*** ***
RateissignificantlyhigherthanColoradorate. RateissignificantlylowerthanColoradorate. ***Indicatesfewerthan3eventsinthiscategory
109
All Cancers Incidence Trend Average Annual Rate - Males

Age-adjusted Rate per 100,000 700 600 500 400 300 200 100 0 1992-1998 1999-2000 Years 2001-2002 2003-2005
Figures 40ad. In Colorado, the cummulative lifetime risk for developing cancer of any type is 1 in 2 for males and 2 in 5 for females.
Delta Garfield Mesa Montrose Colorado
All Cancers Mortality Trend Average Annual Rate - Males

Age-adjusted Rate per 100,000 300 250 200 150 100 50 0 1992-1998 1999-2000 Years 2001-2002 2003-2005 Delta Garfield Mesa Montrose Colorado
110
All Cancers Incidence Trend Average Annual Rate - Females

Age-adjusted Rate per 100,000 500 400 300 200 100 0 1992-1998 1999-2000 Years 2001-2002 2003-2005 Delta Garfield Mesa Montrose Colorado
All Cancers Mortality Trend Average Annual Rate - Females

Age-adjusted Rate per 100,000 200 150 100 50 0 1992-1998 1999-2000 Years 2001-2002 2003-2005 Delta Garfield Mesa Montrose Colorado
111
Prostate Cancer Incidence Trends Average Annual Rates

Prostate Cancer Mortality Trends Average Annual Rates

Age-adjusted Rate per 100,000 50 40 30 20 10 0 1992-1998 1999-2000 Years

Figures 41a and 41b. Prostate cancer is the most frequently diagnosed cancer in men. In Colorado,1in5malesislikelyto develop prostate cancer during their lifetime. The only well established risk factors for prostate cancer are age (>65 years), ethnicity (incidence rates are highest among African American men), and family history of the disease. Some studiessuggestthatadiethighin saturated fat may also be a risk factor.
Delta Garfield Mesa Montrose Colorado 2001-2002 2003-2005
112
Female Breast Cancer Incidence Trends Average Annual Rates

Age-adjusted Rate per 100,000 200 150 100 50 0 1992-1998 1999-2000 Years 2001-2002 2003-2005 Delta Garfield Mesa Montrose Colorado
Female Breast Cancer Mortality Trends Average Annual Rates

Figures 42a and 42b. Breast cancer is the most frequently diagnosed cancer in women. In Colorado, it is likely that 1 in 7 women develop breast cancer during their lifetime. Age is the most important risk factor for femalebreastcancer.Otherrisk factors include genetic background(inheritedmutations intheBRCA1andBRCA2genes;a personal or family history of breast cancer), and reproductive factors such as a long menstrual history, never having children or having a first child after age 30, and recent use of oral contraceptives. Other factors suchasobesity,postmenopausal hormone therapy, physical inactivity and alcohol use may playaroleinthedevelopmentof breastcancer.
113
Colorectal Cancer Incidence Trends Average Annual Rates - Males

80 70 60 50 40 30 20 10 0 1992-1998 1999-2000 Years 2001-2002 2003-2005 Age-adjusted Rate per 100,000
Colorectal Cancer Mortality Trends Average Annual Rates - Males

Age-adjusted Rate per 100,000 30 25 20 15 10 5 0 1992-1998 1999-2000 Years
Figures 43ad. Nationally, colorectal cancer is the third most common cancer in both men and women. In Colorado the lifetime risk for developing colorectal cancer is 1 in 14 for males and 1 in 18 for females. Risk factors for developing colorectal cancer include increasing age, certain genetic mutationsorapersonalorfamily history of colorectal cancer, polyps, or chronic inflammatory boweldisease.Otherriskfactors include obesity, physical inactivity, smoking, alcohol consumption, and a diet high in red or processed meat and low infruitsandvegetables.
114
Colorectal Cancer Incidence Trends Average Annual Rates - Females

Colorectal Cancer Mortality Trends Average Annual Rates - Females

115
Lung Cancer Incidence Trends Average Annual Rates - Males

Lung Cancer Mortality Trends Average Annual Rates - Males

Age-adjusted Rate per 100,000 100 80 60 40 20 0 1992-1998 1999-2000 Years
Figures 44ad. Although the incidence rates for lung cancer aredeclining,itisstillthesecond most frequently diagnosed cancerinbothmenandwomen, and accounts for the most cancerrelated deaths. In Colorado,thelifetimeriskoflung canceris1in10formalesand1 in 15 for females. Cigarette smoking is by far the most important risk factor for developing lung cancer. Other risk factors include genetic background history of tuberculosis and occupational or environmental exposure to secondhand smoke, radon, asbestos, certain metals, organic chemicals, and other air pollutants,andradiation.
116
Lung Cancer incidence Trends Average Annual Rates - Females

Lung Cancer Mortality Trends Average Annual Rates - Females

117
Bladder Cancer Incidence Trends Average Annual Rates - Males

Bladder Cancer Incidence Trends Average Annual Rates - Females

Age-adjusted Rate per 100,000 20 15 10 5 0 1992-1998 1999-2000 Years
Figures45aand45b.Nationally, bladder cancer incidence rates are nearly four times higher in men than in women. In Colorado, the lifetime risk for bladder cancer is 1 in 20 for males and 1 in 76 for females. Smoking is the most important riskfactorfordevelopingbladder cancer. Workers in the dye, rubber or leather industries and individuals who live in communities that have high levelsofarsenicintheirdrinking water are also at increased risk fordevelopingbladdercancer.
118
Melanoma Incidence Trends Average Annual Rates - Males

Figures46aand46b.Melanoma is the most common, serious form of skin cancer, and occurs primarily (although not exclusively)inwhites.Themajor risk factors for melanoma are personal or family history and thepresenceoflargenumbersor atypicalmoles.Otherriskfactors includeahistoryofexcessivesun orultravioletlightexposure;sun sensitivity (light or red hair and lightskin);diseasesthatsuppress the immune system; and occupational exposure to coal tar, pitch, creosote, arsenic compounds, or radiation.
Melanoma Incidence Trends Average Annual Rates - Females

Age-adjusted Rate per 100,000 30 25 20 15 10 5 0 1992-1998
Delta Garfield Mesa Montrose Colorado 1999-2000 Years 2001-2002 2003-2005
119
Leukemia Incidence by County for 2003-2005 Average Annual Rates

Colorado Montrose Mesa Garfield Delta 0 5 10 15 20 25 Male Female
Age-adjusted Rate per 100,000
Leukemia Mortality by County for 2003-2005 Average Annual Rates

Colorado Montrose Mesa Garfield Delta 0 2 4 6 8 10 12 Male Female
Figures 47a and 47b. Although leukemia is diagnosed 10 times more often in adults than in children,itisoftenthoughtofas primarily a childhood disease. The most common types of leukemiadiagnosedinadultsare acute myeloid leukemia and chronic lymphocytic leukemia. Leukemia occurs more frequently in males than in females. In Colorado, the lifetime risk of developing leukemiais1in45formalesand 1 in 79 for females. Individuals with Down Syndrome and certain other genetic abnormalities have higher rates of leukemia. Cigarette smoking andexposuretoradiationandto chemicals like benzene are also riskfactorsforleukemia.

120
Thyroid Cancer Incidence by County for 2003-2005 Average Annual Rates

Colorado Montrose Mesa Garfield Delta 0 2 4 6 8 10 12 14 16 18 Male Female
Cervical Cancer Incidence Rates by County for 2003-2005 Average Annual Rates
Colorado Montrose Mesa Garfield Delta 0 1 2 3 4 5 6 7 8
Figure48.Nationally,thyroidcancers occur about 3 times more often in women than in men, and they are more frequent in younger adults. In Colorado, the lifetime risk of developing thyroid cancer is 1 in 204 for males and 1 in 92 for females. Major risk factors for developing thyroidcancerareadietlowiniodine, exposure to radiation, and certain inheritedmedicalconditions. Figure49.Incidenceratesforcervical cancer have decreased significantly, and as Pap screening has become morecommon,cervicalcancerismore often detected before it becomes invasive.InColorado,1in154women develop invasive cervical cancer over their lifetime. The primary cause of cervicalcancerisinfectionwithhuman papillomavirus (HPV). The ability of the infection to progress to cancer is related to factors such as suppressed immunity, having many children, cigarette smoking, and nutrition. Longtermuseoforalcontraceptivesis also associated with an increased risk ofcervicalcancer.
121
WestNileVirus There are two reasons for looking at countybased statistics for West Nile virus both are relatedtoconcernsvoicedbyGarfieldCountyresidents.First,inpublicmeetings,focusgroups, andinterviews,citizenshaveexpressedtheconcernthatbirdsappearedtobefewerinnumber incertainareasofGarfieldCountythanremembered.BirdsarethereservoirhostforWestNile virus.Theymaybecomeillanddieasaresultoftheviralinfection. Second,manyofthesymptomsassociatedwithhumanWestNilevirusinfectionmimicthose reported by Garfield County residents and attributed to exposures related to natural gas industryactivities. WestNileviruswasfirstobservedinColoradoin2002.Itwasnotuntil2003,however,before the virus had a significant Western Slope presence. The coincidence of the presence of this virus and the beginnings of the surge in natural gas industry activities on the Western Slope makeitimportanttoconsiderthepossibilitythatWestNilevirusinfectioncouldbethecauseof someofthehumanphysicalsymptomsandanimalimpactsthathavebeenreportedbyGarfield Countyresidents.Atthispoint,itwouldbeextremelydifficulttoeitherruleoutWestNilevirus asthecauseofatleastsomeoftheseimpactsortoproveconclusivelythatitwasthevirusand notexposuretovolatileorganiccompoundsorothertoxicantspresentintheenvironmentasa resultofnaturalgasactivity. NaturalHistoryofWestNileVirus(www.cdphe.state.co.us/dc/Zoonosis/zoonosis.asp) WestNilevirusismaintainedinabirdmosquitobirdcycle.Mosquitoesareinfectedbyfeeding onaninfectedbird;virusisthentransmittedtoanewhostwhenthemosquitobitesaperson orotheranimal.Humansandhorsesarethemostfrequent,deadendhosts(i.e,theycannot infectothermosquitoes).VirusesareprevalentfromMaytoSeptember,whenmosquitoesare mostabundant. West Nile virus infection has been detected in more than 70 wild bird species. The highest deathratesareseenamongbirdsinthecorvidfamily.Horsesaresusceptibletoinfectionwith West Nile virus and can die as a result of the infection. Other clinical signs include lack of coordination or muscle control, weakness of limbs, and inability to rise. Other mammalian speciesmaybesusceptibletoinfectionwiththisvirus(e.g.,therehavebeenreportsofinfection indogs). MostpeoplewhoareinfectedwithWestNilevirusdonotbecomeillandhavenosymptoms. Forthosewhodobecomeill,theonsetofsymptomsoccurs515daysafterbeingbittenbyan infected mosquito. The clinical symptoms of West Nile virus infection range from a flulike syndrome(WestNilefever)tomeningitisandencephalitis.ThesymptomsofWestNilefever have been reported to last significantly longer than the usual flu. Individuals who have developed West Nile fever report not feeling well for many months after the onset of the illness.Insomecases,infectionwiththevirusresultsininflammationofthebrain,resultingin either meningitis or encephalitis. In these cases, symptoms begin with a high fever and headachethatmayprogresstostiffneck,disorientation, tremors, coma and death. Severe infections have resulted in longlasting neurological syndromesthatresemblepolio.
122
ThefollowinggraphsandmapsillustratethetimetrendsforWestNileVirusinColoradofrom 2002 through 2007 (Colorado Department of Public Health and Environment). Information is providedforbothhumanandanimalinfections.Itisimportanttonotethatthenumbersfor humancasesrepresentonlythosethatwereclinicallyvalidated(eitherbybloodtestsorbythe factthattheindividualwashospitalizedasaresultoftheirinfection).Becauseofthetendency ofindividualssufferingflulikesymptomsNOTtobetestedforWestNilevirus(duetocostor unavailability of the test) and to either be treated by their doctors or to treat themselves symptomatically, it is very likely that the actual numbers of individuals who developed West Nile Fever were much higher than those listed in the official records (CDPHE; www.cdphe.state.co.us)andreportedhere.
Figure50.WestNileVirusInfections2003
40 Total deaths 35 Total Cases Encephalitis Meningitis Fever
Number of validated cases
30
25
20
15
10
0 Delta Garfield Mesa Montrose
County
2003 Colorado cases: Fever = 2325; Meningitis = 288; Encephalitis = 234; Total Cases = 2947; Deaths = 63
123
300 Total deaths Total Cases 250 Encephalitis Meningitis Fever
200
150
100
50
County
25 Total deaths Total Cases 20 Encephalitis Meningitis Fever 15
10
0 Delta Garfield
County
Mesa Montrose 2005 Colorado cases: Fever = 85; Meningitis = 13; Encephalitis = 8; Total Cases = 106; Deaths = 2
124
80
Total deaths Total Cases Encephalitis Meningitis Fever
70
60 Number of validated cases
50
40
30
20
10
County
80
Total deaths Total Cases Encephalitis Meningitis Fever
70
60
50
40
30
20
10
0 Delta Garfield
County
Mesa Montrose 2007 Colorado cases: Fever = 459; Meningitis = 43; Encephalitis = 53; Total Cases = 555; Deaths = 6
125
Figure55.ColoradoWestNileVirusHumanCases2002

126
127
128
Figure61.ColoradoWestNileVirusSurveillanceMap2002
129
130
131
ReportableConditions TheColoradoDepartmentofPublicHealthandEnvironment,bystatestatute,collectsdataonavariety of reportable conditions, including a number of infectious diseases and cancer. Reportable conditionsforwhichGarfieldCounty,atleastoneofthecomparisoncounties,oranothercountyinthe statereportedoneormorecasesoftheconditionaredetailedinTables25Athrough25Gforeachof the years 2000, 2001, 2002, 2003, 2004, 2005, and 2006. Numbers of cases and rates (per 1,000 population,2000censusdata)forthefourcountiesandthestatearegiven.GarfieldCountycasesare highlightedtoalloweasiercomparisonwiththeothercountydata. In 2000, none of the rates for reportable conditions in Garfield County were ranked highest, whencomparedtothestateandcomparisoncountyrates. In 2001, the only reportable condition for which Garfield County had the highest rate was giardiasis. In 2002, Garfield Countys rates were highest for influenza, meningitis, and invasive strep pneumonia. In2003,GarfieldCountysratesforgiardiasis,hepatitisA,andinfluenzawerethehighest. In2004,GarfieldCountysratesforencephalitis,salmonellosis,andshigellosiswerethehighest. TherateformeningitiswasthesameasMesaCountysrate,andwashigherthanforthestate andtheothertwocomparisoncounties. In 2005, Garfield Countys rates were the highest for encephalitis (1 case out of only 8 in the statethatyear),hepatitisA,malaria,meningitis,shigellosis,andvaricella(chickenpox). In 2006, Garfield Countys rates for hanta virus pulmonary syndrome, hepatitis A and C, salmonellosis,andshigellosiswerethehighest. Atleastonecaseofthefollowingdiseaseswasreportedduringeachyearoftheperiod,20002006: camphylobacter giardiasis influenza salmonella shigellosis hepatitisC hepatitisAwasreportedinfiveofthesevenyears hepatitisBwasreportedinsixofthesevenyears.
133
Delta Garfield Garfield Mesa Delta Cases Rate Cases Rate Cases AMEBIASIS 0 0.00 0 0.00 3 CAMPYLOBACTER 6 2.14 6 1.36 16 CRYPTOSPORIDIOSIS 1 ** 0 0.00 0 E0COLI 1 ** 3 0.68 7 ENCEPHALITIS 0 0.00 0 0.00 2 GIARDIASIS 2 ** 5 1.13 11 HAEMOPHILUSINFLUENZAE 0 0.00 0 0.00 0 HANTAVIRUSPULMONARYSYNDROME 1 ** 0 0.00 0 HEMOLYTICUREMICSYNDRM 0 0.00 0 0.00 0 HEPATITISA 0 0.00 1 0.23 4 HEPATITISB 2 ** 1 0.23 4 HEPATITISC 8 2.86 18 4.07 60 INFLUENZA 3 1.07 2 ** 64 LISTERIOSIS 0 0.00 2 ** 0 MALARIA 1 ** 0 0.00 2 MENINGITIS 0 0.00 1 0.23 5 MENINGOCOCCALDISEASE 0 0.00 0 0.00 1 MUMPS 0 0.00 0 0.00 0 PERTUSSIS 0 0.00 0 0.00 31 SALMONELLOSIS 6 2.14 7 1.58 15 SHIGELLOSIS 1 ** 4 0.90 3 STREPPNEUMOINVASIVE 0 0.00 0 0.00 0 TOXICSHOCKSTREP 0 0.00 0 0.00 0 TYPHOIDFEVER 0 0.00 0 0.00 0 ColumnTotal 32 11.42 50 11.30 228 Ratesper1,000population,basedoncountypopulationaccordingto2000censusdata. **ratesnotcalculatedforfewerthan3cases. Table25A.2000ReportableConditions Mesa Rate 0.25 1.36 0.00 0.59 ** 0.93 0.00 0.00 0.00 0.34 0.34 5.10 5.44 0.00 ** 0.42 ** 0.00 2.63 1.27 0.25 0.00 0.00 0.00 19.38 Montrose Cases 1 2 0 1 0 4 0 0 0 1 3 2 2 0 0 0 0 0 0 12 0 0 0 0 28 Montrose Rate ** ** 0.00 ** 0.00 1.19 0.00 0.00 0.00 0.30 0.89 ** ** 0.00 0.00 0.00 0.00 0.00 0.00 3.56 0.00 0.00 0.00 0.00 8.32 Colorado Total 42 592 72 157 15 695 33 8 11 223 831 6802 592 11 30 211 35 1 487 692 269 226 2 2 12039 Colorado Rate 0.10 1.36 0.17 0.36 0.03 1.60 0.08 0.02 0.03 0.51 1.92 15.68 1.36 0.03 0.07 0.49 0.08 ** 1.12 1.59 0.62 0.52 ** ** 27.75
134
Delta Garfield Garfield Mesa Delta Cases Rate Cases Rate Cases AMEBIASIS 1 ** 1 0.22 1 CAMPYLOBACTER 6 2.09 8 1.73 19 CRYPTOSPORIDIOSIS 0 0.00 0 0.00 2 E0COLI 1 ** 1 0.22 2 ENCEPHALITIS 0 0.00 0 0.00 1 GIARDIASIS 4 1.39 10 2.17 15 HAEMOPHILUSINFLUENZAE 0 0.00 0 0.00 1 HEMOLYTICUREMICSYNDRM 1 ** 0 0.00 0 HEPATITISA 0 0.00 1 0.22 1 HEPATITISB 1 0.35 0 0.00 10 HEPATITISC 15 5.22 24 5.20 104 INFLUENZA 5 1.74 12 2.60 25 LISTERIOSIS 0 0.00 0 0.00 0 MALARIA 0 0.00 1 0.22 0 MENINGITIS 1 ** 2 ** 9 MENINGOCOCCALDISEASE 0 0.00 0 0.00 2 MUMPS 0 0.00 0 0.00 0 PERTUSSIS 0 0.00 0 0.00 14 SALMONELLOSIS 1 ** 3 0.65 9 SHIGELLOSIS 0 0.00 1 0.22 2 STREPPNEUMOINVASIVE 0 0.00 0 0.00 2 TOXICSHOCKSTREP 0 0.00 0 0.00 0 TYPHOIDFEVER 0 0.00 0 0.00 0 ColumnTotal 36 12.54 64 13.86 219 Ratesper1,000population,basedoncountypopulationaccordingto2000censusdata. **ratesnotcalculatedforfewerthan3cases. Table25B.2001ReportableConditions Mesa Rate ** 1.58 ** ** ** 1.25 ** 0.00 ** 0.83 8.66 2.08 0.00 0.00 0.75 ** 0.00 1.17 0.75 ** ** 0.00 0.00 18.23 Montrose Cases 0 8 0 0 0 2 0 0 0 2 11 1 0 0 0 0 0 1 4 0 0 0 0 29 Montrose Rate 0.00 2.31 0.00 0.00 0.00 ** 0.00 0.00 0.00 ** 3.18 ** 0.00 0.00 0.00 0.00 0.00 ** 1.16 0.00 0.00 0.00 0.00 8.38 Colorado Total 34 766 44 95 10 632 38 11 89 748 5708 1743 10 25 388 38 3 389 593 245 317 2 1 11929 Colorado Rate 0.08 1.72 0.10 0.21 0.02 1.42 0.09 0.02 0.20 1.68 12.84 3.92 0.02 0.06 0.87 0.09 0.01 0.87 1.33 0.55 0.71 ** ** 26.83
135
Table25C.2002ReportableConditions Delta Garfield Garfield Mesa Delta Cases Rate Cases Rate Cases AMEBIASIS 0 0.00 0 0.00 0 CAMPYLOBACTER 2 ** 3 0.63 3 CRYPTOSPORIDIOSIS 1 ** 0 0.00 5 E0COLI 0 0.00 1 ** 1 ENCEPHALITIS 0 0.00 0 0.00 0 GIARDIASIS 2 ** 8 1.69 7 HAEMOPHILUSINFLUENZAE 0 0.00 0 0.00 0 HANTAVIRUSPULMONARYSYNDROME 0 0.00 0 0.00 0 HEMOLYTICUREMICSYNDRM 0 0.00 1 ** 1 HEPATITISA 1 ** 0 0.00 2 HEPATITISB 0 0.00 1 ** 14 HEPATITISC 11 3.77 34 7.17 96 INFLUENZA 35 11.99 126 26.56 259 LISTERIOSIS 0 0.00 0 0.00 0 MALARIA 0 0.00 0 0.00 0 MENINGITIS 1 ** 5 1.05 3 MENINGOCOCCALDISEASE 2 ** 0 0.00 0 MUMPS 0 0.00 0 0.00 0 PERTUSSIS 6 2.06 0 0.00 8 SALMONELLOSIS 4 1.37 4 0.84 7 SHIGELLOSIS 3 1.03 4 0.84 3 STREPPNEUMOINVASIVE 0 0.00 6 1.26 0 TOXICSHOCKSTREP 0 0.00 0 0.00 0 TYPHOIDFEVER 0 0.00 0 0.00 0 WESTNILEVIRUS 0 0.00 0 0.00 0 ColumnTotal 68 23.29 193 40.68 409 Ratesper1,000population,basedoncountypopulationaccordingto2000censusdata. **ratesnotcalculatedforfewerthan3cases. Mesa Rate 0.00 0.24 0.41 ** 0.00 0.57 0.00 0.00 ** ** 1.14 7.84 21.15 0.00 0.00 0.24 0.00 0.00 0.65 0.57 0.24 0.00 0.00 0.00 0.00 33.40 Montrose Cases 0 1 0 1 0 9 0 0 0 0 0 6 5 0 0 0 0 0 0 6 0 0 0 0 0 28 Montrose Rate 0.00 ** 0.00 ** 0.00 2.54 0.00 0.00 0.00 0.00 0.00 1.69 1.41 0.00 0.00 0.00 0.00 0.00 0.00 1.69 0.00 0.00 0.00 0.00 0.00 7.90 Colorado Total 32 616 57 109 15 578 36 1 14 75 789 6128 3363 7 25 254 25 2 464 615 213 392 7 5 12 13834 Colorado Rate 0.07 1.36 0.13 0.24 0.03 1.28 0.08 ** 0.03 0.17 1.74 13.55 7.44 0.02 0.06 0.56 0.06 ** 1.03 1.36 0.47 0.87 0.02 0.01 0.03 30.59
136
Table25D.2003ReportableConditions Delta Garfield Garfield Mesa Delta Cases Rate Cases Rate Cases AMEBIASIS 0 0.00 0 0.00 0 CAMPYLOBACTER 8 2.70 1 ** 10 CRYPTOSPORIDIOSIS 0 0.00 0 0.00 7 E0COLI 1 ** 1 ** 3 ENCEPHALITIS 0 0.00 0 0.00 0 GIARDIASIS 0 0.00 8 1.65 7 HAEMOPHILUSINFLUENZAE 1 ** 0 0.00 3 HANTAVIRUSPULMONARYSYNDROME 0 0.00 0 0.00 0 HEMOLYTICUREMICSYNDRM 0 0.00 1 ** 0 HEPATITISA 0 0.00 8 1.65 0 HEPATITISB 4 1.35 1 ** 14 HEPATITISC 27 9.10 36 7.44 120 INFLUENZA 144 48.54 444 91.75 413 LISTERIOSIS 0 0.00 0 0.00 0 MALARIA 0 0.00 1 ** 0 MENINGITIS 2 ** 1 ** 10 MENINGOCOCCALDISEASE 0 0.00 0 0.00 1 MUMPS 0 0.00 0 0.00 0 PERTUSSIS 1 ** 0 0.00 3 SALMONELLOSIS 1 ** 3 0.62 17 SHIGELLOSIS 2 ** 3 0.62 2 STREPPNEUMOINVASIVE 3 1.01 3 0.62 1 TOXICSHOCKSTREP 0 0.00 0 0.00 0 TYPHOIDFEVER 0 0.00 0 0.00 0 WESTNILEVIRUS 10 3.37 2 ** 19 ColumnTotals 204 68.77 513 106.00 630 Ratesper1,000population,basedoncountypopulationaccordingto2000censusdata. **ratesnotcalculatedforfewerthan3cases. Mesa Rate 0.00 0.80 0.56 0.24 0.00 0.56 0.24 0.00 0.00 0.00 1.12 9.59 33.02 0.00 0.00 0.80 ** 0.00 0.24 1.36 ** ** 0.00 0.00 1.52 50.37 Montrose Cases 0 2 0 0 0 2 0 0 0 1 1 13 41 0 0 0 1 0 1 0 0 1 0 0 9 72 Montrose Rate 0.00 ** 0.00 0.00 0.00 ** 0.00 0.00 0.00 ** ** 3.60 11.35 0.00 0.00 0.00 ** 0.00 ** 0.00 0.00 0.28 0.00 0.00 2.49 19.93 Colorado Total 25 605 38 78 3 458 39 5 8 59 684 5977 14880 9 22 709 27 1 368 492 331 365 3 4 2947 28137 Colorado Rate 0.05 1.32 0.08 0.17 0.01 1.00 0.09 0.01 0.02 0.13 1.49 13.03 32.44 0.02 0.05 1.55 0.06 ** 0.80 1.07 0.72 0.80 0.01 0.01 6.42 61.34
137
Delta Delta Garfield Garfield Mesa Table25E.2004Reportable Mesa Conditions Cases Rate Cases Rate Cases Rate AMEBIASIS 0 0.00 0 0.00 0 0.00 CAMPYLOBACTER 7 2.33 9 1.82 5 0.39 CRYPTOSPORIDIOSIS 0 0.00 0 0.00 3 0.23 E0COLI 0 0.00 0 0.00 4 0.31 ENCEPHALITIS 0 0.00 3 0.61 0 0.00 GIARDIASIS 6 1.99 5 1.01 12 0.94 HAEMOPHILUSINFLUENZAE 1 ** 0 0.00 1 ** HANTAVIRUSPULMONARY 0 0.00 0 0.00 0 0.00 SYNDROME HEMOLYTICUREMICSYNDRM 0 0.00 0 0.00 0 0.00 HEPATITISA 1 ** 1 ** 1 ** HEPATITISB 4 1.33 1 ** 5 0.39 HEPATITISC 30 9.97 49 9.93 153 11.97 INFLUENZA 0 0.00 1 ** 0 0.00 LISTERIOSIS 0 0.00 0 0.00 1 ** MALARIA 0 0.00 1 ** 0 0.00 MENINGITIS 1 ** 8 1.62 21 1.64 MENINGOCOCCALDISEASE 0 0.00 0 0.00 0 0.00 MUMPS 0 0.00 0 0.00 0 0.00 PERTUSSIS 26 8.64 1 ** 112 8.76 SALMONELLOSIS 3 1.00 6 1.22 4 0.31 SHIGELLOSIS 0 0.00 2 ** 1 ** STREPPNEUMOINVASIVE 11 3.66 10 2.03 13 1.02 TOXICSHOCKSTREP 0 0.00 1 ** 0 0.00 TYPHOIDFEVER 0 0.00 1 ** 0 0.00 VARICELLA(CHICKENPOX) 1 ** 41 8.31 28 2.19 WESTNILEVIRUS 27 8.98 5 1.01 127 9.94 Total 118 39.23 145 29.40 491 38.42 Ratesper1,000population,basedoncountypopulationaccordingto2000censusdata. **ratesnotcalculatedforfewerthan3cases.

Montrose Cases 0 3 3 0 0 7 1 0 0 0 0 34 0 0 0 0 0 0 1 1 1 1 0 0 44 11 107
Montrose Rate 0.00 0.81 0.81 0.00 0.00 1.90 ** 0.00 0.00 0.00 0.00 9.21 0.00 0.00 0.00 0.00 0.00 0.00 ** ** ** ** 0.00 0.00 11.91 2.98 28.97
Colorado Total 15 827 59 52 8 517 44 4 6 51 578 4818 26 13 19 424 15 3 1185 542 161 332 2 3 2040 296 12040
Colorado Rate 0.03 1.78 0.13 0.11 0.02 1.11 0.09 0.01 0.01 0.11 1.24 10.35 0.06 0.03 0.04 0.91 0.03 0.01 2.55 1.16 0.35 0.71 ** 0.01 4.38 0.64 25.88
138
Table25F.2005 Delta Delta Garfield Garfield Mesa Mesa Montrose ReportableConditions Cases Rate Cases Rate Cases Rate Cases AMEBIASIS 0 0.00 0 0.00 1 ** 1 CAMPYLOBACTER 4 1.32 10 1.97 10 0.77 6 CRYPTOSPORIDIOSIS 0 0.00 0 0.00 5 0.38 1 ENCEPHALITIS 0 0.00 1 ** 0 0.00 0 GIARDIASIS 2 ** 5 0.99 11 0.84 4 HAEMOPHILUS 1 ** 1 ** 4 0.31 0 INFLUENZAE HANTAVIRUSPULMONARY 0 0.00 0 0.00 0 0.00 0 SYNDROME HEMOLYTICUREMIC 0 0.00 0 0.00 0 0.00 0 SYNDRM HEPATITISA 0 0.00 2 ** 0 0.00 0 HEPATITISB 0 0.00 1 ** 2 ** 1 HEPATITISC 24 7.93 36 7.10 127 9.74 28 INFLUENZA 10 3.31 11 2.17 36 2.76 1 LISTERIOSIS 0 0.00 0 0.00 1 ** 0 MALARIA 0 0.00 1 ** 0 0.00 0 MENINGITIS 1 ** 4 0.79 5 0.38 0 MENINGOCOCCALDISEASE 0 0.00 0 0.00 1 ** 0 MUMPS 0 0.00 0 0.00 0 0.00 0 PERTUSSIS 5 1.65 15 2.96 81 6.21 3 SALMONELLOSIS 5 1.65 5 0.99 13 1.00 5 SHIGELLOSIS 0 0.00 6 1.18 0 0.00 0 STEC(shigatoxinproducing 1 ** 1 ** 6 0.46 0 E0coli) STREPPNEUMOINVASIVE 2 ** 4 0.79 16 1.23 0 TOXICSHOCKSTREP 0 0.00 0 0.00 3 0.23 0 TYPHOIDFEVER 0 0.00 0 0.00 0 0.00 0 VARICELLA(CHICKENPOX) 7 2.31 57 11.25 89 6.83 33 WESTNILEVIRUS 1 ** 1 ** 10 0.77 0 Total 63 20.82 161 31.77 421 32.29 83 Ratesper1,000population,basedoncountypopulationaccordingto2000censusdata. **ratesnotcalculatedforfewerthan3cases.
Montrose Rate ** 1.58 ** 0.00 1.06 0.00 0.00 0.00 0.00 ** 7.39 ** 0.00 0.00 0.00 0.00 0.00 0.79 1.32 0.00 0.00 0.00 0.00 0.00 8.71 0.00 21.91
Colorado Total 17 868 50 8 534 42 11 10 49 435 4235 1032 6 25 322 19 6 1383 582 170 83 421 6 7 1797 110 12228
Colorado Rate 0.04 1.84 0.11 0.02 1.13 0.09 0.02 0.02 0.10 0.92 8.97 2.19 0.01 0.05 0.68 0.04 0.01 2.93 1.23 0.36 0.18 0.89 0.01 0.01 3.81 0.23 25.89
139
Table25G.2006 ReportableConditions AMEBIASIS CAMPYLOBACTER CRYPTOSPORIDIOSIS ENCEPHALITIS GIARDIASIS HAEMOPHILUS INFLUENZAE HANTAVIRUS PULMONARYSYNDROME
HemolyticUremicSyndrm
Delta Cases 0 6 0 0 3 0 0 0 0 1 27 11 0 0 0 0 0 17 6 2 0 6 0 0 15 34 128
Delta Rate 0.00 1.93 0.00 0.00 0.97 0.00 0.00 0.00 0.00 ** 8.71 3.55 0.00 0.00 0.00 0.00 0.00 5.48 1.93 ** 0.00 1.93 0.00 0.00 4.84 10.96 41.28
Garfield Cases 0 0 0 0 3 1 1 0 1 2 51 1 1 0 4 1 1 12 12 6 0 7 0 0 23 2 129
Garfield Rate 0.00 0.00 0.00 0.00 0.57 ** ** 0.00 ** ** 9.77 ** ** 0.00 0.77 ** ** 2.30 2.30 1.15 0.00 1.34 0.00 0.00 4.41 ** 24.72
Mesa Cases 0 8 2 1 7 5 0 0 0 10 115 17 0 0 5 2 0 9 9 2 1 10 0 0 50 38 291
Mesa Rate 0.00 0.60 ** ** 0.53 0.38 0.00 0.00 0.00 0.75 8.63 1.28 0.00 0.00 0.38 ** 0.00 0.68 0.68 ** ** 0.75 0.00 0.00 3.75 2.85 21.85
Montrose Cases 0 5 0 0 3 0 0 0 0 2 22 5 0 0 5 0 0 15 3 0 1 3 0 0 14 13 91
Montrose Rate 0.00 1.41 0.00 0.00 0.84 0.00 0.00 0.00 0.00 ** 6.19 1.41 0.00 0.00 1.41 0.00 0.00 4.22 0.84 0.00 ** 0.84 0.00 0.00 3.94 3.66 25.60
Colorado Total 0 830 77 11 554 51 6 8 44 521 3490 775 12 25 241 22 51 710 625 238 109 422 1 7 1504 345 10679
Colorado Rate 0.00 1.73 0.16 0.02 1.15 0.11 0.01 0.02 0.09 1.08 7.26 1.61 0.02 0.05 0.50 0.05 0.11 1.48 1.30 0.50 0.23 0.88 ** 0.01 3.13 0.72 22.22
HEPATITISA HEPATITISB HEPATITISC INFLUENZA LISTERIOSIS MALARIA MENINGITIS MENINGOCOCCAL DISEASE MUMPS PERTUSSIS SALMONELLOSIS SHIGELLOSIS STEC(shigatoxin producingE0coli)
STREPPNEUMOINVASIVE
TOXICSHOCKSTREP TYPHOIDFEVER VARICELLA(CHICKENPOX) WESTNILEVIRUS Total

140
SexuallyTransmittedDiseases(STDs).Oftheconditionsthatarereportabletothestate health department, other than cancer, a recent trend in the frequency of STD diagnoses is worth noting. The worksheets and data to support the graphs shown below are provided in Appendix N; data are from the STD/HIV Surveillance Program, Disease Control and EnvironmentalEpidemiologyDivision,ColoradoDepartmentofPublicHealthandEnvironment. Forthetimeperiod,2003through2007,therateofreportedcasesofChlamydiainboth GarfieldandMesaCountieshassteadilyincreased.Itisnotclear,atthistime,whether thisrateincreaseisduetoanactualincreaseindiseaseorisanartifactofchangesin screeningpatternsinthetwocounties.Overall,GarfieldCountysratesarelowerthan those for Mesa County and the state, but higher than those for Delta and Montrose Counties. Both Delta and Montrose Counties show an increase in reported cases for 20042005, but then the rates are flat (Delta) or decrease (Montrose) from 2005 through2007.
Figure25H.Rateper100,000PopulationofReportedChlamydiaCasesinDelta,Garfield,Mesaand MontroseCountiesComparedtoColoradoChlamydiaCaseRateper100,000(20002007)
400 350 Rate per 100,000 300 250 200 150 100 50 0 2000 2001 2002 2003 Year 2004 2005 2006 2007 Delta County Garfield County Mesa County Montrose County Colorado
Similarly,therateofreportedcasesofGonorrheainGarfieldandMesaCountiessteadily increasedfrom2003through2007.GarfieldCountysratesareagainlowerthanthose for both Mesa County and the state, but higher than those for Delta and Montrose Counties.DeltaCountysGonorrheacasesshowedthesamepatternasforChlamydia; Montrose Countys Gonorrhea cases increased slightly over the period 2004 through 2007.
Figure25I.Rateper100,000PopulationofReportedGonorrheaCasesin Delta,Garfield,MesaandMontroseCountiesComparedtoColoradoGonorrheaCaseRateper 100,000andHealthyPeople2010Goal(20002007)
90 80 70 Rate per 100,000 60 50 40 30 20 10 0 -10 2000 2001 2002 2003 Year 2004 2005 2006 2007 Delta County Garfield County Mesa County Montrose County Colorado Healthy People 2010 Goal
141
GarfieldCountyhadthehighestrateofreportedHIVcasesamongthefourcountiesfor theyears2000,2001,2003,and2005.Theserateswerehigherthanthestateratefor theyears2001and2003.(Therewerenocasesreportedin2002,2006,or2007.)

Figure25J.Rateper100,000PopulationofReportedHIVCasesin Delta,Garfield,MesaandMontroseCountiesComparedtoColoradoHIVCaseRateper100,000 (20002007)
7 6 Rate per 100,000 5 4 3 2 1 0 2000 2001 2002 2003 2004 2005 2006 2007 Year of Diagnosis Delta County Garfield County Mesa County Montrose County Colorado
GarfieldCountyhadthehighestrateofreportedAIDScasesamongthefourcountiesfor theyears2000,2003,2004,and2005.(Therewerenocasesreportedin2001,2002,or 2006.)

Figure25K.Rateper100,000PopulationofReportedAIDSCasesin Delta,Garfield,MesaandMontroseCountiesComparedtoColoradoAIDSCaseRateper100,000 (20002007)
8 7 Rate per 100,000 6 5 4 3 2 1 0 2000 2001 2002 2003 2004 2005 2006 2007 Year of Diagnosis Delta County Garfield County Mesa County Montrose County Colorado
142
Hospital and Insurance Provider Data. Several other data sources were used to expand the picture of the health of Garfield County residents. Hospital discharge data and emergency room visit data provided comparative information on the more serious conditions for which Garfield County residents sought medical care. Health insurance claims data from a major health insurance provider operating on the Western Slope was another resource for information regarding health and healthcare usage among the comparison counties. Unfortunately,ourattemptstoobtainprimarycareproviderinformationwerenotsuccessful. HospitalDischargeData Table26providesinformationonhospitalizationratesandhospitalchoiceforGarfieldCounty residents during 2005 and 2006. Out of more than 4,000 hospitalizations of Garfield County residentsduringboth2005and2006,approximately60%ofthosepatientswerehospitalizedat ValleyViewHospitalinGlenwoodSprings.Approximately16%and12%ofthepatientswere hospitalizedinSt.MarysHospitalandMedicalCenter(GrandJunction)orGrandRiverHospital District(Rifle),respectively.Therestwerehospitalizedinotherareaorregionalhospitals. Table 27 shows the top 10 DRG (DiagnosisRelated Group) categories for Garfield County residentshospitalizedin2000through2005.DiagnosisRelatedGroupisasystemthatwas developedfortheU.S.HealthCareFinanceAdministrationtoclassifyhospitalcasesintooneof approximately500groups(DRGs)expectedtohavesimilarhospitalresourceuse.Thesystemis usedbyMedicareandMedicaidaspartofaprospectivepaymentplan.Ittakesintoaccount diagnoses, procedures, age, gender and the presence of complications or coexisting medical conditions. Forthepurposesofthisstudy,theapproximately500DRGcategorieswerecollapsedinto130 consolidatedcategoriesforwhichdatawereobtained.These130categorieswerethenfurther groupedintolarger,relationshipcategories,thecontentsofwhich,aredescribedbelow. Duringthissixyearperiod,withtheexceptionof2003,circulationdisordersandboneandjoint disorders were either the first or second most common reasons for hospitalization (in 2003, birthing and pregnancy disorders was number one, and circulation disorders was third). Becauseofthefrequencywithwhichrespiratoryandnervoussystemconcernswerementioned by Garfield County residents, and the potential for air quality issues to cause or exacerbate respiratoryconditions,thesetwoDRGcategoriesarehighlighted.Respiratorydisorderswere eitherthefourthorfifthmostcommonreasonforhospitalizationinallyears;nervoussystem disorderswererankedeitherseventhoreightheachyear.Digestivedisordersrankedthirdin fouroutofthesixyears.Neonataldisordersweresixthinfrequencyduringtheentiresixyear period.
141
Table26.HospitalUtilization,GarfieldCountyResidents20052006(ColoradoHospitalAssociation)
Patient Origin Garfield County Valley View Hospital Saint Marys Hospital and Medical Center Grand River Hospital District Aspen Valley Hospital
University Hospital
Presbyterian Saint Lukes Medical Center Cnty Cases 69 19 14 18 3 9 6 Cnty Share 1.6% 1.7% 1.4% 2.0% 0.6% 2.1% 2.2% 0.0%
Vail Valley Medical Center
FY 2005 TOTAL GLENWOOD SPRINGS RIFLE CARBONDALE PARACHUTE NEW CASTLE SILT BATTLEMENT MESA
Cnty Cases 2,384 914 415 519 95 294 146 1
Cnty Share 55.9% 80.0% 42.9% 57.0% 17.9% 68.4% 52.7% 10.0%
Cnty Cases 668 65 206 65 229 48 49 6
Cnty Share 15.7% 5.7% 21.3% 7.1% 43.2% 11.2% 17.7% 60.0%
Cnty Cases 515 9 261 4 172 22 46 1
Cnty Share 12.1% 0.8% 27.0% 0.4% 32.5% 5.1% 16.6% 10.0%
Cnty Cases 277 28 5 224 14 5 1
Cnty Share 6.5% 2.5% 0.5% 24.6% 0.0% 3.3% 1.8% 10.0%
Cnty Cases 79 34 11 15 1 7 11 -
Cnty Share 1.9% 3.0% 1.1% 1.6% 0.2% 1.6% 4.0% 0.0%
Cnty Cases 62 20 9 16 4 9 4 0
Cnty Share 1.5% 1.8% 0.9% 1.8% 0.8% 2.1% 1.4% 0.0%
142
FY 2006 COUNTY TOTAL GLENWOOD SPRINGS RIFLE CARBONDALE PARACHUTE NEW CASTLE SILT BATTLEMENT MESA
Cnty Cases 2,346 863 371 543 103 292 171 3
Cnty Share 55.8% 78.8% 38.7% 58.7% 19.6% 74.9% 57.2% 27.3%
Cnty Cases 674 84 237 40 217 42 49 5
Cnty Share 16.0% 7.7% 24.7% 4.3% 41.3% 10.8% 16.4% 45.5%
Cnty Cases 513 4 280 2 157 25 43 2
Cnty Share 12.2% 0.4% 29.2% 0.2% 29.8% 6.4% 14.4% 18.2%
Cnty Cases 272 26 6 229 5 6 -
Cnty Share 6.5% 2.4% 0.6% 24.8% 0.0% 1.3% 2.0% 0.0%
Cnty Cases 66 21 6 23 6 6 4 -
Cnty Share 1.6% 1.9% 0.6% 2.5% 1.1% 1.5% 1.3% 0.0%
Cnty Cases 76 25 11 17 4 4 15 -
Cnty Share 1.8% 2.3% 1.1% 1.8% 0.8% 1.0% 5.0% 0.0%
Cnty Cases 62 21 14 17 2 6 2 0
Cnty Share 1.5% 1.9% 1.5% 1.8% 0.4% 1.5% 0.7% 0.0%
143
Table26.HospitalUtilization,GarfieldCountyResidents20052006(continued)
Patient Origin Garfield County The Childrens Hospital Cnty Cases TOTAL GLENWOOD SPRINGS RIFLE CARBONDALE PARACHUTE NEW CASTLE SILT BATTLEMENT MESA FY 2006 COUNTY TOTAL GLENWOOD SPRINGS RIFLE CARBONDALE 37 15 4 8 2 7 1 0 Cnty Cases 35 14 9 3 Cnty Share 0.9% 1.3% 0.4% 0.9% 0.4% 1.6% 0.4% 0.0% Cnty Share 0.8% 1.3% 0.9% 0.3% Other Hospitals Cnty Cases 176 38 43 41 24 20 9 1 Cnty Cases 161 37 25 51 Cnty Share 4.1% 3.3% 4.4% 4.5% 4.5% 4.7% 3.2% 10.0% Cnty Share 3.8% 3.4% 2.6% 5.5% Total
FY 2005
Total Cases 4,267 1,142 968 910 530 430 277 10 Total Cases 4,205 1,095 959 925
144
PARACHUTE NEW CASTLE SILT BATTLEMENT MESA
4 2 2 1
0.8% 0.5% 0.7% 9.1%
33 8 7 0
6.3% 2.1% 2.3% 0.0%
526 390 299 11
Table27.Top10DRGcategoriesbyyearinGarfieldCounty(ColoradoHospitalAssociation)
2000 per 1000 2001 per 1000 2002 per 1000 2003 per 1000 2004 per 1000 2005 per 1000
DRG category
DRG category
DRG category
DRG category Birthing and Pregnancy Disorders Bone and Joint Disorders
DRG category
DRG category
Circulation Disorders Bone and Joint Disorders Birthing and Pregnancy Disorders
9.87
Circulation Disorders Bone and Joint Disorders
9.72
Bone and Joint Disorders Circulation Disorders
9.55
8.86
8.98
8.96
8.74
9.05
8.87
8.82
8.96
8.11
8.54
Digestive Disorders Birthing and Pregnancy Disorders
7.78
Digestive Disorders
8.39
Circulation Disorders
8.64
7.85
7.76
Respiratory Disorders
8.45
7.52
Respiratory Disorders
8.24
Digestive Disorders
7.67
7.52
7.56
145 Birthing and Pregnancy Disorders Neonatal Disorders Nervous System Disorders Female Reproductive Disorders Nutritional, Metabolic and Thyroid Disorders
Digestive Disorders Neonatal Disorders Nervous System Disorders Female Reproductive Disorders Nutritional, Metabolic and Thyroid Disorders
8.38
Respiratory Disorders Neonatal Disorders Female Reproductive Disorders Nervous System Disorders
6.89
7.74
Respiratory Disorders Neonatal Disorders Nervous System Disorders Female Reproductive Disorders
7.36
6.73
6.93
6.21
6.04
6.07
5.29
5.09
4.78
3.82
3.92
3.96
3.51
4.22
3.59
3.77
3.66
3.86
3.37
3.22
3.53
2.58
Pancreas and Liver Disorders Nutritional, Metabolic and Thyroid Disorders
2.95
2.51
Pancreas and Liver Disorders
3.14
2.88
Pancreas and Liver Disorders Nutritional, Metabolic and Thyroid Disorders
2.51
2.12
2.69
Urinary Tract Disorders
2.49
2.52
2.47
2.49
146
AmoredetailedanalysisofthehospitalizationDRGsbycountyofresidencefortheyears2000 through2006andforthefirstquarterof2007follows.Theratesreferredtoandshowninthe graphsareper1,000individuals,basedonthe2000censusdataforcountypopulation.(The completedatasetandmoredetailedsubcategorygraphsmaybefoundinAppendixO.) AllHospitalizations Delta County has the highest rate over the 6.25year period for hospitalizations. Garfield County has either the lowest or the nexttolowest rate for this period, and showed a downwardtrendinhospitalizationratefrom2005intothefirstquarterof2007. Figure67.AllHospitalizationsbyCountyofResidence
AllTrauma Garfield Countys trauma rates were generally the same as or lower than the rates for the comparisoncountiesforthe6.25yeartimeperiod.
147
Figure68.AllTraumaHospitalizationsbyCountyofResidence
AlcoholandDrugDisorders This area was of concern to some Garfield County residents because of a perception that an influxofnaturalgasindustryworkerscouldresultinanincreaseintheuseandabuseofalcohol anddrugs.Withrespect,specifically,tohospitalizationasaresultofalcoholanddrugrelated disorders,MesaCountyhasthehighestratesforthestudyperiod,althoughtheirratedropped in2006toessentiallythesameasfortheothercounties.GarfieldCountysratesweresimilar tothoseDeltaCountyandlowerthantheratesforMontroseCountythrough2003.In2003, Montrosesawadecreaseinalcoholanddrughospitalizations,whichbroughttheirratetoone thatwassimilartothoseofGarfieldandDeltaCounties.GarfieldCountysrateappearedtobe increasingin2007,at whichtimeithadthehighestrateofthefourcounties.Caution:one quartersdataisnotgenerallysufficienttosafelypredictatrend. Figure69.AlcoholandDrugDisordersHospitalizationsbyCountyofResidence
148
BirthingandPregnancyDisorders Garfield Countys overall rates for this category were consistently higher than those for the comparison counties. The percentages of complicated births, out of all births, for Delta, Garfield, Mesa, and Montrose Counties were 53.4, 58.0, 61.4, and 53.0, respectively. Postpartumcomplications,inparticularhavebeensteadilyincreasinginGarfieldCountysince 2004.Thiscategoryalsoincludesabortions,cesareansections,ectopicpregnancies,falselabor, other antepartum diagnoses with or without surgical procedures, threatened abortion, and vaginaldeliverieswithandwithoutcomplications.Furtherbreakdownofinformationforthis categoryisprovidedbelow. Figure70.BirthingandPregnancyDisordersHospitalizationsbyCountyofResidence
NeonatalDisorders Neonatal disorder hospitalization rates in Garfield County have been relatively stable, and generally lower than the rates for Mesa and Montrose Counties and similar to the rates for Delta County. Mesa Countys neonatal disorder hospitalization rates have been consistently higher than for the other three counties. This category includes extreme immaturity or respiratory distress at birth, prematurity, fullterm neonates with significant problems, and neonataldeathortransfertoanotherfacilityforcare.
149
Figure71.NeonatalDisordersHospitalizationsbyCountyofResidence
NormalNewbornHospitalizations Ontheotherhand,GarfieldCountyalsohasthehighestratesfornormalnewbornsborninthe hospital.Thesehigherratesforbothnormalandabnormalbirthsmostlikelyreflectthefact thatGarfieldCountyhasthehighestbirthratesamongthefourcountiesfortheyearsstudied (Tables1013.VitalStatistics). Figure72.NormalNewbornsbyCountyofResidence
150
FemaleandMaleReproductiveDisordersHospitalizations To complete the picture of factors that might influence pregnancy and birth outcomes, the following graphs show data for hospitalizations related to hospitalizations for treatment of reproductive organ disorders. Garfield County hospitalization rates for female reproductive disorders were generally the same as or lower than the rates for the other three counties. MontroseCountyhadthehighestratesforthiscategoryforallyears,withtheexceptionofthe Year2000,inwhichDeltaCountyhadthehighestrate.Thiscategoryincludesinfectionsand malignancies of the female reproductive organs, menstrual disorders, surgical procedures involvingfemalereproductiveorgans,andtuballigations. GarfieldCountyhospitalizationratesformalereproductivedisordersweregenerallythesame asorlowerthantheratesfortheotherthreecounties.DeltaCountyhadthehighestratesfor this category for all years. This category includes complications of benign prostatic hypertrophy, transurethral prostectomy, inflammation and malignancies of the male reproductiveorgans,andothersurgicalproceduresinvolvingmalereproductiveorgans.
Figure73.FemaleReproductiveDisordersHospitalizationsbyCountyofResidence Figure74.MaleReproductiveDisordersHospitalizationsbyCountyofResidence
151
BoneandJointDisorders Delta County has the highest rates throughout the 6.25year period. Garfield Countys rates werethelowestofthefourcounties,andweregenerallystable.Thiscategoryincludesback and neck and other joint procedures, fractures, strains, sprains and dislocations, and musculoskeletalandconnectivetissuedisorders. Figure75.BoneandJointDisordersHospitalizationsbyCountyofResidence
CirculationDisorders DeltaCountyhasthehighestratesthroughoutthe6.25yearperiodforthiscategory.Garfield Countysrateswerethelowestofthecomparisoncountiesandweregenerallyconsistentover theperiod.Therewasaslightdownwardtrendin2007,atwhichtime,theratesfortheother three counties were trending up. This category includes acute myocardial infarction, angina pectoris, atherosclerosis, cardiothoracic procedures, cardiovascular disorders and procedures, chest pain, circulatory system disorders, coronary bypass, heart failure and shock, hypertension,peripheralvasculardisorders,andsyncopeandcollapse. Figure76.CirculationDisordersHospitalizationsbyCountyofResidence
152
DigestiveDisorders MesaCountyhasthehighestoverallratesforthiscategory.GarfieldCountysoverallratesare essentially the same as those in Delta and Montrose Counties until 2005, when Montrose County began to see an increasing trend in these disorders. Garfield Countys rate began a downward trend for the same time period. This category includes appendectomy; digestive malignancy; gastrointestinal hemorrhage; gastrointestinal obstruction; hernia procedures; pepticulcer;smallandlargebowelprocedures;stomach,esophagealandduodenalprocedures; and other digestive syndrome diagnoses. For some of these subcategories, Garfield Countys rateswerethelowestofthefourcounties;GarfieldCountysrateswereneverhighestforany subcategory. Figure77.DigestiveDisordersHospitalizationsbyCountyofResidence
Ear,Nose,ThroatDisorders Overall,andingeneral,thehospitalizationratesforthisgroupofdisordersweresimilaramong thefourcounties,althoughDeltaCountyhasapeakinthefirstquarterof2007.Whenlooking at the subcategories, however, Garfield Countys hospitalization rates for otitis media and upper respiratory infections for children, ages 017 years, are the highest among the four counties.Thisisalsotrueforthehospitalizationratesforbronchitisandasthmainchildren.In contrast, the hospitalization rates for otitis media, upper respiratory infection, bronchitis and asthmainadultsarethelowestamongthefourcounties.GarfieldCountyalsohasthelowest rate of COPD hospitalizations. This category includes dysequilibrium; ear, nose, mouth and throatdiagnoses;otitismediaanupperrespiratoryinfection.
153
Figure78.Ear,Nose,ThroatDisordersHospitalizationsbyCountyofResidence
InfectiousandParasiticDiseases HospitalizationsforinfectiousandparasiticdiseaseswerehighestinDeltaCountythroughout themajorityofthetimeperiodstudied.Ratesforallcountiesappeartobedecreasingfroma highin2005,withtheexceptionofratesforMontroseCounty,whichhaveremainedessentially stable.GarfieldCountyshospitalizationratesforthesediseasesweregenerallylowerthanfor theotherthreecounties,increasingtoapeakin2005,andthendroppingsteadilythroughthe 1stquarterof2007.Thiscategoryincludespostoperativeandposttraumaticinfections,fevers of unknown origin, viral illnesses, septicemia, and other infectious or parasitic disease diagnoses. Figure79.InfectiousandParasiticDiseasesHospitalizationsbyCountyofResidence
154
MentalDisorders Garfield Countys hospitalization rates for mental disorders are the lowest among the four countiesthrough2003;theratesremainstablethrough2005,afterwhichthereisadownward trendintheratethroughthefirstquarterof2007.MesaCountysratesarehighestamongthe four counties. After a significant drop in the rate in 2006, the Mesa County rate becomes essentially that same as for the other counties. This category includes acute adjustment reactionandpsychosocialdysfunction,childhoodmentaldisorders,depressiveneuroses,other neuroses,psychoses,andothermentaldisorders. Figure80.MentalDisordersHospitalizationsbyCountyofResidence
NervousSystemDisorders GarfieldCountyhastheoveralllowestratesforthe6.25yearperiod;DeltaCountysratesare the highest for the years 2001 through 2005. With respect to the subcategory, seizure and headache for children (ages 017), Garfield Countys rates show an upward trend since 2004, with a big jump in the first quarter of 2007. Seizure and headache hospitalization rates for individualsover17yearsofageshowadownwardtrendsince2004,andaresimilartothoseof the other counties. [Subcategory data are not shown here, but are provided in Appendix N. This category includes cerebrovascular disorders, concussion, cranial and peripheral nerve disorders,craniotomy,extracranialprocedures,intracranialhemorrhageorcerebralinfarction, nervous system infection except viral meningitis, nervous system neoplasms, seizure and headache, spinal disorders and injuries, stupor and coma (nontraumatic and traumatic), transientischemia,viralmeningitis,andothernervoussystemdisorders.
155
Figure81.NervousSystemDisordersHospitalizationsbyCountyofResidence
Nutritional,MetabolicandThyroidDisorders MontroseCountyhasthelowestratesuntil2005;ratesincreasedbetween2004and2005,then leveledoff.GarfieldCountysratesshowedadownwardtrendbetween2002and2003,aspike in2005,anddecreasesagainin2006.GarfieldCountysdiabetesratesarethelowestamong thecountiesforbothchildrenandadults.Thiscategoryincludesdiabetes(ages035and>35); endocrine, nutritional and metabolic disorders, nutritional and miscellaneous metabolic disorders(ages017and>17),andobesityprocedures. Figure82.Nutritional,MetabolicandThyroidDisordersHospitalizationsby
CountyofResidence
156
PancreasandLiverDisorders DeltaCountysratesareconsistentlythehighestamongthefourcounties.Ratesfortheother counties across the time period are similar. This category includes cirrhosis and alcoholic hepatitis, malignancy of hepatobiliary system or pancreas, and pancreas and liver disorders. Garfield Countys hospitalization rates are the highest among the counties for most years, surpassedonlybyDeltaCountyin2003and2006. Figure83.PancreasandLiverDisordersHospitalizationsbyCountyofResidence
RedCellandClottingDisorders GarfieldCountysratesshowasteadydownwardtrendfrom2001through2006,withaslight increaseinthefirstquarterof2007.GarfieldCountysratesaregenerallylowerthanthosefor theotherthreecounties. Figure84.RedCellandClottingDisordersHospitalizationsbyCountyofResidence
157
RespiratoryDisorders Overall,GarfieldCountyratesforrespiratorydisordersarelowerthanfortheothercounties; Delta Countys rates are the highest. Subcategory data comparisons are more variable, however. (Subcategory data by age are not shown here, but are available in Appendix N.) GarfieldCountyhasthelowestratesforbronchitisandasthmainindividuals>17years,butthe highestratesforchildren(ages017).ChildhoodbronchitisandasthmaratesinGarfieldCounty took a big drop in 2005, but increased to 2004 levels in the first quarter of 2007. Chronic obstructivepulmonarydisease(COPD)ratesinGarfieldCountyarethelowestamongthefour counties.GarfieldCountyratesforrespiratoryinfectionsandinflammationsforbothadultsand childrenarethelowestforthefourcounties.Ratesforsimplepneumoniaandpleurisywere the lowest among the four counties for adults (>17 years), but highest for children (017) through 2005. (Childhood rates in other counties surpassed those for Garfield County from 2005throughthefirstquarterof2007.)Theotherdisordersincludedinthiscategoryaremajor chest procedures, interstitial lung disease, major chest trauma, pulmonary embolism, respiratoryneoplasms,andotherpulmonarydisorders. Figure85.AllRespiratoryDisordersHospitalizationsbyCountyofResidence
158
Figure86.RespiratoryDisordersHospitalizationsbyCountyofResidenceBronchitis, Asthma,COPDandRespiratoryInfectionsandInflammation
SkinDisorders Delta County has the highest ratesamong the four counties; the ratesfor the other counties aresimilaracrossthetimeperiod.Thiscategoryincludescellulitis;malignantbreastdisorders; skin graft or debridement; skin ulcers; trauma to skin, subcutaneous tissue and breast; and otherskindisorders. Figure87.SkinDisordersHospitalizationsbyCountyofResidence
159
Skingrafts,trauma,poisoningandallergicreactions GarfieldCountyshospitalizationratesforthiscategoryaregenerallythelowestamongthefour counties,althoughGarfieldCountysawajumpintraumaticinjuryhospitalizationforthefirst quarter of 2007. This category includes allergic reactions injury, poisoning and toxic effect diagnosis;surgicalproceduresforinjuries;poisoningandtoxiceffectsofdrugs;traumaticinjury; wounddebridementforinjuries. Figure88.SkinGrafts,TraumaandPoisoningHospitalizationsbyCountyofResidence
Urinarytractdisorders DeltaCountysratesarethehighestacrossthe6.25yearperiod;GarfieldCountysratesarethe lowest. This category includes kidney and urinary tract disorders and infections, kidney and urinarytractneoplasms,kidneytransplant,andrenalfailure. Figure89.UrinaryTractDisordersHospitalizationsbyCountyofResidence
160
AsthmaHospitalizationRates Coloradostatedataforcountyratesofhospitalizationdischargeswithaprincipaldiagnosisof asthma for the years 19932001 are provided in the graph below (Health Statistics Section, Colorado Department of Public Health and Environment). The ageadjusted rates are per 10,000 county residents. Rates were adjusted to the 2000 U.S. population using the direct methodappliedto10yearagegroups. Figure90.
Rocky Mountain Health Plans Member Data for Health Plan Usage by County; Grand River HealthDistrictandValleyViewHospitalEmergencyRoomAdmissionsDatabyZipCode Rocky Mountain Health Plans (RMHP) is a nonprofit health insurance provider, operating in Colorado. The companys main office is in Grand Junction, Colorado. RMHP provides health insurancetoalargeproportionofinsuredindividualsinthefourcomparisoncounties,although only about 7% of the population of Garfield County is insured by RMHO. We were able to obtain ambulatory, emergency room, inpatient and outpatient hospital visit data for RMHP membersinallfourcountiesfortheyears2000through2007.Thelimitationsofthesedata, withrespecttoprovidingvalidcomparativeinformationarethefactthatfewerindividualsare insured by RMHP in Garfield County than in the other three counties (although all rates are adjusted to member months), and the demographics of the RMHPinsured populations most likely varies among the counties. Thus, these comparisons should be viewed with caution. CompletedatasetsandgraphsmaybefoundintheAppendixP.
161
OnlythoseDRGdisordercategoriesthathadatleast200visitsper1000MemberMonthswere includedintheanalysis.Ratesareper1,000membermonths. The Grand River Hospital District (located in Rifle, CO) provided emergency room admission data for the years 2004, 2005, and 2006. These data were organized into the same DRG categoriesasusedfortheRMHPmemberdata.Admissionsdataweresortedbycommunityof residence.Asmightbeexpectedbythelocationofthehospitalrelativetocommunitieswithin Garfield County, there are few admissions recorded for residents of Carbondale, Glenwood Springs,orNewCastle.Thus,onlythedataforresidentsofParachute,Rifle,andSiltareusable andwillbediscussedinthisreport.WhereGrandRiverERvisitdatadifferedsubstantivelyfrom theRMHPdataforGarfieldCounty,asawhole,theGrandRiverdataarehighlighted.Please note that, because of the limitations described above for these data, it is not reasonable to directlycomparerates,butitispossibletocompareandobserveratetrends.Achangeinrate overaoneyearperiodisNOTsufficienttodeterminewhetheranapparenttrendissignificant oranoutlier.Tables2831discusstheserelativetrendsinthedata.Ratesareper1,000visits. Complete datasets and graphs may be found in the Appendix Q. Of note is the fact that Garfield County residents had the lowest usage rates for annual physical exam visits (adults) andwellcheckups(children)ofallofthecomparisoncountiesovertheentire8yearperiod. Withregardtotheselowrates,itisalsoimportanttonotethattheseratesareforaninsured populationforwhichannualphysicalexamsandwellchildcheckupsareprovided. Valley View Hospital in Glenwood Springs provided emergency room data by zip code of residenceforvisitsthatoccurredbetweenJanuary1,2004andDecember31,2006.Thesedata werealsoorganizedintotheDRGcategoriesusedforGrandRiverHospitalDistrictandRMHP memberdata.Similartowhatisdescribedabovefortheotheremergencyroomdata,thereare limitationstothepossibleuseandinterpretationoftheValleyViewHospitaldata.Byfar,the largest numbers of emergency room visits were from residents of the Glenwood Springs zip codes.Becausetheratecalculations(visitsbyDRGcategoryper1000persons)useanestimate ofthepopulationresidingineachzipcodeareaandthemarginoferrorincreasesinversely withthesizeofthearea(i.e.,thesmallerthegeographicareathelargerthemarginoferrorin the population estimate) the rates calculated for each population should only be taken as estimates.Toaddtotheuncertaintyofcomparingratesamongthesezipcodepopulations,the dataaresubdividedintoAdultandChildcategories.Theagecategoriesfordataprovidedby Valley View Hospital were different than those for data provided by Grand River Hospital DistrictandRMHP;ValleyViewHospitalcategorized025yearoldindividualsaschildren,while RMHPandGrandRiverHospitalDistrictdesignated018yearoldindividualsaschildren.Thus, theagecategoriesarenotentirelycomparable.Finally,thetimeperiodsfordataprovidedby Valley View Hospital and the other data sources are not completely overlapping. Because of theselimitationsincomparingandinterpretingtheValleyViewHospitalemergencyroomdata, with respect to the other two emergency room data sources and because of the relatively smallnumbersofemergencyroomvisitsforresidentsofzipcodeareasotherthanGlenwood Springs,onlythedatafortheGlenwoodSpringszipcodes(81601and81602)arediscussedin thereport.Itisalsoimportanttonotethat,inthecaseofsmallerpopulationsandshortertime periods(e.g.,theSeptemberthroughDecember2006period),asinglevisithasagreaterimpact
162
on the rate, thus making the rates unreliable and the data sets useful only for estimating trends.ThecompleteValleyViewHospitaldatasetandanalysisareavailableinAppendixR.
163 Table 28. Emergency Room Admissions Data DRGCategory
Accident,Injury,Trauma
CHAdata (20001stQ2007) GarfieldCountys injuryhospitalization ratesincreasedfor adults(>18years) after2005,butstill remainedlowerthan theratesforMesa andDeltaCounties. Accident,injuryand trauma hospitalizationrates forchildren(<18 years)residingin GarfieldCountyare thelowestamong thefourcounties. Notapplicable.
RMHP(20002007) InpatientHospital GarfieldCountys rateswerethe lowestamongthe fourcountiesuntil 2005forbothadults andchildren,and remainedthelowest forchildrenthrough 2007.
Outpatient/Ambulatory Sameasforinpatient visitrates.
EmergencyRoom Adultemergency roomvisitsrelated toaccident,injuryor traumaincreased from2003through 2005,thenslightly decreasedthrough 2007.
GRHDER (20042006) Emergencyroomvisit dataforGrandRiver HospitalDistrictshow increasingratesfor Silt, Parachute/Battlement Mesa,andRifle residentsforthe period20042006.
ValleyViewER* (20042006) AdultERvisitrates werestable throughoutthe period.Ratesfor children/young adultsdecreased between2004and 2005,then remainedstablefor therestofthe period.
Notapplicable.
AnnualPhysicals/Well ChildCheckups Ratesare consistentlyhigher forGarfieldCounty thanforDelta,Mesa andMontrose Counties. Birthing,Pregnancy,GYN GarfieldCounty residentshadlower utilizationratesthan didresidentsof Delta,Mesaand MontroseCounties. Thisislikelya reflectionofthe differencein utilizationpatterns ofinsuredand uninsured populations.
GarfieldCounty'srates arethelowestamong thefourcounties.Note: TheseareRMHPdata forinsuredindividuals! Sameasforinpatient visits.
Notapplicable.
Notapplicable.
Notapplicable.
Sameasfor inpatientvisits.
Datashowanincrease inbirthingand pregnancyrelated patientvisitsfor2005 through2006forSilt, Parachute/Battlement MesaandRifle residents.
AdultERvisitrates werestableover theperiod;ERvisit ratesforthis categorydecreased amongyoung adults.ERvisitsfor conditionsinthe perinatalperiod increasedthrough thetimeperiod.
Abbreviations:DRG=DiagnosisRelatedGroup;CHA=ColoradoHospitalAssociation;RMHP=RockyMountainHealthPlans;GRHD=GrandRiver HospitalDistrict.*DatapresentedforGlenwoodSpringszipcodeareasonly.
164
DRGCategory
CHAdata (20001stQ2007) GarfieldCounty's ratesarethelowest amongthefour counties.
Circulation/Cardiac
RMHP(20002007) InpatientHospital Outpatient/Ambulatory GarfieldCounty's Sameasforinpatient ratesarethelowest visits. amongthefour countiesforboth adultsandchildren. Ratesshowan increasingtrend from20032006.
EmergencyRoom Sameasfor inpatientvisits.
GRHDER (20042006) RatesforSilt, Parachute/Battlement MesaandRifle residentsshowan increasingtrendforthe timeperiod.
ValleyViewER* (20042006) AdultERvisitsfor thiscategorywere stableoverthe period;child/young adultvisits decreased.
Endocrine/Metabolic
Overall,Garfield County hospitalizationrates forthesedisorders arenotdifferent thantheratesforthe otherthreecounties. However,diabetes related hospitalizationrates forbothadultsand childreninGarfield Countyarethe lowestamongthe fourcounties.
Ratesare consistentwiththe CHAdataforboth adultsandchildren.
SameasforERvisits.
ERvisitratesfor thesedisordersare generallylowerfor bothadultandchild residentsofGarfield Countythanforthe otherthree counties;rates showaslight increasingtrend overthetime period.
RatesforSilt, Parachute/Battlement MesaandRifle residentsshowan increasingtrendforthe timeperiod.
AdultERvisits decreasedbetween 2004and2005,then remainedstablefor theremainderof theperiod. Child/youngadult ratesincreased between2004and 2005,then decreasedfrom 2005through2006.
165 CHAdata (20001stQ2007) GarfieldCountyhas thehighestratefor otitismedia,upper respiratory infections, bronchitis,and asthmainchildren. Asthma hospitalizationshave beenhigherthanin eitherMesaorDelta Countiessince1993, butwerelowerthan inMontroseCounty fortheperiod,1993 2001.Garfieldadult hospitalizationrates fortheseconditions werelowestamong thecounties,as weretheratesfor COPDandother respiratory infectionsand inflammationsfor bothadultsand children.Exceptions aresimple pneumoniaand pleurisyinchildren, which,until2005, werehighestamong thefourcounties. RMHP(20002007) InpatientHospital Outpatient/Ambulatory GarfieldCounty Hospitalization outpatient/ambulatory ratesforadult visitsforadultswerethe ENTand lowestamongthefour respiratory counties,buthighest conditionswere thelowestamong amongthefourcounties thefourcounties forchildren.Garfield foradultsand Countyratesdecreasedfor similartothe bothadultsandchildren othercountiesfor from2005through2007. children. GRHDER (20042006) ERvisitratesforSilt, Parachute/Battlement Mesa,andRifle residentsarevariable, butgenerally increasingoverthe timeperiod. ValleyViewER* (20042006) AdultERvisitsforotitis mediaincreasedbetween 2004and2005,then decreasedfrom2005 through2006.Ratesfor respiratoryconditions,in general,andasthma, specifically,werestable throughoutthetime period.Child/youngadult ERvisitsforotitismedia decreasedsteadily throughthetimeperiod. Ratesforrespiratory conditions,overall,were stable,howeverasthma ratesincreased.
DRGCategory
EmergencyRoom GarfieldCountyER visitratesfor adultswerethe lowestamongthe fourcountiesand werestableover the8yearperiod. Ratesforchildren inGarfieldCounty increasesteadily from20002003, butaresimilarto thosefortheother threecounties.
Ear,Nose,Throat, Respiratory

166
DRGCategory
CHAdata (20001stQ2007) Nodataforthis category.
InpatientHospital GarfieldCounty's ratesarethe lowestamongthe fourcountiesfor bothadultsand children.
RMHP(20002007) Outpatient/Ambulatory Sameasfor hospitalizationrates.
EmergencyRoom Sameasfor hospitalization rates.
GRHDER (20042006) RatesforSilt increasedsharply between2004and 2005;rateincreases forParachuteand Rifleweremore gradualforthesame timeperiod.Rifle's rates continuedtoincrease through2006,while ParachuteandSilt showeddecreasing rates. RatesforSilt, Parachute/Battlement MesaandRifle residentsshowan increasingtrendfor thetimeperiod.
ValleyViewER* (20042006) AdultERvisitswerelow andstablethroughout theperiod.Child/young adultratesshoweda steadydecreaseoverthe timeperiod.
Eye
GarfieldCounty's ratesareeitherthe lowestamongthe fourcounties,or aresimilartothose Gastrointestinal/Urinary oftheother countiesforthis groupofdisorders.
Ratesare consistentwiththe CHAdataforboth adultsand children.
Ratesareconsistentwith theCHAdataforboth adultsandchildren.
GarfieldCounty ratesaresimilarto thosefortheother counties.
Infection
GarfieldCounty outpatient/ambulatory visitsforadultswerethe lowestamongthefour counties,buthighest amongthefourcounties forchildrenduringthe timeperiod2000through 2002.
ERvisitratesfor thesedisordersare lowerforadultsin GarfieldCounty thanfortheother threecounties,and alsogenerallylower forchildren,but increasingoverthe timeperiod. ERratesforGarfield Countyadultswere inconsistent, withoutasustained trendineither direction,but similartothosefor theothercounties. ChildERvisitsfor infectionrelated conditionswere generallyhigher thanothercounties.
AdultERvisitsforthis categorywerestable throughouttheperiod. Child/youngadultvisits increasedbetween2004 and2005,thendecreased between2005and2006.
Ratesforinfection relatedERvisitswere variableoverthetime period,without consistenttrendsfor thecommunitiesof Parachute/Battlement Mesa,SiltandRifle.
AdultERvisitsfor infectionsdecreased throughouttheperiod,as didchild/youngadult visits.
167 CHAdata (20001stQ2007) GarfieldCounty's ratesarethe lowestamongthe fourcounties. RMHP(20002007) Outpatient/Ambulatory EmergencyRoom Sameasfor Sameasfor hospitalizationrates. hospitalizationrates. GRHDER (20042006) RatesforRifleandSilt showagradual increase;ratesfor Parachute/Battlement Mesaincreased steeplybetween2005 and2006. ValleyViewER* (20042006) OveralladultERvisitrates decreasedbetween2004 and2005,thenincreased to2004levelsbetween 2005and2006. Drug/alcoholrelated visitsshowedasteady increaseovertheperiod. Thesamepatternwas seenforchildrenand youngadults.
DRGCategory
Mental
InpatientHospital GarfieldCounty ratesarelower thanfortheother countiesand steadilydecreasing overthetime period.
Musculoskeletal
GarfieldCounty's ratesarethe lowestamongthe fourcounties. Overall,Garfield hospitalization ratesforthese disordersarelower thanfortheother counties.Forthe subcategoryof seizureand headache,ratesfor childreninGarfield Countyhave increasedsince 2004,while decreasingin adults.
Ratesare consistentwiththe CHAdataforboth adultsand children. Ratesare consistentwiththe CHAdataforboth adultsand children.
AdultERvisitswere stableoverthetime period,aswere child/youngadultvisits. AdultERvisits,including thoserelatedto headachesormigraines, werestableoverthe period.Child/youngadult visitsoverallincreased between2004and2005, thendecreasedthrough 2006.Visitsrelatedto headachesormigraines increased.
GarfieldCountyER visitrates,overall, werethelowest amongthefour counties.
NervousSystem
168
DRGCategory
Pancreas/Liver
CHAdata (20001stQ2007) InpatientHospital Ratesforthese Nodataforthis conditionsforDelta category. Countywere consistentlythe highestamongthe fourcounties. GarfieldCounty's rateswerenot differentfrom thoseofMesaand Montrose Counties.
RMHP(20002007) Outpatient/Ambulatory EmergencyRoom Nodataforthis Nodataforthis category. category.
GRHDER (20042006) Nodataforthis category.
ValleyViewER* (20042006) Includedwith Gastrointestinal/urinary category.
RedCell/Clotting
GarfieldCounty's ratessteadily decreasedfrom 2001through2006.
Skin/Allergy
Ratesforthese conditionsforDelta Countywere consistentlythe highestamongthe fourcounties. GarfieldCounty's rateswerenot differentfrom thoseofMesaand Montrose Counties.
GarfieldCounty's ratesarethe lowestamongthe counties,showing periodsof moderaterate decrease(2001 2003)andincrease (20032005).
GarfieldCounty'srates arethelowestamong thefourcounties, increasingslightlyfrom 2003to2004,then decreasingfrom2005 through2007.
AdultERratesincreased between2004and2005/ Child/youngadultrates werestablethrough 2005,thendecreased through2006. GarfieldCounty's RatesforSilt, AdultERrateswere ratesarethelowest Parachute/Battlement stable.Child/youngadult amongthefour MesaandRifle ratesdecreasedthrough counties,andwere residentsshowan theperiod. generallystable increasingtrendfor throughoutthe8year thetimeperiod. period,exceptforan increasebetween 2003and2004, decreasingto previousratesin 2006.(Ratesfor childreninGarfield Countydecreased steadilyfrom2005 through2007.)
169
Table29.InpatientHospitalAdmissionsData
RMHPMemberData:20002007* DRGCategory Accident,Injury&Trauma Adult(18years) GarfieldCounty'sratesarethelowestamongthe countiesuntil2005,whentheratesrisetobecome higherthanthoseforMontroseCounty(butstill lowerthanforDeltaorMesaCounties.Garfield Countyratesremainstablethrough2007. GarfieldCounty'sratesarethelowestamongthe counties,slightlyincreasingfrom2004to2006. GarfieldCounty'sratesarethelowestamongthe counties,slightlydecreasingfrom2001through 2006 GarfieldCounty'sratesaregenerallythelowest amongthecounties,andremainrelativelystable. GarfieldCounty'sratesaresimilartothoseofDelta andMontroseCountiesformostofthetimeperiod, increasingin2001,thendecreasingthrough2007. GarfieldCounty'sratesarerelativelystableandlower thanthoseoftheothercounties,withtheexception ofbeinghigherthanDeltaCounty'sratesin2002, 2003,and2006. Child(<18years) GarfieldCounty'sratesarethelowestamongthe counties,andarerelativelystableacrossthe8year period.
Cardiac
Ear,Nose&Throat
Endocrine
Eye
GarfieldCounty'sratesarethelowestamongthe counties,andarestableacrossthetimeperiod. GarfieldCounty'sratesarethelowestamongthe counties,andarerelativelystableacrossthe8year timeperiod. GarfieldCounty'srateswerehighlyvariableover GarfieldCounty'srateshaveincreasedduringthe 8yeartimeperiod,butaregenerallylowerthan thoseoftheothercounties GarfieldCounty'sratespeakedin2003,dropping
Gastrointestinal&Urinary
Infection
170
the8yeartimeperiod,aswerethosefortheother counties.
steadilythrough2006,therisingagainin2007.These ratesweregenerallysimilartothoseoftheother counties. GarfieldCounty'srateswerelowerorthesameas ratesfortheothercountiesthroughoutthe8year timeperiod.Therewasaspikeinratesduring2004. GarfieldCounty'sratesarelowandsimilartothoseof theothercounties,withtheexceptionofMontrose County,whichhadaspikeinratesfor2005,thendropping toreturntothelowratesoftheothercountiesby2007.
MentalHealth
Withtheexceptionof2000and2001,Garfield County'srateswerelowerthanthosefortheother counties,anddecreasedsteadilyfrom2001to2007. GarfieldCounty'sratesarethelowestamongthe counties,andarerelativelystableacrossthe8year timeperiod.
Musculoskeletal
Neurological
GarfieldCounty'sratesarethelowestamongthe counties,andarerelativelystableacrossthe8year timeperiod. GarfieldCounty'sratesarethelowestamongthe counties,andarerelativelystableacrossthe8year timeperiod,withamoderateincreasefrom2002to through2004,droppingagainin2005. GarfieldCounty'sratesarethelowestamongthe counties,andarerelativelystableacrossthe8year timeperiod,followinganincreasefrom2000to2001. GarfieldCounty'sratesincreasedsteadilyfrom2000through 2002,remainingstablethrough2005,andthendecreasing steadilythrough2007.Theseratesaregenerallysimilarto thosefortheothercounties,andfollowthesametrends.
Pregnancy/Gynecological
RespiratoryCondition
Skin/Allergy
GarfieldCounty'sratesarethelowestamongthe counties,showingperiodsofmoderateratedecrease (20012003)andincrease(20032005).
171
Table30.OutpatientHospitalVisitsorLabWork
RMHPMemberData:20002007* DRGCategory Accident,Injury&Trauma Adult(18years) GarfieldCounty'sratesarethelowestamongthe comparisoncountiesuntil2005,whentherewasan anincreaseinrates,makingGarfieldCounty'srateshigher thanthoseofMontroseCounty,butstilllowerthan thoseforDeltaandMesaCounties.Allofthecounties showedadecreaseinratesbetween2006and2007except GarfieldCounty,whoseratesremainedstable. GarfieldCounty'sratesarethelowestamongthefour countiesforannualphysicalusage.NOTE:Thisrateisforan insuredpopulation,forwhichannualphysicalsareprovided. GarfieldCounty'sratesarethelowestamongthefour counties,andweregenerallystablethroughoutthe8year period. GarfieldCounty'sratesarethelowestamongthefour counties,andweregenerallystablethroughoutthe8year period. Withtheexceptionof2001,GarfieldCounty'sratesarethe lowestamongthefourcounties,andaregenerallystable throughoutthe8yearperiod. GarfieldCounty'sratestendtofallinthemiddleoftherates forthefourcomparisoncounties,increasingbetween2000 and2001,thenincreasingthrough2007. GarfieldCounty'sratesincreasedfrom2001through2003, duringwhichtimetheywerehigherthanthoseforDeltaCounty butlowerthanthoseforMesaandMontroseCounties.After 2003,GarfieldCounty'srateswerestablethrough2006, droppingin2007,andarelowerthantheothercountiesfor GarfieldCounty'sratesincreasedfrom2001through2003, belowGarfieldratesfor2006only. Child(<18years) GarfieldCounty'srateswereamongthelowestforthefour counties,remainingessentiallystableoverthe8yearperiod.
AnnualPhysical WellCheckups(child)
GarfieldCounty'srateswerethelowestamongthefour countiesforannualphysicalusage.NOTE:Thisrateisforan insuredpopulation,forwhichwellcheckupsareprovided.
Cardiac
Ear,Nose&Throat
Endocrine
172
Eye
GarfieldCounty'sratesarethelowestamongthefour counties,andweregenerallystablethroughoutthe8year period. GarfieldCounty'sratesarethelowestamongthefour counties,andweregenerallystablethroughoutthe8year period.
GarfieldCounty'sratesarethelowestamongthefour counties,andshowadownwardtrendfrom2005through2007. From2001until2003,GarfieldCounty'sratesarelowerthan thosefortheothercounties.After2003,theratesarehigher andmoresimilartothoseforDeltaandMontroseCounties (MesaCounty'sratesarehigher).Ratesincreasesharplyin 2007forGarfieldandDeltaCounties,decreasingforMesaand MontroseCounties. GarfieldCounty'sratesincreasedsteeplybetween2000and 2001,remaininghigh(relativetotheothercounties)through 2002,thendecreasingthrough2006.In2007,GarfieldCounty's ratewasagainhigherthanthatfortheothercounties. GarfieldCounty'sratesaregenerallylowerthanthoseforthe othercounties.Ratesincreasebetween2000and2001,again between2003and2004,decreasethrough2006,andturn upwardslightlyin2007. WiththeexceptionofsteepincreaseinMontroseCountyrates in2005(followedbyasteepdownwardtrendthrough2007), ratesforallcountiesarelow.
Infection
GarfieldCountyratesaregenerallyinthemiddleofthe ratesforthefourcounties,andarerelativelystableexcept foranincreasein2001andadecreasein2006.
MentalHealth
Withtheexceptionofaspikein2001,GarfieldCounty'srates arelowerthanthosefortheothercounties,andsteadily decreasedthrough2007.
Musculoskeletal
GarfieldCounty'sratesarethelowestamongthefour counties,andweregenerallystablethroughoutthe8year period. GarfieldCounty'sratesarethelowestamongthefour counties,andweregenerallystablethroughoutthe8year period. GarfieldCounty'sratesarethelowestamongthefour counties,andweregenerallystablethroughoutthe8year period.
Neurological
WiththeexceptionofasteepincreaseinratesforMontrose Countyin2007,ratesforallcountiesarelow.
173 RespiratoryCondition Skin/Allergy GarfieldCounty'sratesarethelowestamongthefour counties,andweregenerallystablethroughoutthe8year period. GarfieldCounty'sratesarethelowestamongthefour counties,increasingslightlyfrom2003to2004,then decreasingfrom2005through2007. RatesforGarfieldCountyareamongthehighestforthefour counties(ratesforMontroseCountyaregenerallyhigher). Ratesincreasefrom2000through2002,thendecrease from2006through2007. GarfieldCounty'sratesaregenerallythelowestamongthe fourcounties,increasingfrom2003to2004,then decreasingfrom2005through2007.
174
Table31.AmbulatoryOfficeVisits
RMHPMemberData:20002007* DRGCategory Accident,Injury&Trauma Adult(18years) Until2005,GarfieldCounty'sratesarethelowestamongthe fourcounties.Anincreaseinratesbetween2004and2005puts GarfieldCounty'srateshigherthanthoseforMontroseCounty, butlowerthanDeltaandMesaCounties,for2006through2007. GarfieldCounty'sratesfor2005through2007arestable,while theothercountiesshowdecreasingratesforthesameperiod. GarfieldCounty'sratesarethelowestamongthefour countiesforannualphysicalusage.NOTE:Thisrateisforan insuredpopulation,forwhichannualphysicalsareprovided. GarfieldCounty'sratesarethelowestamongthefour countiesfortheentire8yearperiod,despiteagradualincrease inratesfrom2004through2006. GarfieldCounty'sratesarethelowestamongthefour counties,andweregenerallystablethroughoutthe8year period. Withtheexceptionof2001,whenDeltaCounty'sratesarelower, GarfieldCounty'sratesarethelowestamongthefourcounties forthe8yeartimeperiod. GarfieldCounty'sratesarethelowestamongthefour counties,andweregenerallystablethroughoutthe8year period. GarfieldCounty'sratesarethelowestamongthefour counties,andweregenerallystablethroughoutthe8year GarfieldCounty'sratesgenerallyfallinthemiddleamongthefour counties,increasingbetween2000and2001,thengenerally decreasingfortheremainderofthetimeperiod. GarfieldCounty'sratesarehigherthanthoseofDeltaCountyfor 2000through2003andagainin2006.Otherwise,GarfieldCounty's ratesarethelowestamongthefourcounties. GarfieldCounty'sratesarethelowestamongthefour counties,withdecreasingtrendsfrom2005through2007. Child(<18years) GarfieldCounty'sratesgoupanddownduringthe8yearperiod, butarestillgenerallylowerthanthosefortheotherfourcounties.
AnnualPhysical WellCheckups(child)
GarfieldCounty'srateswerethelowestamongthefour countiesforannualphysicalusage.NOTE:Thisrateisforan insuredpopulation,forwhichwellcheckupsareprovided.
Cardiac
Ear,Nose&Throat
Endocrine
Eye
GarfieldCounty'sratesaregenerallylowerorsimilartothoseof theothercounties,increasingbetween2003and2004,and
175 period. Infection GarfieldCounty'sratesarehighlyvariableoverthe8yeartime period,remaininginthemiddleofthosefortheothercounties foreveryyearexcept2006.In2006,GarfieldCountiesratesare lowerthanthosefortheothercounties. againbetween2006and2007. GarfieldCounty'sratesincreasedsharplybetween2000and2001, makingthemthehighestamongthecountiesfor2001and2002. Between2002and2006,ratesdecreasedcontinuously, puttingGarfieldCountyinthemiddleoftheothercounties.In 2007,ratesagainincreased,makingGarfieldCountythehighest amongthefourcounties. GarfieldCountyratesincreasedbetween2000and2001andagain in2004,thendecreasedbetween2004and2006,increasing slightlyin2007.Overthe8yeartimeperiod,GarfieldCounty ratesareeitherlowerthanorsimilartotheratesfortheother counties.MesaCountyhasthehighestratesfortheentireperiod. WiththeexceptionofMontroseCounty'sratesfor2005through 2006,ratesforallcountiesarelowandessentiallysimilar. MontroseCounty'sratesrosesteeplybetween2004and2005, thendecreasedthrough2007.
MentalHealth
GarfieldCountyratesincreasedbetween2000and2001,then decreasedsteadilythrough2007.After2002,GarfieldCounty's ratesarethelowestamongthefourcounties.
Musculoskeletal
GarfieldCounty'sratesarethelowestamongthefour counties,andweregenerallystablethroughoutthe8year period. GarfieldCounty'sratesarethelowestamongthefour counties,andweregenerallystablethroughoutthe8year period. GarfieldCounty'sratesarethelowestamongthefour counties,increasinggraduallythrough2004,decreasingfrom 2004through2005,andincreasingagainin2007. GarfieldCounty'sratesarethelowestamongthefour counties,andweregenerallystablethroughoutthe8year period,exceptforanincreasebetween2000and2001.
Neurological
WiththeexceptionofaspikeinMontroseCounty'sratesfor2007, allcountieshavelowratesoverthe8yeartimeperiod.
RespiratoryCondition
GarfieldCounty'sratesincreasefrom2000to2002,remainstable through2005,thendecreasesteadilythrough2007.Theserates aregenerallyinthemiddleoftheratesforthefourcounties.
176
Skin/Allergy
GarfieldCounty'sratesarethelowestamongthefour counties,andweregenerallystablethroughoutthe8year period,exceptforanincreasebetween2003and2004, decreasingtopreviousratesin2006.
GarfieldCounty'sratesarethelowestamongthefourcounties exceptfor2001,whenDeltaCounty'srateswerelower.Garfield County'sratesincreasedbetween2003and2004,decreasing steadilyfrom2005through2007.
177 Careflight Data One other measure of traumatic injury rates (and potentially of serious work-related injuries), is statistical information from St. Marys Hospital Careflight emergency flight ambulance service. Complete data sets are available in Appendix S. Figure 91 shows emergency service for residents of the four counties for all types of emergencies. Figure 92 shows emergency calls to drill rigs in two counties, Garfield and Rio Blanco, both of which have seen a significant increase in natural gas activity over the past four years. These numbers give a comparison of worker-related, serious injuries between the two counties.
Figure91.CareFlights:RatesbyCounty
Figure92.CareFlights:RatesofDrillRigFlights
178
HouseholdSurvey.ThefullACCESSdatabasecontainingresponsestothehouseholdsurvey,EpiInfoanalysisofthesurveyresponsesand ExcelspreadsheetsusedtocollateandchartthedataareprovidedinAppendixU.Thefollowingdiscussioncontainsthehighlightsandmost significantresultsfromthisphaseofthestudy.Inmostcases,thegraphicresultsareportrayedbyzipcodeareaandforGarfieldCounty, overall. Please note that SL in all of the graphs and tables refers to households in which Spanish is the primary language. These interviewsandsurveyswereconductedinSpanish. RespondentDemographics Table 32 provides an overview of the sample population, numbers of surveys attempted and completed, and the percentage of the population surveyed, by zip code area. Table 33 provides information on response rates and reasons for refusing to participate in the survey. Table32.HouseholdSurveyDemographics Households %of %of Zip Census Surveys Individuals (withland Sample households population Code pop. Completed surveyed** lines) completed surveyed 81601* 12,768 4,680 Mail 440 49 1.05 135 1.06 Phone 438 56 1.20 151 1.18 81623 13,008 4,784 Mail 362 26 0.54 63 0.48 Phone 355 48 1.00 140 1.08 81635 5,041 1,947 Mail 186 24 1.23 52 1.03 Phone 186 22 1.13 57 1.13 81647 4,410 1,672 Mail 170 11 0.66 25 0.57 Phone 170 23 1.38 84 1.90 81650 10,319 3,752 Mail 401 35 0.93 87 0.84 Phone 401 50 1.33 147 1.42 81652 3,107 1,632 Mail 172 21 1.29 54 1.74 Phone 169 20 1.23 53 1.71 Totals 48,653 18,467 3,450 385 2.08 1,048 2.15 *ThecityofGlenwoodSpringsincludeszipcodes81601and81602.Forthesakeofsimplicity,inthisreport,thedataforthesetwo zipcodesarecombinedandreferredtocollectivelyaszipcode81601. **Thisnumberincludesthenumberoftotalindividuals(householdmembers)forwhomdatawerecollectedasaresultofcompleting thedesignatednumberofhouseholdsurveys.
179
Table33.HouseholdSurveyResponseRateTelephoneSurvey Reasongivenfornotparticipating 16 17 3 4 16 4 1 7 1 1 5 3 1 0 0 0 3 0 5 12 3 1 3 3 2 6 4 6 5 6
Total Carbondale(81623) GlenwoodSprings (81601&81602) NewCastle(81647) Parachute(81635) Rifle(81650) Silt(81652) 25(34.2%)* 42(42.8%) 11(32.4%) 12(35.3%) 37(43.5%) 16(44.4%)
Gender**
Female
Age**
Young Middleaged Elderly
RespondentLocation NumberRefused HungUp Toobusy Surveytoolong Notinterested Other Male
8 7 0 2 7 3
7 16 5 2 8 5
6 4 1 1 5 2
6 5 4 1 4 5
1 3 1 0 2 0
BattlementMesa (81635) 1(4.3%) 1 *Percentageofactualtelephonecontacts(doesnotincludeattemptedcallstonumbersnolongerinserviceorhouseholdcontactsthat wereunsuccessfulasaresultofrepeatedcallstoaphonenumberthatresultedinnoansweroransweringmachineresponseonly.With theexceptionofBattlementMesa(whichislargelyaretirementcommunity),therearenosignificanttrendsamongimpactedandnon impactedcommunitieswithrespecttorefusaltoparticipateinthetelephonesurvey. **Itwasoftennotpossibletodefinitivelyidentifythegenderandageofthepersonwhoansweredthephoneandrefusedtocompletethe survey.Thus,thesenumbersaregivenforroughinformationonly,andarenotintendedtoprovideareliabledescriptionofthepopulation thatrefusedtoparticipateinthissurvey.
180
Figure93demonstratesgraphicallythatthequalityofthevastmajorityofcompleted householdsurveyswasgood,withaminimalnumberthatweremarginal,withrespecttothe reliabilityofthecollecteddata,orhadsignificantlevelsofmissingdata. Figure93.QualityofHouseholdSurveyInterviews Figure94showsthelengthofresidenceinGarfieldCountyforsurveyrespondenthouseholds. ThevastmajorityofrespondentsfromeveryzipcodeareahavelivedinGarfieldCountyfor morethanfiveyears,andmorethan90%havelivedintheircurrentresidenceformorethan oneyear(Figure95).Figures96and97showthelocationofrespondentresidencesandthe numberofpersonsresidingintherespondenthouseholdsasapercentageofthehouseholds surveyed.Figures98through103provideinformationonotherdemographiccharacteristicsof thesurveyedhouseholdsandrespondents,suchasage,gender,ethnicity,percentageof ethnicity,educationandhealthinsurance.Themajorityoftherespondentsweremarried.
181
Figure94.LengthofresidenceinGarfieldCounty
Figure95.Lengthoftimeincurrentresidence
182
Figure96.LocationofresidenceinGarfieldCounty
Figure97.HouseholdSize(residentsperhousehold)
183
Figure98.HouseholdEthnicity Figure99.MeanAgeof HouseholdResidents Figure100.GenderofResidents

184
Figure101.PercentageofChildren inRespondentHouseholds Figure102.Educationlevel ofSurveyRespondent Figure103.Percentageof HouseholdswithHealth Insurance
185
HouseholdSurveyOutcomes Greaterthan80%ofindividualsfromeveryzipcodeareainGarfieldCountyratedtheircurrent healthaseitherexcellentorgood,andlessthan10%ofindividualsineveryzipcodeareafelt thattheircurrenthealthissomewhatworseormuchworsethanitwasoneyearago. Figure104.CurrentHealth Figure105.HealthOneYearAgo

186
Approximately12%ofindividuals,countywide,reportedthattheyhadsufferedanillnessor injuryduringthepastyearthathadaffectedtheirhealthforgreaterthan5days. Figure106.Percentageofindividualsreportingselecteddiseasesandsymptoms: GarfieldCountyOverall
187
Approximately8%ofindividuals,countywide,reportedsufferingfromdepression. 20%ofindividualsreportedsufferingfromavarietyofallergies,includinghayfever. 8% of individuals suffer from frequent headaches or migraines; a zip code comparison shows that the lowest frequency of headache sufferers live in zip code 81623 (4%), whilethehighestfrequencywasreportedfromzipcodes81635and81647(8.3%each). PleasenotethatthesedataareNOTageadjusted.
Figure107.Percentageofindividualssufferingfromfrequentheadachesormigraines byzipcode Wemadesomeattempttolookatanycorrelationwithhomedrinkingwatersupplyand neurological symptoms or complaints such as frequent headaches (see Figure 107 below), dizziness, twitching, and weakness, and lung or kidney diseases For a completesetofsymptomsorconditionsthatwerecorrelatedwithhomewatersupply, pleaseseeAppendixU.Thoserespondentswholiveinthe81647,81650and81652zip code areas were more likely to be using bottled or vended water as a home drinking watersource.Inallcases,thenumbersofindividualsreportingtheseconditionswere toosmalltoshowstatisticalsignificanceforanycorrelation.However,themajorityof respondentsinallzipcodeareasusedfilteredorunfilteredtapwater(municipalwater supplies) as their primary drinking water source. Despite the fact that municipal drinking water sources are required to meet Federal drinking water standards, many respondents living in the 81647, 81650 and 81652 reported concerns about a

188
relationship between their health conditions and their drinking water source (Figure 110). Please see our recommendation regarding testing for private wells and small homeownerassociationdrinkingwatersources.
Figure108.Mainwatersupplyandnumberofindividualsreportingfrequentheadaches Figure109.Primarysourceofdrinkingwaterathome
189
Figure110.PercentageofIndividualsConcernedaboutHealthProblems RelatedtoWaterSupply Wealsoaskedquestionsaboutpersonalbehaviorsthatcouldinfluencerespondenthealth;e.g., alcohol,tobacco,andrecreationaldruguse.Theseresponseswerestratifiedbyzipcodeand age.(PleaseseeAppendixUforcompletedatasets.)Figures110through112providedataon frequencyandamountofalcoholandrecreationaldrugusereportedfortheCounty,overall. Figure111.DaysPerMonthofAlcoholConsumptionbyAge:CountyOverall

190
Figure112.NumberofDrinksPerDayforIndividualsWhoConsumedAlcohol DuringthePast30Days:CountyOverallbyAge
Figure113. Percentage of Household Members Who Reported Using Recreational Drugs Duringthe Previous30 DaysbyZip Code
191
Figures114through116provideinformationoncigarettesmokingwithinGarfieldCountybyzip code,agegroupandfrequency. Figure114.PercentageofIndividualsWhoHaveSmokedMorethan 100CigarettesbyZipCode Figure115.PercentageofIndividualsWhoHaveSmokedMoreThan100Cigarettes byAgeGroup:CountyOverall

192
Figure116.FrequencyofSmokingforindividualsWhoHaveSmokedMoreThan100 CigarettesinTheirLives:byZipCode Becauserespiratorycomplaintswereexpressedsofrequentlyininterviewsandfocusgroups, weaskedanumberofspecificquestionsaboutrespiratoryconditionsandcontributingfactors suchassmoking.Figures117through122showcorrelationsbyzipcode,betweenself reportedsmokingfrequencyandselfreportedrespiratoryconditions.Figures123through126, andtheaccompanyingdiscussions,provideadditionalinformationonselfreportedrespiratory conditionsbyzipcode.
193
Figure 117. Smoking Frequency of Individuals Who Have Respiratory Conditions:ZipCode81601 Figure 118. Smoking Frequency of Individuals Who Have Respiratory Conditions:ZipCode81623 Figure 119. Smoking Frequency of Individuals Who Have Respiratory Conditions:ZipCode81635
194
Figure120.SmokingFrequencyof Individuals Who Have Respiratory Conditions:ZipCode81647 Figure 121. Smoking Frequency of Individuals Who Have Respiratory Conditions:ZipCode81650 Figure 122. Smoking Frequency of Individuals Who Have Respiratory Conditions:ZipCode81652
195
6.5%ofindividuals,countywide,reportedhavingadiagnosisofasthma;thehighest frequencyofindividualswithasthmawasinzipcodearea81647(8.3%),whilethe lowestfrequencyofindividualswithasthmawasinzipcodearea81652(4.7%). Bothzipcodeareashavesignificantnaturalgasindustryactivity. Figure123.PercentageofHouseholdMemberswithAsthmabyZipCode Similartowhatwasobservedforasthma,thezipcodeareahavingthehighest frequencyofotherrespiratoryconditions,suchasChronicObstructive PulmonaryDisease(COPD),emphysemaandotherlungorbreathingproblems, wasamongthosemosthighlyimpactedbynaturalgasindustryactivity,butthe lowestornexttolowestfrequencywasalsofoundamongtheseimpactedzip codeareas(Figures125through127). Ageandsmokingarefactorsthatclearlyinfluencetheincidenceofthese conditions.Withtheexceptionofasthma,individualswhoreportedhavinglung conditionstendedtobeolder(e.g.,65+forCOPDandemphysema).Thesedata arenotshownhere,butareavailableinAppendixU.27%ofcountyresidents, overall,reportedhavingsmokedatleast100cigarettesduringtheirlifetime;60% ofthesehavequitsmoking(Figure116above).85%ofrespondentsreported thatsmokingisNOTallowedwithinthehome
196
Figure124. Percentageof HouseholdMembers withCOPDbyZip Code Figure125. Percentageof Household Memberswith Emphysemaby ZipCode
197
Figure126.PercentageofHouseholdMemberswithOtherLungProblemsbyZipCode 3.8% of individuals report living with diabetes and its side effects such as kidney problems, loss of feeling or pain in hands and feet, and eye problems. There was no statistically significant difference among the zip code areas for frequency of diabetes. Completedata,includingagebreakdownforindividualswithdiabetes,maybefoundin AppendixU.Figure127providessurveydataontheprevalenceofdiabetesbyzipcode. Figure127.PercentageofIndividualswithDiabetesbyZipCode

198
Inordertoaddresscausesofphysicalsymptoms,suchaslossofperipheralnervefunction,pain inhandsandfeet,eyeproblems,and/orkidneyproblems,thatcouldberelatedtoexposureto emissions from natural gas industry activities, we correlated the selfreported occurrence of these symptoms with selfreported diabetes among household survey respondents. All of thesesymptomsarepossiblecomplicationsofhavingdiabetes(Figures128through131).
Figure128.DiabetesComplications:PercentageofIndividualswithDiabetesAND PaininHandsorFeet Figure129. Diabetes Complications: Percentageof Individualswith DiabetesAND LossofFeelingin HandsorFeet, CountyOverall (Comparisonwith LossofFeeling inNonDiabetic, Survey Population)
199
Figure130.DiabetesComplications:PercentageofIndividualswithDiabetesAND KidneyProblems/ProteininUrine,CountyOverall (ComparisonwithKidneyProblemsinNonDiabetic,SurveyPopulation) Figure131.DiabetesComplications:PercentageofIndividuals withDiabetesANDEyeProblems

200
24%ofindividualshavecoronarydisease,acategorythatincludesheartattackorheart surgery,highbloodpressure,strokeandangina.Therewerenostatisticallysignificant differencesamongthezipcodeareasforfrequencyofcoronarydisease.(Figures132 and133.CompletedataareavailableinAppendixU.)
Figure132.PercentageofIndividualswithCoronaryDiseasebyZipCode Figure133. Percentageof Individualswith CoronaryDisease byAgeGroup: CountyOverall
201
Oneseriesofquestionsinthehouseholdsurveydealtwithreproductiveissues.Figure 134providesdataontherelativenumberofpregnanciesperadultfemalehousehold membersbyzipcode(womenlivinginGarfieldCounty1year,and18and54years ofage). Figure134.NumberofPregnanciesPerWomanforWomenWhoBecamePregnant Figure135. Percentage of Pregnancies that Resultedin Miscarriages byZipCode

202
6% or less of children born in any zip code area of Garfield County had birth defects (as defined by the survey respondent). 10% or less of children born in any zip code area of GarfieldCountywerereportedtohavedevelopedhealthordevelopmentalproblemswithin 5yearsoftheirbirth.(Figures136and137;completedataareavailableinAppendixU.) Figure136. Percentageof ChildrenBorn WithBirth Defects(Self Defined)byZip Code Figure137. Percentageof Children DevelopingHealth orDevelopmental ProblemsWithin5 YearsofBirthby ZipCode
203
5%ofindividuals,countywide,reportedhavingsomekindofcancerduringtheirlifetime. These data are summarized in Figures 138 through 140. The complete data are availableinAppendixU. ~2Xasmanyindividualsresidinginzipcodearea81635reportedhavingcancer thanwasreportedforthecountyoverall.Itisimportanttonote,however,that the average age of the respondents from this zip code area was considerably olderthanfortheotherzipcodeareas. 53.3%ofthereportedcancerswerediagnosedinindividualswhowere55years orolder;TherewereNOcancersreportedinindividualsunderage25. Themostfrequentlyreportedcancerswerefemalebreastcancer(20.7%),non melanomaskincancers(26.4%),prostatecancer(15.1%),cervicalcancer(9.4%), andcoloncancer(7.5%).Malignantmelanomaandlymphomaeachaccounted for 3.8% of the reported cancers. Uterine, thyroid, liver, kidney, and bladder cancers, along with leukemia, glandular carcinoma, made up the remainder of thecancersreported(1.9%each). Figure138. Percentageof Individuals Reporting CancerbyZip Code

204
Figure139. Percentageof reported CancersbyAge: CountyOverall Figure140.ReportedCancerTypes:CountyOverall
205
OccupationandDisease Responses to questions about occupational history (current and longest job titles and industry affiliations) allowed correlations with diseases and symptoms reported by survey respondents. Figures 140 through 149 provide a graphical description of the longtermoccupationsandcurrentjobcategoriesforhouseholdsurveyrespondentsand household members. There was broad representation of types of jobs, both in the whitecollar and bluecollar occupations, giving us confidence that no one occupationalorjobdemographicwasoverorunderrepresentedinthissurvey.Some individualsdidrefusetoreporteithertheirjobortheindustryforwhichtheywork.Jobs and industries were categorized using standard National Institute for Occupational HealthandSafety(NIOSH)categories.
Figure141.CurrentOccupation:CountyOverall
206
Figure142.CurrentOccupationbyZipCode Figure143.CurrentOccupationbyZipCode:PrimarilySpanishSpeakingHouseholds
207
Figure144.CurrentIndustryEmploymentbyZipCode
Figure145.CurrentIndustryEmploymentbyZipCode: PrimarilySpanishSpeakingHouseholds
208
Figure146.LongestOccupationbyZipCode Figure147.LongestOccupationbyZipCode:PrimarilySpanishSpeakingHouseholds
209
Figure148.LongestIndustryEmploymentbyZipCode Figure149.LongestIndustryEmploymentbyZipCode: PrimarilySpanishSpeakingHouseholds
210
Figure150.PercentageofIndividualsWhoDidntKnoworRefusedtoAnswer OccupationorIndustryEmploymentQuestionsbyZipCode Itisimportanttonotethatthisstudywasnotdesignedasanoccupationalhealthstudy, andthus,noconclusionsmaybedrawnregardingoccupationalexposuresanddisease outcomes. The numbers of individuals within any occupational category that report havingaparticulardiseaseorconditionaretoolowforstatisticalsignificance.However, thefollowingobservationsmaybemade: o Individualswhoreportedthattheircurrentand/orlongestoccupationwasinthe professionalandrelatedservicesindustries(e.g.,healthcareproviders,attorneys, etc.), personal services occupations (e.g., housekeepers, hair stylists, etc.), construction industries or transportation (including truck drivers), communicationsandpublicutilitiesindustriesweremostlikelytohavereported having respiratory conditions; neurological symptoms such as dizziness, numbness,weakness;skinproblems;andfrequentheadaches/migraines. o Thoseindividualswhorefusedtoanswerquestionsabouttheiroccupationand/or industry affiliation were most likely to have reported having frequent headaches/migraines; neurological symptoms such as dizziness, numbness, weakness; anemia; seizures; skin problems; and cancer (but no bladder, kidney, liver,lymphomaorthyroidcancersorleukemia). o Figures151through168provideasummaryofthesedata.PleasenotethattheY axisscalesdifferamongthegraphs.
211 Figure151. Percentageof IndividualsEmployed intheAgricultural IndustryReporting SelectedDiseasesand Symptoms:County Overall Figure152. Percentageof Individuals Employedin theMining Industry Reporting Selected Diseasesand Symptoms: CountyOverall
212
Figure153. Percentageof IndividualsEmployed intheManufacturing IndustryReporting SelectedDiseasesand Symptoms:County Overall Figure154. Percentageof Individuals Employedin theWholesale TradeIndustry Reporting Selected Diseasesand Symptoms: CountyOverall
213 Figure155. Percentageof Individuals Employedinthe Finance,Insurance orRealEstate IndustryReporting SelectedDiseases andSymptoms: CountyOverall Figure156. Percentageof Individuals Employedin thePersonal Services Industry Reporting Selected Diseasesand Symptoms: County Overall
214
Figure157. Percentageof Individuals Employedinthe Professionaland RelatedServices IndustriesReporting SelectedDiseases andSymptoms: CountyOverall Figure158. Percentage of Individuals Employedin theForestry orFisheries Industries Reporting Selected Diseases and Symptoms: County Overall
215 Figure159. Percentageof Individuals Employedinthe Construction IndustryReporting SelectedDiseases andSymptoms: CountyOverall Figure160. Percentageof Individuals Employedinthe Transportation, Communications orPublic Utilities Industries Reporting Selected Diseasesand Symptoms: CountyOverall
216
Figure161. Percentageof IndividualsEmployed intheRetailTrade IndustryReporting SelectedDiseases andSymptoms: CountyOverall Figure162. Percentageof Individuals Employedin theBusiness andRepair Services Industries Reporting Selected Diseasesand Symptoms: CountyOverall
217 Figure163. Percentageof Individuals Employedinthe Recreationand Entertainment IndustriesReporting SelectedDiseases andSymptoms: CountyOverall Figure164. Percentage of Individuals Whose Industry Employment CouldNotBe Classified Reporting Selected Diseasesand Symptoms: County Overall
218
Figure165. Percentageof Currently Unemployed Individuals ReportingSelected Diseasesand Symptoms:County Overall Figure166. Percentage ofCurrently Disabled Individuals Reporting Selected Diseases and Symptoms: County Overall
219 Figure167. Percentageof Currently Retired Individuals Reporting Selected Diseasesand Symptoms: CountyOverall Figure168.Percentage ofAdultsWhoDidnt KnoworRefusedto ProvideanIndustry AffiliationReporting SelectedDiseasesand Symptoms:County Overall
220
RelationshipsBetweenHealthandEnvironmentalExposures:HouseholdMemberConcern Aseriesofquestionsregardingperceptionsofriskrelatedtohomeandoutsideenvironmental exposures and their relationship to health outcomes were asked at the end of the survey. These questions were intended to serve as measures of concern and perceptions among a randomly selected population within Garfield County (as opposed to the more selfselected populationthatprovidedcommentsduringfocusgroups,interviews,andpublicmeetings),and to provide some measure of the potential bias with which survey respondents might have respondedtoquestionsabouttheirhealth. When asked whether or not they are concerned that their home drinking water source was related to any of their health problems, between 5.5 and 17% of individuals who liveinareaswithhighnaturalgasindustryactivity(zipcodeareas81635,81647,81650, and81652)respondedthattheyareconcerned,whileonly35%ofindividualswholive in the areas least impacted by natural gas industry activity (zip code areas 81601 and 81623)respondedinthesamemanner(Figure169). Figure169.PercentageofIndividualsConcernedAboutHealthProblems RelatedtoHomeDrinkingWaterSupplybyZipCode When asked whether or not they are concerned that their health problems may be relatedtochemicalsinorneartheirhomes,between6and16%ofindividualswholive inareaswithhighnaturalgasindustryactivity(zipcodeareas81635,81647, 81650,and 81652)respondedthattheyareconcerned;between1.4and7%ofindividualswholive in the areas least impacted by natural gas industry activity (zip code areas 81601 and 81623)respondedinthesamemanner(Figure170).Therewasnocorrelationbetween levelofeducationand concernaboutarelationshipbetweenchemicalsinornearthe homeandhealthproblems(Figure171).
221
Figure170.PercentageofHouseholdsConcernedthatHealthProblemsMayBeRelatedto ChemicalsInorNeartheHomebyZipCode Figure171. Concernthat ChemicalsInor NeartheHome MayBerelated toHealth Problems: CorrelationWith Education.
222
Whenaskediftheyareconcernedthateitherenvironmentalorchemicalhazardsintheir neighborhoodsmayberelatedtohealthproblems(Figure172), o Between24and38%ofindividualsresidinginzipcodeareas81635,81647,81650, and 81652 responded that they are not worried at all. Between 76 and 62% of individualsinthesezipcodeareasrespondedthattheyarealittleworried,very muchworried,ordontknow/notsure. o Between 56 and 43% of individuals residing in zip code areas 81601 and 81623 respondedthattheyarenotworriedatall.Between44and57%ofindividualsin these zip code areas responded that they are a little worried, very much worried,ordontknow/notsure.
Figure172.PercentageofHouseholdsConcernedthatHealthProblemsMaybe RelatedtoEnvironmentalorChemicalHazardsinTheirNeighborhoodsbyZipCode
223
o Countywide,individualswhohaveahighschooleducationorlessareslightlyless worried about the relationship between their health and environmental or chemicalhazardsintheirneighborhoods(Figure173). Figure173.ConcernthatEnvironmentalorChemicalHazardsintheNeighborhood MayBeRelatedtoHealthProblems:CorrelationwithEducation o Individualswithhigherincomes(>$100,000peryear)tendedtobelessworriedabouta relationship between environmental or chemical hazards in their neighborhoods and health problems. Those who refused to report an income level tended to express a little concern more often that individuals who reported an income at any level. Respondents in those households with reported incomes of between $25,000 and $50,000weremostlikelytoreportthattheyareveryconcernedaboutenvironmental orchemicalhazardsandanimpactonhumanhealth(Figure174).

224
Figure174.ConcernthatEnvironmentalorChemicalHazardsMayBeRelatedto HealthProblems:CorrelationwithHouseholdIncome When asked specifically whether they are concerned that natural gas industry activities mayberelatedtohealthproblems(Figure175), o Between69and92%ofindividualsresidinginzipcodeareas81601and81623 respondedthattheyarenotconcern.Between8and31%ofindividualsinthese zip code areas responded either that they are concerned or that they dont knoworarenotsure. o 90% of individuals residing in zip code areas 81601 and 81623 responded that theyarenotworriedatall.10%ofindividualsinthesezipcodeareasresponded that they either that they are concerned or that they dont know or are not sure. o There was essentially no difference related to education between individuals who responded that they are concerned about healthrelated impacts of the natural gas industry and those who are not concerned or are not sure (Figure 176).
225
Figure175.PercentageofHouseholdsConcernedthattheNaturalGasIndustryMay BeRelatedtoHealthProblemsbyZipCode Figure176. Concernthatthe NaturalGas IndustryMayBe Relatedto Health Problems: Correlationwith Education
226
STUDY CONCLUSIONS AND RECOMMENDATIONS

Based on the data available to us from state, hospital association, and healthcare provider sources,thehealthofpeopleinGarfieldCountyisnotdifferentfromthehealthofresidentsof otherWesternSlopecounties.However,citizenperception,asevidencedbyfocusgroupand interviewdataandthehouseholdsurvey,indicatesthatatleastsomeindividualsinimpacted areasofGarfieldCountybelieveotherwise.Citizensexpressconcernandexperiencefrustration and stress over activities that they fear could have negative impacts on their health and environment.Suchstress,frustrationandthefearoftheunknowncan,ofitself,leadtohealth problems. Paul Slovics research (Slovic P. Science 1987;236:280285) on risk perception documentsthatindividualstendtofearandreactmostnegativelytosituationsforwhichthere areconsiderableunknownsregardingtheoutcomes,bothcurrentandfuture,andoverwhich theyfeeltheyhavelittlecontrol.EnergyrelatedactivitiesinWesternColoradopresentsucha situationformanyindividuals. Asdescribedinthisreport,therearenumerousgapsanduncertaintiesinourunderstandingof pollution from natural gas operations in Garfield County. The motivation for the following recommendations is to reduce these uncertainties and fill data gaps. Better information is neededinordertomakeacompleteandaccurateevaluationofthreatstohumanhealth. RECOMMENDATIONS. Establish a medical monitoring system especially through primary care networks to identifyanychangesinbaselinedataortrendsand/oranomaliesinmedicalpractices. Conductathoroughstudyofairemissionsduringdrilling,includingenoughsitestocover therangeofdrillingapproaches. o Collect 24hour samples daily around the perimeter of the drill pad to achieve continuousmonitoringduringseveralcyclesofwellinstallation. o Monitormeteorologicalconditions. Identifythecomponentsofhydraulicfracturingfluids. o Wouldallowopenevaluationofdegreeofthreat. o Wouldimprovepublicacceptanceofnaturalgasoperations. Inspectsurfacesoilsatcompletionofdrillingoperations. o Minimizepossibleexposureoflandownerstoresidualsoilcontamination. o Sampleandanalyzeareassuspectedtobecontaminated. o Cleanupareasexceedingactionlevels.
227
Usegreencompletionsandapplicablebestmanagementpractices,includinglocating drilling and production facility operations far enough from public buildings and residencestoreducetheriskofexposuretoairtoxics,suchasbenzene,toluene,and xylenes. Establishamonitoringprogramforprivatewells o Provides the most direct way to assess contamination of drinking water resources o Analyze for methane, benzene and other volatile organic compounds, and selected components of hydraulic fracturing fluids having the greatest potentialtoaffecthumanhealth STUDYLIMITATIONS Inadditiontolimitationsanddatagapsdiscussedinthebodyofthisreport,thefollowingstudy limitationsareworthpointingout: RARE CANCERS Cancers that occur very rarely within a population cannot be identified through the states Tumor Registry, because of requirement to protect individual identity. Thus, the occurrence of rare cancers in persons living in Garfield County, whether or not those cancers could have some relationship to environmental exposures,werenotlikelytohavebeenpickedupthroughthisstudy.Thoseincidences couldbeidentifiedthroughacountybasedmedicalmonitoringsystem. PRIMARYCAREDATAAsnotedearlier,wewerenotabletoacquireprimarycaredata directly,despiteanattempttosurveylocalprimarycarepractices.Primarycarepractice dataarethemostreliablesourceofinformationonunusualoccurrencesofdiseasesor symptoms, especially if the data from all local practices can be combined for analysis. This requires, however, that local care providers are networked electronically or have some other means of sharing patient data in a deidentified (to protect individual identities),butnearlyrealtimemanner. AVAILABLE DATA SETS Because the health data used for this study (other than the HouseholdSurveydata)werenotcollectedspecificallyforresearchpurposes,theyare limitedinscope,havesignificantgaps,andmaynotbecomparablewithotherdatasets. Thisisacommonlimitationforstudiesofthistype.However,thedatasetsusedforthis studyareoftentheonlydataavailable,andthus,areoftenusedtodescribecommunity healthwiththequalificationsnoted.

Garfield Co

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Garfield Co

Uploaded by

Copyright:

Available Formats

COMMUNITY HEALTH RISK ANALYSIS

OIL AND GAS INDUSTRY IMPACTS

Teresa Coons, Ph.D. Saccomanno Research Institute

Russell Walker, Ph.D. Mesa State College

LIST OF TABLES AND FIGURES

169 171 174

PAGE 94 95 96 97 98 99 100 101 102 103 109

PAGE 194 195 196 196 197 197 198

PAGE 176.Concernthatthenaturalgasindustrymayberelatedtohealthproblems 225 correlationwitheducation

Garfield County % 2.6 18.7 79.2 16 41.8 12.6

MesaCounty 95% % CI* 5.4 20.7 80.1 10.6 57.3 20.9

Colorado 95% % CI* 4.6 19.9 84.4 17.2 60.7 14.1

Birthing,Pregnancy, Reproductive,and Neonatal

GarfieldCounty'srates arethelowestamong thefourcounties. Note:TheseareRMHP dataforinsured individuals! Sameasforinpatient visits.

Datashowan increaseinbirthing andpregnancy relatedpatientvisits for2005through 2006forSilt, Parachute/Battlement MesaandRifle residents.

CHAdata (20001stQ2007) GarfieldCounty's ratesarethelowest amongthefour counties.

InpatientHospital GarfieldCounty's ratesarethelowest amongthefour countiesforboth adultsandchildren. Ratesshowan increasingtrendfrom 20032006.

RMHP(20002007) Outpatient/Ambulatory EmergencyRoom Sameasforinpatient Sameasforinpatient visits. visits.

GRHDER (20042006) RatesforSilt, Parachute/Battlement MesaandRifle residentsshowan increasingtrendfor thetimeperiod.

ValleyViewER* (20042006) AdultERvisitsfor thiscategorywere stableoverthe period;child/young adultvisits decreased.

Ratesareconsistent withtheCHAdatafor bothadultsand children.

RatesforSilt, Parachute/Battlement MesaandRifle residentsshowan increasingtrendfor thetimeperiod.

InpatientHospital GarfieldCounty's ratesarethelowest amongthefour countiesforboth adultsandchildren.

GarfieldCounty's ratesareeitherthe lowestamongthe fourcounties,orare similartothoseof theothercounties forthisgroupof disorders.

Ratesareconsistent withtheCHAdata forbothadultsand children.

Ratesareconsistent withtheCHAdatafor bothadultsand children.

GarfieldCountyrates aresimilartothose fortheother counties.

GarfieldCounty outpatient/ambulatory visitsforadultswere thelowestamongthe fourcounties,but highestamongthefour countiesforchildren duringthetimeperiod 2000through2002.

Ratesforinfection relatedERvisitswere variableoverthetime period,without consistenttrendsfor thecommunitiesof Parachute/Battlement Mesa,SiltandRifle.

InpatientHospital GarfieldCountyrates arelowerthanfor theothercounties andsteadily decreasingoverthe timeperiod.

Ratesareconsistent withtheCHAdata forbothadultsand children. Ratesareconsistent withtheCHAdata forbothadultsand children.

Ratesareconsistent withtheCHAdatafor bothadultsand children.

GarfieldCountyER visitrates,overall, werethelowest amongthefour counties.

RatesforSilt, Parachute/Battlement MesaandRifle residentsshowan increasingtrendfor thetimeperiod.

Outpatient/Ambulatory EmergencyRoom Nodataforthis Nodataforthis category. category.

GarfieldCounty's ratesarethelowest amongthecounties, showingperiodsof moderaterate decrease(2001 2003)andincrease (20032005).

INTRODUCTION TO THE STUDY

Exposure Potential (risk assessment)

Health Assessment (current health status)

Determine the type and magnitude of potential human exposures to toxins

Relationship between exposure and health

Figure 1. Relationship between Exposure PotentialandHealthAssessment

Mesa Montrose County County

Delta W e ld La Cheyenne County County Plata County County

PART 1. RISK ASSESSMENT

Detectedcompounds CAS# 71432 108883 100414 Compound Benzene Toluene Ethylbenzene

13677761 m,pXylenes 2 95476 oXylene

PART 2. HEALTH STUDY

Figure 8. World Health Organization Definition of Human

100,000 Delta Garfield Mesa Montrose

Figure10.PopulationEstimatesbyCounty,20002006. Allfourcountieshaveundergonedecreases,andsubsequentincreases,inpopulationoverthe pastdecade.Currently,allfourcountiesareexperiencingpopulationgrowth. Figure11.AverageAnnualPercentChangeinPopulation,20002006.

Delta Garfield Mesa Montrose COLORADO

0 00-01 01-02 02-03 03-04 04-05 05-06

Percentage of the Total Population (Year 2000 Census Data)

Residence Time in the County Range: 9 months to 20 years

How Did Individual Hear About Study?

Female: 12 (52.2%) Male: 11 (47.8%)

Rifle: 5 (21.7%) Parachute: 10 (42.6%) Silt: 8 (34.8%)

Range: 150 feet to 6.5 miles

Mean: 9.8 years Median: 9.5 years

Mean: 0.93 miles

Median: 0.25 miles

Assault/Legal Intervention 12.4 15.4 28.0 7.3** 27.1