Research Articles

Timely Follow-Up of Positive Fecal Occult

Blood Tests
Strategies Associated with Improvement
Adam A. Powell, PhD, MBA, Amy A. Gravely, MA, Diana L. Ordin, MD, MPH, James E. Schlosser, MD,
Melissa R. Partin, PhD

Background: In light of previous research indicating that many patients fail to receive timely diagnostic
follow-up of positive colorectal cancer (CRC) screening tests, the Veterans Health
Administration (VA) initiated a national CRC diagnosis quality-improvement (QI) effort.
Purpose: This article documents the percent of patients receiving follow-up within 60 days of a
positive CRC screening fecal occult blood test (FOBT) and identifies improvement
strategies that predict timely follow-up.
Methods: In 2007, VA facilities completed a survey in which they indicated the degree to which they
had implemented a series of improvement strategies and described barriers to improve-
ment. Three types of strategies were assessed: developing QI infrastructure, improving care
delivery processes, and building gastroenterology capacity. Survey data were merged with
a measure of 60-day positive-FOBT follow-up. Facility-level predictors of timely follow-up
were identified and relationships among categories of improvement strategies were
assessed. Data were analyzed in 2008.
Results: The median facility-reported 60-day follow-up rate for positive screening FOBTs was 24.5%.
Several strategies were associated with timeliness of follow-up. The relationship between
the implementation of QI infrastructure strategies and timely follow-up was mediated by
the implementation of process-change strategies. Although constraints on gastroenterol-
ogy capacity were often sited as a key barrier, implementation of strategies to address this
issue was unassociated with timely follow-up.
Conclusions: Developing QI infrastructure appears to be an effective strategy for improving FOBT
follow-up when this work is followed by process improvements. Increasing gastroenterology
capacity may be more difficult than improving processes of care.
(Am J Prev Med 2009;37(2):87–93) Published by Elsevier Inc. on behalf of American Journal of
Preventive Medicine

Background Multiple tests for CRC screening are available,

however most VA patients are screened using fecal

C
olorectal cancer (CRC) is the third leading cause
of cancer death among American men and
women.1 It is estimated that one in 18 Americans
will be diagnosed with CRC over their lifespan. Screening
programs aimed at early diagnosis of cancer and removal
of precancerous polyps have been demonstrated to re-
duce both CRC incidence and mortality.2
From the Center for Chronic Disease Outcomes Research, Veterans
Affairs Medical Center (Powell, Gravely, Partin), Minneapolis, Min-
nesota; Veterans Affairs Office of Quality and Performance (Ordin),
Washington, DC; and the Edith Nourse Rogers Memorial Veterans
Hospital (Schlosser), Bedford, Massachusetts
Address correspondence and reprint requests to: Adam A. Powell,
PhD, MBA, Center for Chronic Disease Outcomes Research
(CCDOR), One Veterans Drive (111-0), Minneapolis MN 55417.
E-mail: adam.powell@va.gov.
The full text of this article is available via AJPM Online at
www.ajpm-online.net.
occult blood tests (FOBTs).3 To benefit from FOBT
screening, positive results should be followed with a
diagnostic colonoscopy.4 Research indicates that
while many U.S. healthcare plans have systems in
place to promote CRC screening, few have a process
to monitor follow-up of positive results.5,6 It is there-
fore not surprising that several studies have found
that the majority of patients with a positive screening
test fail to receive a compete diagnostic exam in
a timely manner.7–10 Although the relationship be-
tween positive-FOBT follow-up intervals and clinical
outcomes is unknown, excessive delay in cancer
diagnoses has been linked to increased patient stress
and decreased satisfaction with care.11–13
In 2006, the Veterans Health Administration (VA)
launched a national effort to increase the proportion of
patients receiving a colonoscopy within 60 days of a

Am J Prev Med 2009;37(2) 0749-3797/09/$–see front matter 87

Published by Elsevier Inc. on behalf of American Journal of Preventive Medicine doi:10.1016/j.amepre.2009.05.013
positive FOBT lab result. This 60-day interval was
determined based on consensus among VA clinical and
operations leadership. The project was based on the
premise that the first step to sustainable improvement
is the development of a quality-improvement (QI)
infrastructure. All VA medical centers were encouraged
to establish multidisciplinary quality improvement
teams, work collaboratively with other facilities, and
develop data-collection and patient tracking systems.
Sites were to use this infrastructure to identify barriers,
choose strategies to improve processes of care and/or
gastroenterology capacity, and monitor the effects of
implementing these strategies. It was therefore antici-
pated that the development of QI infrastructure would
not have a direct (unmediated) effect on timely follow-
up, but would instead lead to the implementation of
changes to processes of care and/or efforts to increase
colonoscopy capacity that would in turn improve rates
of timely follow-up.
This article reports the proportion of VA patients
with a positive FOBT CRC screening test between
March and June 2007 who received a colonoscopy
within 60 days and examines the association between
the implementation of various improvement strategies
and the proportion of patients receiving timely follow-
up. The hypotheses that the development of QI infra-
structure indirectly affects timely follow-up through
improvements-to-care processes and/or increases in
gastroenterology capacity are assessed, and the relation-
ship between facility-reported barriers to improvement
and timely follow-up is examined.
Methods
Data Sources and Measures
Key variables in these analyses were obtained from two
sources: (1) a survey of VA medical centers that assessed
activities to improve the timely follow-up of positive FOBTs
and identified perceived barriers to improvement; and (2) a
VA performance monitor that included facility self-reported
rates of 60-day follow-up of patients with positive FOBT
results.
Medical center survey. In June 2007, a web-based survey was
completed by 95% of VA medical centers (n⫽132) on their
FOBT follow-up quality improvement activities. The survey
instrument was developed by a committee of VA researchers,
clinical experts, and administrative leadership and was pre-
tested at six VA facilities. The survey was completed by the staff
member most familiar with their facility’s current positive-FOBT
follow-up improvement activities. Participants were asked to
indicate the degree to which their facility had implemented
each of 16 strategies to improve follow-up of positive FOBTs.
The list of strategies was identified from a 2006 survey of 21
VA facilities participating in a pilot project to improve the
timeliness of positive-FOBT follow-up.
Participants indicated whether each strategy had been fully
implemented, partially implemented, never implemented, or
previously implemented but no longer in use. To assess
predictors of timely follow-up, responses were transformed
into dichotomous variables (fully implemented⫽1, all other
responses⫽0). After the data were collected, strategies were
classified into three general categories (Table 1). Strategies
designed to identify problems, organize resources to focus on

Table 1. Percent of facilities implementing strategies to improve the timeliness of positive FOBT follow-up
Fully In the process Have not Implemented in the
Improvement strategy implemented of implementing implemented past but no longer use
Development of QI infrastructure
Track colonoscopy supply and demand 56 28 12 4
Form an active multidisciplinary 56 22 19 2
improvement team
Participate in an improvement collaborative 51 21 28 1
Create system for tracking follow-up of 42 38 20 0
⫹FOBTs
Track number of inappropriate FOBTs 33 28 37 2
Track number of incomplete colonoscopies 28 20 48 4
Process changes
Strategies to decrease cancellations/no-shows 82 12 5 1
Create/revise primary care/gastroenterology 64 22 11 4
service agreement
Consult template revision 59 25 15 1
Revise colonoscopy prep 54 21 24 1
education/protocols
Revise CRC screening clinical reminder 43 31 26 0
Gastroenterology capacity building
Initiate/increase use of fee-based 44 16 35 5
colonoscopies
Hire additional nurses/other staff 29 33 37 1
Hire additional colonoscopists 23 35 42 0
Contract additional onsite colonoscopists 15 18 65 2
Add additional endoscopy suites 15 27 55 3
CRC, colorectal cancer; FOBT, fecal occult blood test

88 American Journal of Preventive Medicine, Volume 37, Number 2 www.ajpm-online.net

improvement, or measure the effect of system change were
classified as development of QI infrastructure. Strategies that
involved modifications to the care delivery system were clas-
sified as process changes. Attempts to increase gastroenterology
resources were classified as gastroenterology capacity building.
Variables were created representing these three categories by
adding the number of items that had been fully implemented
by each facility within each category.
Responses to an open-ended survey question, What have
been the most significant barriers to improvement? were coded
independently by two coders (Kappa⫽0.97). Ten response
categories were mentioned by seven or more facilities (⬎5%)
and were included in analyses.
FOBT follow-up performance monitor. In 2006, the VA
implemented an FOBT follow-up performance monitor that
required facilities to flow-chart their FOBT follow-up process,
initiate improvement efforts, and report data on the timeli-
ness of positive-FOBT follow-up. As part of this monitor, in
October 2007, facilities reported the number of patients
receiving a positive CRC screening FOBT lab result between
March and June 2007 and, of those tracked, the number
receiving a colonoscopy within 60 days of the FOBT lab result.
(Although facilities were not given explicit instructions re-
garding tests completed through digital rectal exam, VA
guidelines consider this method unacceptable for CRC
screening.) Data were reported by 133 (96%) medical cen-
ters, including 125 facilities that had completed the July 2007
medical center survey.
Analysis
The facility 60-day positive-FOBT follow-up rate is pre-
sented (median and interquartile range), as well as fre-
quency distributions indicating the degree to which each
improvement strategy was implemented. To assess which if
any of the improvement strategies predicted timely FOBT
follow-up, each of the dichotomous improvement-strategy
variables was entered into separate regression models using
the percent of positive-FOBT patients receiving 60-day
follow-up as the dependent variable. Models were also
created using the total number of strategies fully imple-
mented by each medical center and each of the dichoto-
mous barriers to improvement variables as independent
variables.
Two hypotheses regarding the relationship between the
implementation of QI infrastructure strategies and 60-day
follow-up were tested: (1) that this relationship is mediated
through the implementation of process-change strategies;
and (2) that this relationship is mediated through the imple-
mentation of gastroenterology capacity building strategies.
For each hypothesis, Baron and Kenny’s three-step regression
procedure was used to ascertain the relationships between a
proposed independent, dependant and mediator variable.14
In the first step, the dependent variable was regressed on the
independent variable to determine if there was a relationship
to explain. Second, to ascertain whether the independent
variable was associated with the proposed mediator, the
mediator variable was regressed on the independent variable.
Third, the dependent variable was regressed on both the
mediator and the independent variable. This assessed
whether the mediator was associated with the dependent
variable while controlling for any direct effect of the indepen-
August 2009
dent variable. This third regression also assessed the direct
(unmediated) effect of the independent variable on the
dependent variable. In addition to conducting this three-step
analysis, the significance of the mediated path was tested.15,16
All models were first developed using linear least squares
regression analysis; however, the error terms of some
models were not normally distributed. Parameters of each
model were therefore also estimated using bootstrapping
techniques, which are robust to nonnormality.17 This com-
putational method involves resampling the data multiple
times to generate a mean estimate of coefficients and
confidence intervals. Both bootstrapped and nonboot-
strapped results were very similar. Bootstrapped unstand-
ardized beta coefficients and 95% CIs are reported. The
unstandardized beta coefficient quantifies the change in
the dependent variable attributed to one unit change in
the independent variable.
Although five variables were initially considered for
inclusion as covariates in the regression models—number
of clinical gastroenterology staff, number of total gastroen-
terology staff, number of outpatients seen by the facility,
median patient age, and median patient income— only the
number of clinical gastroenterology staff was associated
with the 60-day follow-up dependent variable. However,
data on this measure were unavailable for 26 facilities.
Analyses conducted on facilities with complete data on
gastroenterology staffing indicated that controlling for this
measure had little effect on the results and did not alter
our conclusions. Therefore, the results presented here do
not control for this factor.
All analyses were conducted in R version 2.7.1 with the
exception of the bootstrapped mediation analyses which were
generated in SAS version 9.1 with macros created by Preacher
and Hayes.18
Results
The median facility-reported 60-day follow-up rate for
patients receiving positive FOBT lab results between
March and June 2007 was 24.5% (interquartile range⫽
13.8% to 40.7%). Table 1 depicts the degree to which
each of the 16 change strategies had been imple-
mented at VA facilities. Strategies to decrease cancella-
tions or no-shows and to create/revise primary care/
gastroenterology service agreements were most common.
The effect of each improvement strategy on the timely
follow-up of positive FOBTs is reported in Table 2.
Strategies that were significantly associated with
timely follow-up include two strategies to develop QI
infrastructure and three process-change strategies. Ad-
ditionally two QI infrastructure strategies and one
gastroenterology capacity strategy approached signifi-
cance. On average, facilities indicated that they had
fully implemented 6.84 improvement strategies. The
total number of strategies fully implemented was posi-
tively associated with 60-day follow-up (␤⫽1.9, 95%
CI⫽0.9, 3.0).
Results of two mediation analyses are depicted in
Figure 1. The hypothesis that the effect of QI infrastruc-
Am J Prev Med 2009;37(2) 89
ture on timely follow-up Table 2. Relationship between implementation of improvement strategies (fully implemented
is mediated by process versus any other response) and 60-day follow-up
changes was supported. Improvement strategy (fully implemented versus any other
The number of QI infra- response) ␤ (95% CI)
structure strategies im- DEVELOPMENT OF QI INFRASTRUCTURE
plemented significantly Track colonoscopy supply and demand 8.5 (0.7 to 16.4)
predicted timely FOBT Form an active multidisciplinary improvement team 7.2 (⫺0.3 to 14.7)
follow-up (Path a, ␤⫽2.9, Participate in an improvement collaborative 5.6 (⫺1.9 to 13.2)
95% CI⫽0.6, 5.2). The Create system for tracking follow-up of ⴙFOBTs 7.7 (0.3 to 15.1)
Track number of inappropriate FOBTs 2.0 (⫺6.6 to 10.6)
QI infrastructure variable Track number of incomplete colonoscopies 5.8 (⫺2.5 to 15.1)
also predicted the num- PROCESS CHANGES
ber of process-change Strategies to decrease cancellations/no-shows 7.9 (0.1 to 15.7)
variables implemented Create/revise primary care/gastroenterology service agreement 11.2 (4.2 to 18.3)
(Path bprocess, ␤⫽0.36, Consult template revision 5.1 (⫺1.5 to 11.8)
Revise colonoscopy prep education/protocols 10.6 (3.9 to 17.4)
95% CI⫽0.24, 0.48). Revise CRC screening clinical reminder 7.6 (⫺0.2 to 15.4)
When timely follow-up GASTROENTEROLOGY CAPACITY BUILDING
was regressed on both Initiate/increase use of fee-based colonoscopies 1.3 (⫺6.7 to 9.2)
QI infrastructure and Hire additional nurses/other staff 7.8 (⫺.2 to 15.7)
process-change variables, Hire additional colonoscopists 4.9 (⫺4.2 to 14.0)
Contract additional onsite colonoscopists 3.5 (⫺7.1 to 14.0)
the effect of process Add endoscopy suites 3.7 (⫺7.5 to 14.9)
change was significant
Boldface indicates p⬍0.05
(Path cprocess, ␤⫽4.0, 95% CRC, colorectal cancer; FOBT, fecal occult blood test
CI⫽1.8, 6.3) but the ef-
fect of QI infrastructure
was no longer significant (Path a=process, ␤⫽1.8, 95% Discussion
CI⫽⫺0.5, 4.0). The coefficient associated with the The VA’s adoption of a CRC screening performance
mediated path was significant (Path bcprocess, ␤⫽1.2, measure and clinical reminder system has led to sub-
95% CI⫽0.3, 2.6). stantial increases in VA screening rates, primarily
The hypothesis that the effect of QI infrastructure on through the extensive use of FOBTs.3 As a result, VA
timely follow-up is mediated by changes in gastroenter- patients are more likely to be screened for CRC than
ology capacity was not supported. The first mediation are individuals covered by commercial, Medicare or
analysis had already established that QI infrastructure Medicaid plans.19 However, this increase in FOBT
predicted timely follow-up (Path a). QI infrastructure screening has led to a downstream increase in the
also predicted the number of gastroenterology capacity demand for colonoscopies following positive screens.
changes implemented (Path bcapacity, ␤⫽0.17, 95% Results of this study suggest that the benefit to
CI⫽0.05, 0.28). However, when timely follow-up was veterans of the VA’s high screening rate may be
regressed both on QI infrastructure and gastroenterol- limited by lower rates of timely follow-up. Only one
ogy capacity, the relationship between capacity and in four patients received follow-up colonoscopies
timely follow-up was not significant (Path ccapacity, ␤⫽2.4, within 60 days of a positive FOBT lab result. Other
95% CI⫽⫺0.4, 5.3) and the relationship between QI studies with follow-up windows ranging from 6 to 12
infrastructure and timeliness remained significant months have documented CRC screening follow-up
(Path a=capacity, ␤⫽2.7, 95% CI⫽0.4, 5.0). Additionally rates of 34 to 65%.8,10,20 –22
the mediated path coefficient was not significant (Path
bccapacity, ␤⫽0.3, 95% CI⫽⫺0.2, 1.1).
Improving Timely FOBT Follow-Up
Table 3 displays commonly mentioned barriers to
timely FOBT follow-up, their frequency of mention, There does not appear to be a single “silver bullet” for
and the relationship between each barrier and 60-day improving timely follow-up of positive FOBTs. Several
follow-up. The most commonly sited barrier to im- improvement strategies that were implemented by VA
provement was the number of endoscopists on staff. facilities significantly predicted the proportion of patients
The second and third most commonly mentioned receiving timely follow-up. These include two QI infra-
barriers also concerned limitations in gastroenterol- structure strategies (tracking colonoscopy supply and
ogy capacity (number of non-endoscopist staff and demand and creating a system for tracking follow-up of
gastroenterology clinic space issues). Facilities that ⫹FOBTs) and three process-improvement strategies
mentioned either the number of endoscopists or the (creating/revising primary care/gastroenterology ser-
number of non-endoscopist staff as a barrier had a vice agreement, revising colonoscopy prep education
significantly lower rate of FOBT follow-up than those protocols, and decreasing cancellations and/or no-
that did not. shows). These strategies, either individually or in com-

90 American Journal of Preventive Medicine, Volume 37, Number 2 www.ajpm-online.net

 The hypothesis that
the implementation of
process changes medi-
ates the relationship be-
tween the development
of QI infrastructure and
timely FOBT follow-up
was supported. Creating
QI infrastructure takes
substantial resources,
especially if collecting
the necessary data for
tracking events and
outcomes is not easily
automated. This research
suggests that in order
for the development of
QI infrastructure to trans-
late into improvement
there must be enough
resources remaining and
enough tension for con-
tinued change so that
the infrastructure created
is used to facilitate the
implementation of pro-
cess changes. By plan-
ning up front for work
beyond the development
of QI infrastructure,
there is a greater chance
that efforts will lead to
improvement.
The hypothesis that
the relationship between
QI infrastructure and
timely follow-up would
be mediated by the im-
plementation of strate-
gies to increase gastro-
enterology capacity was
not supported. One
possible reason for this
is that although a strong
QI infrastructure may
help in identifying what
type of additional gas-
Figure 1. Analysis of process-change and capacity-building strategies as mediators of the relationship troenterology clinic re-
between building QI infrastructure and 60-day follow-up. Dotted lines represent non-significant paths.
sources are needed, a
QI, quality improvement
facility’s ability to obtain
these additional resources
bination, should be considered in the development of may be unrelated to QI infrastructure. Instead changes
future interventions to improve the timeliness of positive- in gastroenterology capacity are more likely to be a
FOBT follow-up. Additionally these results suggest that function of the facility’s capital planning process, avail-
the percent of patients receiving timely follow-up may able funds, and the local labor pool for appropriate
be partially a function of the total number of improve- staff (e.g., gastroenterology physicians and nurses).
ment strategies implemented. Another possibility is that process changes have greater

August 2009 Am J Prev Med 2009;37(2) 91

potential for impact than Table 3. Relationship between barriers to improvement and 60-day follow-up
increases in capacity. Re-
search on delay in health- Barrier
care systems suggests that
CAPACITY-RELATED BARRIERS
access limitations are Number of endoscopists
more frequently the re- Number of non-endoscopist staff
sult of suboptimal man- Number of procedure/recovery rooms
agement of current ca- Resource/staff limitations (type not specified)
pacity than insufficient Fee basis limitations
23,24 OTHER BARRIERS
levels of capacity. A Patient cancellations/no-shows
third possibility is that Tracking/record keeping problems
the medical centers that Poor primary care/gastroenterology communication
were the poorest per- Patient travel issues
formers to start with Inappropriate FOBTs
were more likely to allo- Boldface indicates p⬍0.05
cate financial resources FOBT, fecal occult blood test
to increasing gastroen-
terology capacity. A single-point measure of timely
positive-FOBT follow-up is unable to account for any
such variations in baseline performance.
Barriers to Improvement
The most commonly cited barriers to improvement in-
volved capacity constraints. Sites listing insufficient gas-
troenterology staff as a barrier had a lower percent of
patients receiving timely follow-up than those that did
not. One interpretation of this finding is that higher
performing facilities have adequate staff to meet their
demand for colonoscopies. It is also possible that some
facilities believe that long waits for colonoscopies are
unavoidable because their gastroenterology clinic is
understaffed. This belief may keep these facilities from
attempting to identify and implement process improve-
ments. The fact that none of the improvement strate-
gies designed to address gastroenterology capacity con-
straints were associated with timely follow-up suggests
that this barrier may be more difficult or take more
time to address than process inefficiencies.
Limitations
Data on timeliness of follow-up were self-reported by
facilities. Data definitions and the completeness of
reporting may have varied across sites. If inaccuracies in
these self-reports varied systematically between facilities
that did and did not fully implement a system to track
positive FOBTs, the reported association between fully
implementing a positive-FOBT tracking system and
timeliness of follow-up may be spurious. It should be
noted however that although 20% of facilities indicated
that they had not fully implemented a positive-FOBT
tracking system, all facilities included in this analysis
had, at a minimum, developed a means of obtaining
positive-FOBT follow-up data for the 4-month FOBT
follow-up performance monitor reporting period.
A second limitation is that because data on imple-
mentation strategies were collected in June of 2007 and
92 American Journal of Preventive Medicine, Volume 37, Number 2
Number
mentioning ␤ (95% CI)
55 ⴚ9.4 (ⴚ17.6, ⴚ1.2)
35 ⴚ9.3 (ⴚ15.5, ⴚ3.1)
33 ⫺1.4 (⫺10.2, 7.5)
19 ⫺1.5 (⫺12.4, 9.3)
10 ⫺5.4 (⫺17.2, 6.4)
19 3.7 (⫺7.9, 15.3)
21 8.8 (⫺2.3, 20.0)
10 1.8 (⫺9.1, 12.8)
9 18.5 (⫺2.6, 39.6)
8 ⫺5.3 (⫺15.8, 5.2)
60-day follow-up data were collected on patients with
positive FOBT results between March and June of 2007,
it is possible that some implemented strategies had not
been in place long enough to affect 60-day follow-up
rates. A third limitation is that no information was
obtained on the context in which improvement strate-
gies were implemented. Contextual factors may play a
critical role in determining each strategy’s success.
Fourth, this work does not assess the relationship
between rates of 60-day follow-up and clinical or psy-
chological outcomes. Further research is necessary to
examine this issue. Finally, because the strategies im-
plemented were measured but not manipulated, this
research does not provide definitive evidence of the
effectiveness of each strategy. This work does however
identify several specific strategies that were associated
with improvement. Perhaps more importantly, this
work suggests a framework for the creation of improve-
ment programs in which initial efforts are focused on
the development of QI infrastructure and are then
followed by targeted strategies designed to improve
processes of care.
We would like to express our appreciation to George Ponte,
George Jackson, Dawn Provenzale, Heidi Martin, David Hagg-
strom, Emily Bliss, and Janice Hersh for their help in the
development of the data-collection instruments used in this
study. Additionally we thank all VA staff members that were
involved in the CRC diagnosis quality improvement project
for their hard work and cooperation with this research.
No financial disclosures were reported by the authors of
this paper.
References
1. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2008. CA Cancer J Clin
2008;58:71–96.
2. Pignone M, Rich M, Teutsch SM, et al. Screening for colorectal cancer in
adults at average risk: a summary of the evidence for the U.S. Preventive
Services Task Force. Ann Intern Med 2002;137:132– 41.
www.ajpm-online.net
 3. El-Serag HB, Petersen L, Hampel H, Richardson P, Cooper G. The use of
screening colonoscopy for patients cared for by the department of veterans
affairs. Arch Intern Med 2006;166:2202– 8.
4. Winawer S, Fletcher R, Rex D, et al. Colorectal cancer screening and
surveillance: clinical guidelines and rationale— update based on new
evidence. Gastroenterology 2003;124:544 – 60.
5. Klabunde CN, Riley GS, Mandelson MT, Frame PS, Brown ML. Health plan
policies and programs for colorectal cancer screening: a national profile.
Am J Manag Care 2004;10:273–9.
6. Sarfaty M, Myers RE. The effect of HEDIS measurement of colorectal
cancer screening on insurance plans in Pennsylvania. Am J Manag Care
2008;14:277– 82.
7. Etzioni DA, Yano EM, Rubenstein LV, et al. Measuring the quality of
colorectal cancer screening: the importance of follow-up. Dis Colon
Rectum 2006;49:1002–10.
8. McGarrity TJ, Long PA, Peiffer LP, Converse JO, Kreig AF. Results of a
television-advertised public screening program for colorectal cancer. Arch
Intern Med 1989;149:140 – 4.
9. Paszat L, Rabeneck L, Kiefer L, Mai V, Ritvo P, Sullivan T. Endoscopic
follow-up of positive fecal occult blood testing observed in the Ontario
FOBT project. Can J Gastroenterol 2007;21:379 – 82.
10. Lurie JD, Welch HG. Diagnostic testing following fecal occult blood
screening in the elderly. J Natl Cancer Inst 1999;91:1641– 6.
11. Risberg T, Sorbye SW, Norum J, Wist EA. Diagnostic delay causes more
psychological distress in female than in male cancer patients. Anticancer
Res 1996;16:995–9.
12. Jakobsson S, Horvath G, Ahlberg K. A grounded theory exploration of the
first visit to at cancer clinic—strategies for achieving acceptance. Eur J
Oncol Nurs 2005;9:248 –57.
13. Yardley SJ, Davis CL, Sheldon F. Receiving a diagnosis of lung cancer: patients’
interpretations, perceptions and perspectives. Palliat Med 2005;15:379 – 86.
What’s new online?
Visit www.ajpm-online.net today to find out how you can search 400 top medical and
health sciences journals online, including Medline.
August 2009
14. Baron RM, Kenny DA. The moderator–mediator variable distinction in
social psychological research: conceptual, strategic, and statistical consid-
erations. J Pers Soc Psychol 1986;51:1173– 82.
15. Sobel ME. Asymptotic confidence intervals for indirect effects in structural
equation models. In: Leinhardt S, ed. Sociological methodology. Washing-
ton DC: American Sociological Association, 1982:290 –312.
16. Preacher KJ, Hayes AF. Asymptotic and resampling strategies for assessing
and comparing indirect effects in multiple mediator models. Beh Res
Methods 2008;40:879 –91.
17. Efron B, Tibshirani R. An introduction to the bootstrap. New York:
Chapman & Hall, 1993.
18. Preacher KJ, Hayes AF. SPSS and SAS procedures for estimating indirect
effects in simple mediation models. Behav Res Methods Instrum Comput
2004;36:717–31.
19. United States Department of Veterans Affairs. Quality of veterans’ health
care rates high marks. www1.va.gov/opa/pressrel/pressrelease.cfm?id⫽
429.
20. Myers RE, Turner B, Weinberg D, et al. Impact of a physician-oriented
intervention on follow-up in colorectal cancer screening. Prev Med 2004;
38:375– 81.
21. Fisher DA, Jeffreys A, Coffman CJ, Fasanella K. Barriers to full colon
evaluation for a positive fecal occult blood test. Cancer Epidemiol Biomar-
kers Prev 2006;15:1232–5.
22. Garman KS, Jeffreys A, Coffman C, Fisher DA. Colorectal cancer screening,
comorbidity, and follow-up in elderly patients. Am J Med Sci 2006;332:
159 – 63.
23. Murray M, Berwick DM. Advanced access: reducing waiting and delays in
primary care. JAMA 2003;289:1035– 40.
24. Murray M, Tantau C. Same day appointments: exploding the access
paradigm. Fam Pract Manag 2000;7:45–50.
Am J Prev Med 2009;37(2) 93