Him ORIGINAL CONTRIBUTION The Spread of the Obesity Epidemic in the United States, 1991-1998 Ali H. Mokdad, Mary K. Se Willian H. Dietz, MD, Barbara A. Bowman, PAD hb ‘Context The increasing prevalence of obesity isa major public health concem, since .. MD, MPH ‘obesity is associated with several chronic diseases. nD ‘Objective To monitor trends in state-specific data and to examine changes in the prevalence of obesity among adults. Design Cross-sectional random-digit telephone survey (Behavioral Risk Factor Sur- Marks, MD, MPH vellance System) of noninstitutionalized adults aged 18 years or older conducted by Mb. MPI the Centers for Disease Control and Prevention and state health departments from 1991 to 1998. LTHOUGH ATTEMPTS TO LOSE Setting States that participated in the Behavioral Rsk Factor Surveillance System. weight are common in the ain Qutcome Measures Body mass index calculated from self-reported weight United States," the preva tnd height ’ . . Jeflrey P. Koplai lence of obesity has in- creased since the 1980s." Such in- Results, The prevalence of obey (ined a body mass index =30 kg in {ressed from 12.0% in 1991 to 17 9% in 1998. A steady increase was obsetved i all creases wll tremendously affect public Stern both sexes, across age groups, races, educational level and occurred regard health since obesity s strongly associ fee of smoking status. The greatest magnitude of mnrease was found in he folowing ated with severlchronicdiseases,such groups: 18-to 29-year-old (71% to 12.1%), those with some college education (10,630 as cardiovascular diseases and diabe- to 17.8%), and those of Hispanic ethnkty (11.6% to 20.8%) The magnitude ofthe tes Recent estimates suggest that obe- increased prevalence varied by region (ranging from 31.9% for mid Atiantic to 67 2% ‘Niy-telated movbidity may account for for South ant the area with the greatest increases and by at ranging fom 113% ae ee for Delaware to 101.8% for Georg, te state withthe greatest increases). Recently published trend daa form Conclustons Obesity continues to increase rapidly inthe United States. To alter this the National Health and Nutstion Ex. tend, strategies and programs for weight maintenance as well as weight reduction amination Surveys (NHANES) show the must become a higher public health priority. percentage of obese persons has in- JAMA 1999:280:1519-1522 re jamacom Greased from 14.5% in the years 1976. Toso tg 22% in 1OBE1O94 Tomant. TREBRESS questionnaire concemsper- tivity were included in the BRESS only sonal behaviors that increase risk for 1 in 1991, 1902, 1994, 1906, and 1908 and tor obesity trends since 1994 and to Cy 'mmore of the 10 leading causes of were tised tocreatea leisure-time physi- present state-specific results, we used data froma large population-based sur- vey to examine changes in the preva- lence of obesity among adults in the United States from 1991 to 1998, death inthe United States cal ectivity score: inactive, iregulaly ac The BRFSS uses a mulistage cluster Wve regular, ot intense, aed regula, in- design based on random digit dialing tense." We used SAS and SUDAAN methods to select a representative statatical software in the analyses and samplefomeschstut'snoninstiution tosccount forthe complex sampling de METHODS alized residents" tachstatcare pooled to produce nation- (more than 100000 participants each ally representative esimates® Purher yea) we have not emphasized sttist- Aletalsabout the BRFSS have been pub- cal testing lished" a - ‘aos latins: Diino ain and Pose Wecalculated body massindex (BMI) {YR Be Modes, Seu, De and Bor Tevet or alder, ieeonlucted bythe velght in kilograms divided by the fan ier ofthe Deca stn Cre fr ‘square of height in meters) based on self- Chronic Disease Prevention and Heath Fromation (Or Centers for Disease Control and Pre- A ‘Mush and Stet he Draco Coe (Suton and state health departments, ported weight and height. Paruic {vforOtae ntl and ventana Co pants werecsiiedasobese fei BML Gap At Reps A Nk was greater than or equal to 30 kg/m." $579 ua ty Ne, Malstop 26, Aleta, CA {Questions on leisure-time physical ac- 3013717 ea shied eo. Data collected from sigh." Because ofthe large sample size We analyzed data from all states that participated in the Behavioral Risk Fac- tor Surveillance System (BRFSS). The BRFSS, a cross-sectional telephone sur- vey of noninstitutionalized adults aged See also Patient Page. (©1900 American Medial Association. All rights reserved, JAMA, October 27, 1993_Vol 282, No. 16 1519THE SPREAD OF OBESITY IN THE UNITED STATES ‘Table 1 Obesty Prevalence m Acute ana Mean Wart by Vea 1991 fo 1908" Year 1001 "902 1998) 1994 1996 "006 1997 | 1998 oe Tela 1200.18) 136017 14a1g 183.020) sea 179/047, ir 117 026) 73.7028) 1461028 156032) 768026 __17.7 025) Women 122024) T3529 142/029 149024) 7631022) 1811023) Tage Toi 73:1 (0.00) T4909 7451009 7500.10) 7551009 762/008) iM 81.50.12) wai) 8.1019 #85 0.19) 228 0.111 eta (0.1) Weren 05.1070) 66.1 (008) B.A LO] 6.9 0.10) are ooy sea ie! South Alantic regions, respee- ‘Table 2. changes in Obesty Prevalence in Adults by Characteristics tively) and by state ranging from 11.3% Te 1098 Diference {in Delaware to 101.8% in Georgia. yrp2H bo In 1901, the level of leisure-time ae hysical activity was 29.78 inactive, tL ds ti irregularly active, 33.2% regular 1211029 50 not intense, and 8.7% regular intense. aaa agen 1998, they were 28.6% inactive, (eat, Sa 25.2% irregularly active, 29.6% regular 3m) 7 not intense, and 13.6% regular intense. waar FSG BE To exclude the possibility that de- Tae aaa 32 ‘mographie dilerences accounted for the variation in obesity prevalence be- vw 166 (018) tween the states, we computed the age-, Tce 5 TBE sex, and race-ajusted prevalence of Sepa FST 20 oheciy. although the adjusted rates Other T9087) 'y 10 ere higher than the unadjusted rates, Ect nas Tigh sco 211088 similar patterns were observed among gna Ta states and overtime, Therefore, we only Sar coo 73 teport the unadjusted estimates Cola Tar Sang se COMMENT tn wro2 so Sanaa aaa a tna Thee dat show iat obesity incensed allage groups, races, educational levels, and smoking statuses. Rarely do chronic We excluded from our tend analy- women and across all sociodemo- conditions such as obesity spread with ses Ostates because they didnotcollect graphic groups (TABLE 2),withthe high- the speed and dispersion characteristic weight and height for 1 or more years: est increaseamong the youngestagesand of a communicable disease epidemic. Arkansas (1992), the DistrictofColum- higher education levels. Among His- However, this rapid trajectory of obe- bia (1005), Kansas (1991), Nevada panic men, the prevalence of obesity in- sity may present both clues to origin and (21991), Rhodelsland (1994), and Wyo- creased from 10.0% n 1991 to 18.3% in measures for efficient attempts to con- ‘ming(1991, 1002, and 1993). However, 1997 and for Hispanic women from tol its spread weincludedthedataforallavailableyears 13.29%t0 23.4%, The prevalence of obe- Since overweight participants in sell fortheseandallotherstatesinourmaps. sity increased steadily rom 1991 10 1998 reported studies tend to underestimate We used the US Bureau of Census inal states (FIGURE) their weight and all participants tend to methodofgroupingstatesintoregions, In 1991, 4 of the 45 participating overestimate their height; true rates of states had obesity ates of 15% orhigher obesity are likely underestimated." RESULTS TABLE 3). By 1998, 37 states had rates Moreover, people without telephones are The prevalence of obesity increased higher than 15%. The magnitude of the _not surveyed throuigh BRFSS, and such from 12.0% in 1991 to 17.9% in 1998 increase varied by region (ranging from individuals are likely to be of lower so- CEABLE 1). Obesity increased in menand — 31.0%-67.2%6inerease in the mid Atlan cloeconomie status, a factor that is as- 1520 JAMA, Oxtober 27,1992 No. Is (©1900 American Medical Association, All rights reserved,sociated with obesity" The net elfect of these limitations is that the preva- lence of obesity reported herein isikely a conservative estimate. In fact, the prevalence of obesity from NHANES I (1988-1904 in which weightand height were measured by health professionals was 22.5% in adults, more than a tied higher than the rates reported in our The BRESS data provide states with unique population-based estimates of sell-eported obesity against which pre- vention efforts may be evaluated. State level population-based estimates of obe- sity should be used to provide each slate with a basis for setting priorities for public health interventions THE SPREAD OF OBESITY IN THE UNITED STATES This rapid increase in obesity in all, segments of the population and re- ‘glons of the country implies that there have been sweeping changes in US s0- ‘eely that are contributing to weight gain by fostering energy intake imbalance. ‘Such changes are unlikely to be due to diminished individual motivation to maintain weight or in genetic or other biological changes in the population, We focused on describing the changes in patierns of obesity instead of its contributory factors, such as al- terations in diet, activity patterns, or other behaviors that affect energy bal- ance. However, our data demonstrate that a major contributor to obesity — physical inactivity—has not changed substantially at the population level between 1991 and 1998, By focusing on the challenge of stopping the obesity epidemic and the profound negative health consequences of obesity, itis important to increase the awareness and involvement of health professionals in dealing with the epidemic. ‘Our data suggest that the develop- ment of strategies and programs for weight maintenance as well as weight reduction must become a higher pri- ority. Public health messages should focus increasingly on balancingenergy intake with physical activity. To con- trol the obesity epidemic, wide range of population groups, including phy- sicians and other health care profes- Figure, Frevalence of Obesity Among US Aduls From Vers 1991, 1993, 1995, and 1998 Date were calclted ng the Behavioral Rik Factor Surveillance Sytem, States ae white nthe years 197 and 193 beau nforationonvweight angele was ot ealected, (©1900 American Medical Association, All rights reserved, ‘Downloaded From: https:/jamanetwork.com/ on 03/3/2023 JAMA October 27, 1993Vol 282, No 16-1524THE SPREAD OF OBESITY IN THE UNITED STATES sionals, public health professionals, leg- ‘Table 3, Changerin Obesty Prevalence n Acute by Regn ane site" es » Ree {slators, communities, work sites, and —— mt _Biferonce te orea® orpanizations, must become engaged in ean Fe ied solution, Mine Tz (408) TF.o (non, ae aoa Working toward a solution. eo pa Toa oR Va 0.09 a TDI Acdowadgmet We thnk David Neon, MD a eee = aaa Nd Se i i Masrachisats 68082) Taamoe 80S Commedia Toa as) aT we a a Wig Rian TET OST TE TORO, a0 EE] eee be anda wy A He Sew Yo rego) sas at 282 wegttconipaies ints A Serene T7Mae T2080) 5 STO VPso atin Stas toa aa Ferns apa T0aTT a S20 Yon ee A La Ea rh canal Ter pas)___TaT aa 50 wo PE Cmegniasatene or Taa(ia—— a5 0.09 5 BIO 5 Heal Cai Kucera hon CL tngana a 089) ras 0g) aS tne ana tends 960.194 it! Obes Rat ies Ee az wo Mab Gor ome 2 95-47 tiga 2a EaT 35 BST cha, Sec ar, Se te Meigen IBGE RTOGY BSS rene over aang aut tors tions v2 709m) rege SA a ant tahoe ay: Be et rr cana Te2paT]__ Teo 3 TET Heath iooeae 99 Mnesota TO (057) TST a5: National Institutes of Health Consensus Develon- Bessa oso) sig Se a re eats, m7 3 z 339 peat ey. hr inn Me. 5 02 Tops _—_Taa ENA z wo sor 6. Pe sine Pt Heath inpaonof best. Am! 10pm 18703) 28 Glin Nutr. 1991;53(suppl 6)-1595S-16035. Tasso) TS 09H) z6 a ai, Cat Cs, Sock ad conic Tea 75 ea 50 fc woah he na hn in 111/034) 186 635 s Bz Benton Sh HY, WtansonDF And wea (rar) ve (.04 17 TIS Geny hogan Co Dngn crac a Tapa Tas 099 a TES ches of setebeed tov tac = oS SS = ee West Vigna 1B 21088} 22a OH] TT S02 3° Wiberg, Matos for random dt lng Noth Cucina Te.01097) Ta) a0 ctae BE Haran Ds ser M Leusinger CL. South Carina rae (098 202/087, ea 487 Conden & Obietves and desen of the Behavioral Georg TZU) Ta 099) a TOE Ratfadersnelarcesyten i oeedngrt the Fa BOTA z Tia Skene Sey eat fe Ane ee cacouteaal 10a) a mi Balen erry 127 a5) Tw9 ETH z SEO VO foe Canes on Phys tatu. The Temes a ea Seo and netalino Arona apr of ara Tezoaa)_—207 (05 5 B55 Word Cganzron 19 Won Heh ses TeIN__2200.00 a OZ Tea kp Sa ee A 30 cong ys Wel sounceniel TSF) TOSS) 6851S a acy aa spec cone ey Ceulsana 77.10 56 ZEB papain Pecado he Naor Me oe Tapa} — Ta ea aa S7e— got Pbtc Heth Contes on Record an Qiao Tee eey STB Se ala for an ing Popo oy 1-1, Toe 27939) Tava) 72 50 garth dna Cert Heat era THAT ——T eo a TSO tect fea ad Hanan Seve ple oe es a tons Menara 9098) 8) Ba Stn computer ogra Vesion 612 ithe T1750 UE a ST Cay NC Sa ae ne Cota Tap 7y 01096) 36 GOs taal GB. SUDHA Ue New Mexico | 7-8 (0.93) 14.7 (0.66) 6a 88.0, Research Tange nattutes 1397. Topas 17 TSG 1, Rowand I Steporied weight nd height — a ancl fososs a5 18 a ee ot ree Rann Ca Pace. 10.2 (0.40) 17.0053) oe SoS parison of self-reported and measured height and Wage 907 Tree 772 ht A eptor 920 8293.250 Seepon T2078 9m a Soi — 1. AdylA Bese ad Coding Hate atra Taos Tes we oH 257 yas fhe 1989-79-90 la Tray aT a ST Hayek Reese tng ST ttt footy: Bento od Fal 104 ay 1831.10) 4 ATA Car Gomi. Phiadapnn Fu 8 Lppest Co ewe SITE Rs pe Ds aes HR ‘aos 4822 JAMA, cxoker 2, 1994_ol 28, No 16 (©1000 American Medical Assocation Al rights reserved Downloaded From: https:/