Attachment 5 Klabunde et al

Attachment 5
Klabunde, Meadow, Frame et al, 2003.

Klabunde CN, Frame PS, Meadow A, Jones E, Vernon SW. A national survey of
primary care physicians’ colorectal cancer screening recommendations and
practices. Preventive Medicine 2003, 36: 352–362.

Available online at www.sciencedirect.com
R

Preventive Medicine 36 (2003) 352–362

www.elsevier.com/locate/ypmed

A national survey of primary care physicians’ colorectal cancer
screening recommendations and practices
Carrie N. Klabunde, Ph.D.,a,* Paul S. Frame, M.D.,b Ann Meadow, Sc.D.,c
Elizabeth Jones, M.D.,d Marion Nadel, Ph.D.,e and Sally W. Vernon, Ph.D.f
a

Health Services and Economics Branch, Applied Research Program, National Cancer Institute, Bethesda, MD, USA
Tri-County Family Medicine, Cohocton, New York, and the Department of Family Medicine, University of Rochester
School of Medicine and Dentistry, Rochester, NY, USA
c
Office of Research, Development, and Information, Centers for Medicare and Medicaid Services, Baltimore, MD, USA
d
Diagnostic Radiology Department, Clinical Center, National Institutes of Health, Bethesda, MD, USA
e
Division of Cancer Prevention and Control, Centers for Disease Control and Prevention, Atlanta, GA, USA
f
University of Texas Health Science Center at Houston, School of Public Health, Houston, TX, USA
b

Abstract
Background. National data on providers’ colorectal cancer (CRC) screening knowledge, attitudes, and practices are sparse. This study
assessed primary care physicians’ (PCPs’) beliefs about the effectiveness of CRC screening, their recommendations for screening, their
perceptions of the influence of published guidelines on their CRC screening recommendations, and how they conduct CRC screening in their
clinical practices.
Methods. A questionnaire was administered to a nationally representative sample of practicing PCPs. Of 1718 eligible physicians, 1235
(72%) responded.
Results. Only 2% of PCPs said they did not recommend CRC screening. Over 80% indicated that they most often recommend CRC
screening with fecal occult blood testing and/or flexible sigmoidoscopy, although colonoscopy was perceived as the more effective screening
modality. Nearly two-thirds of obstetrician/gynecologists and one-fourth of other practitioners reported conducting fecal occult blood testing
exclusively by digital rectal exam. Only 29% of PCPs said they perform sigmoidoscopy. Estimated volumes of ordering, performing, or
referring for CRC screening were low, and ⬍20% reported that three-fourths or more of their older patients were up to date with CRC
screening as recommended by the physician. Many PCPs reported recommending CRC screening at nonstandard starting ages or
too-frequent intervals.
Conclusions. Awareness of CRC screening among PCPs in the United States is high. However, knowledge gaps about the timing and
frequency of screening and suboptimal screening delivery were evident.
© 2003 American Health Foundation and Elsevier Science (USA). All rights reserved.
Keywords: Colorectal cancer; Screening; Primary care; Health services delivery

Introduction
During the latter half of the 1990s, evidence emerging
from clinical studies [1–9] prompted a number of expert
groups [10 –12] to issue or revise their recommendations for
colorectal cancer (CRC) screening. With varying ap* Corresponding author. Applied Research Program, National Cancer
Institute, Executive Plaza North Room 4005, 6130 Executive Boulevard,
Bethesda, MD 20892-7344, USA. Fax: ⫹1-301-435-3710.
E-mail address: ck97b@nih.gov. (C.N. Klabunde).

proaches to evaluating evidence and formulating recommendations, these groups developed guidelines that differ
somewhat in the CRC screening modalities they support as
well as the frequency with which recommended modalities
are to be applied. Nevertheless, a consensus has emerged in
recent years that average-risk adults aged 50 years and older
should be screened for CRC.
Despite the evidence in support of screening for CRC
and the dissemination of screening guidelines, national surveys of the public indicate that the majority of adults aged
50 and older have never been screened for this disease. For

0091-7435/03/$ – see front matter © 2003 American Health Foundation and Elsevier Science (USA). All rights reserved.
doi:10.1016/S0091-7435(02)00066-X

C.N. Klabunde et al. / Preventive Medicine 36 (2003) 352–362

example, data from the 1998 National Health Interview
Survey (NHIS), an in-person household survey of a nationally representative sample of the noninstitutionalized U.S.
population, show 33% of adults aged 50 years and older to
have received a fecal occult blood test (FOBT) in the 2
years before the interview and 38% to have ever undergone
a proctoscopy/sigmoidoscopy examination [13]. The 1999
Behavioral Risk Factor Surveillance System (BRFSS), a
population-based, random-digit-dialed telephone survey of
the noninstitutionalized U.S. population, indicates that 40%
of adults aged 50 years and older have ever had an FOBT
and 44% a sigmoidoscopy or colonoscopy examination
[14]. Trend data from these surveys demonstrate only modest improvement in CRC screening rates over time [13–15].
Health care providers play a key role in recommending
screening to eligible patients. Yet nationally representative
data on providers’ CRC screening knowledge, attitudes, and
practices are sparse. Prior studies have been state or locally
based, with small sample sizes, low response rates, a focus
on opinions about guidelines rather than actual practice or,
if addressing provider practices, have targeted only one or
two of the four currently recognized CRC screening modalities [16 –27]. To address these gaps, the National Cancer
Institute—in collaboration with the Centers for Disease
Control and Prevention and the Centers for Medicare and
Medicaid Services—fielded the Survey of Colorectal Cancer Screening Practices in Health Care Organizations. This
comprehensive survey of primary care and specialty physicians and health plan medical directors is designed to obtain
current, nationally representative data on how CRC screening is being conducted in the United States, and to identify
barriers to screening delivery in community practice [28]. In
this report, we summarize primary care physicians’ (PCPs’)
beliefs about the effectiveness of CRC screening, their recommendations to patients for CRC screening, and their
self-reported CRC screening activities in their clinical practices.

Methods
Sampling methodology
Using the American Medical Association’s Physician
Masterfile as the sampling frame, we surveyed a nationally
representative sample of PCPs in 1999 –2000. The Masterfile contains demographic and practice-related data on all
allopathic and virtually all osteopathic physicians in the
United States. Eligible respondents were PCPs aged 75
years and younger who were listed in the database as having
an active license to practice medicine and whose major
professional activity involves patient care. We selected physicians practicing the primary care specialties of general
practice (GP), family practice (FP), general internal medicine (IM), and obstetrics/gynecology (OBG). Physicians
who were retired, in residency training, or involved in

353

full-time teaching, research, or administration were excluded. Obstetrician/gynecologists were included in the
sample because of their role as providers of preventive
services for many women in the United States [29]. We
selected a systematic, stratified random sample of 2098
physicians using physician specialties as the sampling strata
and selection proportional to the specialty’s representation
in the total U.S. physician population. Sample sizes in each
specialty group were selected to provide point estimates of
population proportions within ⫾3% at a 95% confidence
interval. Sample selection was accomplished after the sampling frame database had been sorted by U.S. Census region
(Northeast, North Central, South, and West), urban vs. rural
practice location, and physician gender to ensure adequate
representation of rural and female physicians and each Census region in the samples. Of the initial sample, 380 physicians subsequently were determined to have retired or died,
to not currently practice primary care medicine, or to be
unlocatable following an extensive search for current contact information. Exclusion of these physicians left 1718
eligible physicians.
Survey methodology
Sampled physicians were sent an advance mailing in the
fall of 1999 that contained a cover letter describing the
objectives of the survey, letters of support from five medical
societies and the U.S. Surgeon General, and a postcard with
a stamped return envelope on which physicians were asked
to (1) verify their specialty and status as a practicing PCP,
and (2) indicate their preferred mode of response to the
survey (i.e., mail, fax, telephone, or secure Internet Web
site). Physicians who responded to the advance mailing
were sent a subsequent mailing that included the mail or fax
version of the questionnaire or instructions on how to complete the survey by telephone or Internet, depending on their
stated preference. Approximately 6 weeks later, physicians
who did not respond to the advance mailing were sent a
follow-up mailing containing the mail version of the questionnaire. A second follow-up mailing of the questionnaire
was sent by express mail to eligible nonrespondents in
February 2000. Telephone follow-up of nonrespondents
was undertaken in March and early April 2000. All respondents received a prepaid $50 honorarium check.
The survey took approximately 20 min to complete and
was organized into the following four sections: cancer
screening beliefs and practices, attitudes toward and training in CRC screening, recommendations for and use of CRC
screening modalities, and practice and other characteristics.
Items inquiring about PCPs’ CRC screening recommendations and practices were specific to asymptomatic, averagerisk patients and encompassed the four CRC screening modalities most commonly mentioned in published guidelines,
i.e., FOBT, sigmoidoscopy, colonoscopy, and double-contrast barium enema (DCBE). Physicians were asked
whether they recommended a specific modality as a CRC

354

C.N. Klabunde et al. / Preventive Medicine 36 (2003) 352–362

Table 1
Characteristics of primary care physicians and their practice settings, Survey of Colorectal Cancer Screening Practices, 1999 –2000

Physician
Gendera (male)
Raceb (white, non-Hispanic)
Board certifiedb
International medical graduateb
Medical school affiliationb
Age ⱖ 50b
Been screened for colorectal cancerb
Practice
Metropolitan locationb
Practice typeb
Solo
Single specialty group
Multispecialty setting
ⱖ50% of patients are age 50⫹b
Sees ⬎100 patients during a typical weekb
⬎50% of patients are covered by managed careb
a
b

Family practice
(n ⫽ 423; %)

General practice
(n ⫽ 100; %)

General internal
medicine
(n ⫽ 488; %)

Obstetrics/
gynecology
(n ⫽ 224; %)

77.9
79.7
81.8
15.1
40.0
36.5
51.5

84.5
62.4
11.1
34.5
12.9
79.0
74.2

79.4
66.8
78.0
27.3
41.1
38.8
47.6

70.8
75.4
82.2
20.6
37.5
47.7
63.1

49.4

57.2

68.7

66.5

24.8
46.0
29.2
32.1
47.3
39.5

42.7
32.6
24.7
41.3
34.5
29.2

21.1
37.2
41.7
69.4
27.8
40.9

29.6
44.0
26.4
14.9
31.6
61.5

P ⬍ 0.05.
P ⬍ 0.001.

screening strategy to their patients and, if so, their policies
on starting age, test frequency, and stopping age. They also
were asked to identify the test or test combination that they
most often recommend for CRC screening. Items assessing
perceived test effectiveness and the influence of published
guidelines on physicians’ CRC screening recommendations
were measured by using three-point Likert scales that included the categories “very effective,” “somewhat effective,” and “not effective,” or “very influential,” “somewhat
influential,” and “not influential,” as appropriate. A separate
set of items inquired about how often physicians order,
perform, or refer patients for CRC screening with each
modality. Physicians reporting use of a specific modality
were asked additional items about such practice details as
home- versus office-based FOBT, whether the physician
performs or refers patients for sigmoidoscopy, and to whom
referrals are made for non-FOBT modalities. Finally, to
assess patients’ screening status, physicians were asked to
estimate the proportion of their patients aged 50 years and
older who are up to date with the CRC screening recommended by the physician, according to the following response categories: ⬍25%, 25–50%, 51–75%, 76 –100%,
and “don’t know.”
Statistical analysis
We used descriptive statistics and contingency tables
with ␹2 tests to examine physicians’ CRC screening recommendations and practices. A sample weight that accounts
for the probability of selection into the sample as well as a
slightly higher rate of nonresponse among GPs, IMs, and
physicians aged 60 years and older was assigned to each

respondent. Sampling weights were applied in the statistical
analyses to permit generalization of the results to the U.S.
population of practicing PCPs. Analyses were stratified by
the four primary care specialties represented in this study to
facilitate the reporting and interpretation of findings in a
format that has relevance to practitioners as well as policy
makers.

Results
Description of respondents
A total of 1235 practicing PCPs responded to the survey
(overall response rate ⫽ 72%; GPs ⫽ 68%, FPs ⫽ 75%,
IMs ⫽ 69%, OBGs ⫽ 75%). Ninety percent responded by
mail, 6% by Internet, 2% by telephone, and 2% by fax. The
majority were male, white, graduates of U.S. medical
schools, and in group practice (Table 1). The mean age was
48.0 years (range, 29 –75 years). Differences in physician
and practice characteristics by specialty were apparent. For
example, GPs tended to be older, to lack board certification
or an affiliation with a medical school (defined as an adjunct, clinical, or other faculty appointment), and to be
graduates of non-U.S. medical schools and in solo practice.
FPs reported the highest patient volumes of the four specialties and were more likely to practice in rural locations.
IMs tended to provide care to an older patient population
and to practice in multispecialty settings. OBGs indicated
that they saw younger patients, many of whom were covered by managed care plans.

C.N. Klabunde et al. / Preventive Medicine 36 (2003) 352–362

355

sigmoidoscopy (7–11%), and flexible sigmoidoscopy alone
(1– 4%). GPs were more likely than other practitioners to
indicate that they most often recommend FOBT alone
(41%). Colonoscopy (3–13%) and DCBE (0 – 4%) were less
frequently mentioned as the CRC screening strategies most
often recommended. Overall, only 2% of respondents said
that they did not recommend CRC screening in their practices.
Modality-specific recommendations for and practice of
colorectal cancer screening
Fig. 1. Primary care physicians’ perceived effectiveness of colorectal
cancer screening tests for average-risk adults aged 50⫹, Survey of Colorectal Cancer Screening Practices, 1999 –2000

Perceived test effectiveness and screening strategy most
often recommended
Over 80% of respondents reported the belief that
colonoscopy is a “very effective” screening procedure in
reducing CRC mortality in patients aged 50 and older,
compared with 43% (IMs) to 59% (GPs) who indicated that
flexible sigmoidoscopy is “very effective” (Fig. 1). Respondents were less likely to rate DCBE (23–38%) and FOBT
(25–34%) as “very effective.” IMs and OBGs more often
rated DCBE than FOBT as “very effective.” Among GPs,
the pattern was reversed, with a higher percentage reporting
FOBT than DCBE to be “very effective.” FPs were equally
likely to rate DCBE and FOBT as “very effective.”
In contrast to their perceptions of test effectiveness, 79%
(GPs and OBGs) to 89% (FPs) of respondents indicated that
they most often recommended FOBT and flexible sigmoidoscopy— either alone or in combination—to their averagerisk patients as a CRC screening strategy (Table 2). In all
specialty groups except GP, the combined use of FOBT and
flexible sigmoidoscopy was most often reported (40 –55%),
followed by FOBT alone (22–30%), either FOBT or flexible

Ninety-five percent of respondents indicated that they
recommend FOBT as a CRC screening modality to asymptomatic, average-risk patients (Table 3). A range of recommended starting ages was reported for this test. Although
age 50 was commonly cited, 49% (IMs) to 69% (GPs)
mentioned a starting age other than 50. Up to 43% of
respondents cited age 40 as their recommended starting age.
Ninety-four percent or more indicated that they recommend
screening with FOBT on an annual or biennial basis. Most
reported that they did not have a “stopping” age at which
they no longer recommended FOBT screening to patients.
GPs reported a lower monthly ordering volume for FOBT
screening than did other PCPs: 23% of GPs compared with
41–51% of other practitioners indicated that they order or
perform the test 21 or more times during a typical month. A
low percentage (11–32%) of PCPs reported that they exclusively use home-based FOBT for the majority of their patients. Of note is the sizeable percentage of OBGs (64%)
who indicated that they exclusively use office-based FOBT.
There was less consistency in recommendations for flexible sigmoidoscopy as a CRC screening modality; 82% of
FPs and IMs reported recommending this test to asymptomatic, average risk patients, compared with 61% of GPs and
68% of OBGs (Table 4). Age 50 was the most commonly
mentioned starting age (65– 87%). A test frequency of once
every 5 years was most often reported by FPs, IMs, and
OBGs. However, 57% of GPs and one-third of other prac-

Table 2
Test or test combination most often recommended to average-risk patients as a colorectal cancer screening strategy by primary care physicians,
Survey of Colorectal Cancer Screening Practices, 1999 –2000a

Both FOBT and FSb
FOBT aloneb
Either FOBT or FSb
Colonoscopyb
DCBE ⫾ FS
FS alone
Other
Do not recommend CRC screeningc

Family practice
(n ⫽ 423; %, 95% CI)

General practice
(n ⫽ 100; %, 95% CI)

General internal medicine
(n ⫽ 488; %, 95% CI)

Obstetrics/gynecology
(n ⫽ 224; %, 95% CI)

47.1 (42.3–51.9)
27.3 (23.1–31.6)
10.5 (7.6–13.4)
3.4 (1.7–5.2)
1.9 (0.6–3.2)
3.7 (1.9–5.5)
4.1 (2.2–6.0)
1.7 (0.4–2.9)

23.7 (15.3–32.1)
41.1 (31.3–50.9)
13.9 (7.0–20.8)
9.1 (3.4–14.9)
4.0 (0.1–8.0)
0.0 (0.0–0.0)
6.0 (1.3–10.8)
2.1 (0.0–5.1)

55.3 (50.9–59.7)
21.8 (18.1–25.4)
7.2 (4.9–9.5)
5.6 (3.5–7.6)
3.1 (1.5–4.6)
2.5 (1.1–3.8)
2.2 (0.9–3.6)
1.8 (0.6–3.0)

40.1 (33.7–46.6)
30.4 (24.3–36.4)
7.1 (3.7–10.5)
13.0 (8.6–17.4)
0.0 (0.0–0.0)
1.3 (0.0–2.8)
2.7 (0.6–4.9)
4.9 (2.1–7.8)

CI ⫽ confidence interval; FOBT ⫽ fecal occult blood testing; FS ⫽ flexible sigmoidoscopy; CRC ⫽ colorectal cancer.
P ⬍ 0.001.
c
P ⬍ 0.05.
a

b

356

C.N. Klabunde et al. / Preventive Medicine 36 (2003) 352–362

Table 3
Recommendations for and practice of colorectal cancer screening with FOBT, Survey of Colorectal Cancer Screening Practices, 1999 –2000a
Family practice
n

%

CI

General practice

General internal
medicine

Obstetrics/gynecology

n

n

n

Recommend FOBT to asymptomatic, average-risk patients
416
98
95.8
93.8–97.7
Recommended starting age (yr)
398
93
ⱕ35
8.1
5.3–10.8
40
39.5
34.6–44.4
45
5.9
3.6–8.3
50
45.7
40.8–50.7
ⱖ55
0.8
0.0–1.6
Recommended test frequency
Every 1 year
85.6
82.1–89.1
Every 2 years
9.9
7.0–12.9
Every ⬎2 years
3.5
1.7–5.2
No longer recommend FOBT screening when patients reach a certain age
14.3
10.9–17.8
No. of times order or perform FOBT screening during a typical month
423
100
Zero
4.5
2.5–6.5
1–10
20.4
16.5–24.2
11–20
32.8
28.3–37.3
21–40
26.1
21.9–30.3
⬎40
15.3
11.9–18.8
How FOBT is conducted
400
91
Office-basedb
23.7
19.5–27.9
Home-basedc
23.5
19.4–27.7
Both office-based and home-based
49.3
44.4–54.2
Other
2.9
1.3–4.6

%

CI

%

CI

94.9

90.5–99.3

19.6
42.6
3.1
31.5
3.1

%

CI

95.4

93.5–97.3

94.8

91.7–97.8

11.4–27.9
32.3–52.9
0.0–6.7
21.9–41.2
0.0–6.7

6.2
35.1
6.0
51.0
1.8

4.0–8.5
30.7–39.4
3.8–8.2
46.4–55.6
0.5–3.0

9.1
42.4
3.5
43.0
2.0

5.1–13.1
35.5–49.3
0.9–6.0
36.1–49.9
0.0–3.9

76.4
18.3
4.2

67.7–85.2
10.3–26.3
0.1–8.4

87.7
5.9
6.1

84.7–90.7
3.8–8.1
3.9–8.3

91.6
4.9
2.0

87.8–95.5
1.9–7.9
0.0–3.9

8.5

2.7–14.2

17.6

14.1–21.1

2.5

0.0–4.7

8.8
38.5
29.4
13.2
10.1

3.2–14.4
28.8–48.3
20.4–38.5
6.4–19.9
4.1–16.1

6.2
20.5
26.3
23.5
23.1

4.0–8.3
16.9–24.1
22.4–30.2
19.8–27.3
19.4–26.8

10.8
20.2
18.3
22.3
28.5

6.7–14.9
14.9–25.5
13.2–23.4
16.8–27.7
22.5–34.4

29.7
31.0
30.5
5.6

20.1–39.3
21.3–40.7
20.9–40.1
0.7–10.4

23.9
32.2
41.3
1.5

20.0–27.8
28.0–36.5
36.7–45.8
0.4–2.7

64.0
10.5
20.5
3.0

57.3–70.7
6.2–14.8
14.9–26.2
0.6–5.3

479

213

457

202

488

224

456

200

FOBT ⫽ fecal occult blood testing; CI ⫽ confidence interval.
Office-based test involves completing a single FOBT card during a digital rectal examination.
c
Home-based test involves giving or mailing patients a set of FOBT cards to complete at home, using multiple stool specimens.
a

b

titioners indicated that they recommend a frequency of
every 1–3 years for this modality. One-third or fewer respondents said they no longer recommend screening sigmoidoscopy when patients reach a certain age. Sixty-four
percent of FPs and IMs, 77% of GPs, and 95% of OBGs
reported that they do not perform or supervise screening
sigmoidoscopy exams during a typical month. However,
70 – 83% of those who do not perform or supervise screening sigmoidoscopy indicated that they usually refer their
patients to another provider, most often a gastroenterologist,
for the procedure. Of the respondents who reported performing or supervising screening sigmoidoscopy, 50%
(IMs) to 68% (GPs) noted that they perform or supervise the
procedure only one to five times during a typical month.
Fewer respondents recommended colonoscopy and
DCBE as CRC screening modalities than FOBT and sigmoidoscopy (Tables 5 and 6). Thirty-four percent of FPs
and IMs and 49% of GPs and OBGs indicated that they
recommend screening colonoscopy to asymptomatic, average-risk patients; 5% of OBGs, 15% of FPs and IMs, and
24% of GPs recommend DCBE. For both procedures, age

50 was the most commonly mentioned starting age, and
every 1–5 years the most often cited test frequency. From
54% to 63% of respondents reported that they order, perform, or refer patients for screening colonoscopy during a
typical month, whereas 13–32% reported doing so for
screening DCBE. Of those who indicated that they order,
perform, or refer patients for screening colonoscopy or
DCBE, three-fourths or more noted that they do so only one
to five times during a typical month, and the majority
(ⱖ72%) refer their patients to a provider outside of their
practice for the procedure.
Influence of published guidelines on screening practice
and screening status of patients
When asked to rate the influence of published guidelines
on their CRC screening practices, 49% of all respondents
noted the 1997 American Cancer Society (ACS) guidelines
as “very influential” in contrast to the 28% who cited the
1996 U.S. Preventive Services Task Force guidelines, 17%
citing the 1997 G.I. Consortium guidelines, and 15% citing

C.N. Klabunde et al. / Preventive Medicine 36 (2003) 352–362

357

Table 4
Recommendations for and practice of colorectal cancer screening with flexible sigmoidoscopy,
Survey of Colorectal Cancer Screening Practices, 1999 –2000a
Family practice
n

%

General practice
CI

n

%

CI

General internal medicine

Obstetrics/gynecology

n

n

%

CI

%

CI

81.8

78.3–85.3

67.5

61.2–73.8

6.5
4.6
83.9
4.9

4.0–9.0
2.5–6.8
80.3–87.6
2.7–7.0

5.1
2.1
87.1
5.7

1.4–8.8
0.0–4.5
81.5–92.7
1.8–9.6

29.8
15.0
49.8
5.4

25.4–34.5
11.5–18.6
44.0–53.9
3.5–8.1

34.9
13.9
45.8
4.9

26.9–42.6
8.2–19.6
37.6–54.1
1.3–8.4

30.5

25.9–35.1

5.5

1.8–9.3

b

Recommend flexible sigmoidoscopy to asymptomatic, average-risk patients
416
98
479
81.6
77.9–85.4
61.1
51.2–70.9
Recommended starting age (yr)
340
60
392
ⱕ40
8.0
5.1–11.0
20.2
9.7–30.7
45
2.1
0.6–3.7
8.2
1.1–15.2
50
83.3
79.3–87.3
64.8
52.4–77.3
ⱖ55
6.5
3.8–9.1
6.7
0.2–13.3
Recommended test frequency
Every 1–3 years
31.1
26.1–36.1
56.9
43.9–69.7
Every 3–5 years
12.2
9.0–16.0
6.6
0.2–13.0
Every 5 years
50.3
44.1–54.8
33.4
21.1–45.7
Every ⬎ 5 years
6.2
4.0–9.4
3.2
0.0–7.7
No longer recommend sigmoidoscopy screening when patients reach a certain age
24.6
20.0–29.2
9.8
2.1–17.4
No of times perform or supervise screening sigmoidoscopy during a typical month
423
100
488
Zero
63.7
59.1–68.3
77.0
68.6–85.4
1–5
23.1
19.1–27.1
15.0
7.9–22.1
6–10
8.0
5.4–10.6
6.2
1.3–11.0
11–20
3.3
1.6–4.9
0.0
0.0–0.0
⬎20
1.2
0.2–2.3
1.0
0.0–2.9
Physicians not performing or supervising screening sigmoidoscopy who reported they usually refer
test
269
77
310
78.5
73.6–83.4
70.3
59.9–80.8
a
b

213

144

224
63.5
59.2–67.8
95.9
93.3–98.5
17.6
14.3–21.0
1.8
0.0–3.6
10.9
8.1–13.6
0.9
0.0–2.1
4.7
2.8–6.5
0.4
0.0–1.3
2.3
0.9–3.6
0.0
0.0–0.0
their average-risk patients to another provider for this
215
82.6

78.4–86.8

83.2

78.1–88.2

CI ⫽ confidence interval.
P ⬍ 0.001.

other specialty society guidelines (Fig. 2). IMs (29%) and
FPs (35%) were more likely than OBGs (15%) and GPs
(18%) to state that the U.S. Preventive Services Task Force
guidelines were “very influential.” GPs (19%) and IMs
(21%) more often noted the G.I. Consortium guidelines as
“very influential” than did FPs (13%) or OBGs (16%).
One-third of OBGs reported other specialty society guidelines to be “very influential,” compared with 8 –12% of
other practitioners. Of particular note is that 24% of all
respondents indicated that they were unfamiliar with the
1996 U.S. Preventive Services Task Force guidelines on
CRC screening, and 37% reported they were unfamiliar
with the G.I. Consortium guidelines. In contrast, only 8%
stated that they were unfamiliar with the ACS guidelines.
Lastly, less than 20% of respondents reported that threequarters or more of their patients aged 50 and older were
up-to-date with CRC screening as recommended by the
physician (Fig. 3). IMs (22%) were more likely than other
PCPs to indicate this. Moreover, a higher percentage of IMs
(60%) than other specialty groups reported that half or more
of their older patients were up to date with CRC screening.
Consistent with these patterns was the finding that GPs and
OBGs (23%) more often estimated that less than a fourth of

their patients aged 50 and older were up to date than did IMs
(6%) or FPs (15%).

Discussion
Given the low documented rates of CRC screening
among older adults in the United States, there is a critical
need to understand current screening beliefs and practices
from the provider perspective. This survey of PCPs’ CRC
screening attitudes and practices is unique because it covers
all of the CRC screening modalities (FOBT, sigmoidoscopy, colonoscopy, and DCBE) most commonly mentioned
in published guidelines, and provides national estimates of
CRC screening recommendations and practices for the four
primary care specialty groups that are most actively engaged in delivering preventive services to older adults.
A key finding of the study is that several of the CRC
screening recommendations and practices reported by physicians were inconsistent with current guidelines, raising
concerns about the appropriateness and quality of screening
delivery within some primary care practice settings. Although most guidelines stipulate that CRC screening in

358

C.N. Klabunde et al. / Preventive Medicine 36 (2003) 352–362

Table 5
Recommendations for and practice of colorectal cancer screening with colonoscopy, Survey of Colorectal Cancer Screening Practices, 1999 –2000a
Family practice
n

%

General practice
CI

n

%

CI

General internal medicine

Obstetrics/gynecology

n

n

%

CI

%

CI

34.0

29.7–38.2

49.4

42.7–56.2

5.5
3.7
77.8
13.0

2.0–9.1
0.8–6.6
71.3–84.2
7.8–18.3

3.0
1.0
86.1
9.9

0.0–6.4
0.0–2.9
79.2–93.0
4.0–15.8

64.4
35.6
0.0

56.3–71.2
28.2–43.0
0.0–0.0

90.4
6.8
2.9

84.6–96.1
1.9–11.7
0.0–6.1

30.2

23.1–37.3

5.8

1.2–10.3

41.5
44.5
8.4
3.5
0.6

37.2–45.9
40.1–48.9
5.9–10.8
1.9–5.1
0.0–1.3

40.1
42.0
9.9
4.5
2.2

33.7–46.6
35.5–48.5
6.0–13.8
1.7–7.2
0.3–4.2

5.1
19.3
74.5
1.1

2.5–7.6
14.7–24.0
69.4–79.7
0.0–2.3

0.0
14.6
83.1
2.3

0.0–0.0
8.5–20.7
76.6–89.6
0.0–4.9

b

Recommend screening colonoscopy to asymptomatic, average-risk patients
416
98
479
33.6
29.0–38.2
49.1
39.1–59.2
Recommended starting age (yr)
137
48
162
ⱕ40
6.1
2.0–10.3
20.8
8.9–32.6
45
4.6
0.9–8.2
10.6
1.5–19.7
50
75.3
67.9–82.7
58.3
43.8–72.7
ⱖ55
14.0
8.1–19.9
10.4
1.4–19.3
Recommended test frequency
Every 1–5 years
78.8
72.2–85.9
91.8
83.9–99.8
Every 5–10 years
20.3
13.5–27.0
6.2
0.0–13.2
Every ⬎10 years
0.0
0.0–0.0
2.0
0.0–6.1
No longer recommend screening colonoscopy when patients reach a certain age
18.6
12.0–25.1
14.4
4.1–24.6
No. of times order, perform, or refer patients for screening colonoscopy during a typical month
423
100
488
Zero
46.4
41.6–51.1
36.7
27.1–46.3
1–5
43.6
38.9–48.4
48.2
38.2–58.2
6–10
7.3
4.8–9.8
10.2
4.1–16.3
11–20
1.6
0.4–2.8
1.9
0.0–4.6
⬎20
0.6
0.0–1.3
0.0
0.0–0.0
How screening colonoscopy is conducted
223
60
278
Respondent conducts
5.1
2.1–8.0
8.1
1.1–15.1
Refers within practice
14.9
10.3–19.6
12.0
3.4–20.5
Refers outside of practice
77.8
72.4–83.3
79.9
69.5–90.4
Did not specify
2.2
0.3–4.1
0.0
0.0–0.0
a
b

213

105

224

131

CI ⫽ confidence interval.
P ⬍ 0.001.

average-risk adults should begin at age 50 [10 –12,30,31],
one-half of PCPs stated that they recommend initiating
FOBT screening in younger patients, a result paralleling
that of other studies [20,21,27]. This finding supports prior
work showing the limited effect of practice guidelines in
influencing physician behavior [32]. It also may reflect
clinical practices established under older ACS guidelines,
which prior to 1992 designated 40 as the recommended
starting age for FOBT screening [33]. Furthermore, it is
consistent with our finding that PCPs cite the ACS guidelines as the most influential on their screening practices, and
illustrates the difficulty of changing physician practice once
a clinical policy has been adopted. In contrast, PCPs were
more consistent in specifying age 50 as the recommended
starting age for screening with sigmoidoscopy, colonoscopy, and DCBE.
While the recommended FOBT screening interval noted
by most PCPs was consistent with guidelines and evidence,
this was not the case for the non-FOBT screening modalities. Over one-third indicated a test interval of less than 3
years for sigmoidoscopy, and two-thirds or more reported
that they recommended a test interval of every 1–5 years for
colonoscopy and DCBE. In addition, a minority of PCPs
indicated a stopping age for each of the four modalities,

with higher proportions of FPs and IMs compared to GPs
and OBGs reporting upper age limits for stopping screening. The variability in recommended screening intervals for
sigmoidoscopy, colonoscopy, and DCBE may reflect a lack
of specificity concerning test intervals in some of the CRC
screening guidelines [11,31,34] as well as the limited available evidence demonstrating optimal intervals for these modalities. Similar reasons may explain the lack of an upper
age cutoff for recommending screening. Screening averagerisk patients at ages at which they are less likely to benefit
from the procedure or at too-frequent intervals has important implications for the efficient use of resources to perform screening endoscopy and follow-up procedures [35],
and may negatively affect patients’ acceptance of screening.
The finding that one-fourth of FPs, GPs, and IMs and
nearly two-thirds of OBGs use office-based FOBT for the
majority of patients screened with this modality also merits
attention. FOBT efficacy, as established in clinical trials,
was based on administration of home testing kits with stool
samples collected over a 3-day period. The sensitivity of
home-based, 3-day FOBT is likely to be greater than officebased FOBT [12], thus generating a lower false-negative
rate. Although the G.I. Consortium [12] and ACS [10] CRC
screening guidelines are explicit in specifying that home test

C.N. Klabunde et al. / Preventive Medicine 36 (2003) 352–362

359

Table 6
Recommendations for and practice of colorectal cancer screening with double contrast barium enema,
Survey of Colorectal Cancer Screening Practices, 1999 –2000a
Family practice
n

%

General practice
CI

n

%

CI

General internal medicine

Obstetrics/gynecology

n

n

%

CI

%

CI

b

Recommend screening double contrast barium enema to asymptomatic, average-risk patients
416
98
479
14.8
11.3–18.2
24.4
15.8–33.0
15.0
11.8–18.2
Recommended starting age (yr)
61
24
72
ⱕ40
10.3
2.2–18.2
29.4
9.6–49.0
7.0
1.0–13.2
45
1.6
0.0–4.8
4.0
0.0–12.3
5.7
0.1–11.2
50
78.3
67.5–89.0
50.2
28.6–71.8
75.9
65.7–86.1
ⱖ55
9.9
2.2–17.7
16.5
0.5–32.4
11.4
3.8–18.9
Recommended test frequency
Every 1–5 years
88.8
80.8–96.9
87.5
73.3–100.0
79.2
69.5–88.8
Every 5–10 years
9.6
2.1–17.1
8.0
0.0–19.5
20.8
11.2–30.5
No longer recommend screening double contrast barium enema when patients reach a certain age
29.3
17.2–41.5
9.1
0.0–22.3
37.9
26.4–49.5
No. of times order, perform, or refer patients for screening double contrast barium enema during a typical month
423
100
488
Zero
74.7
70.5–78.9
67.7
58.4–77.1
72.0
68.0–75.9
1–5
19.8
16.0–23.6
25.3
16.6–33.9
22.3
18.6–26.0
6–10
3.1
1.4–4.8
3.0
0.0–6.3
3.5
1.9–5.1
11–20
0.9
0.0–1.8
0.0
0.0–0.0
0.4
0.0–1.0
⬎20
0.0
0.0–0.0
0.0
0.0–0.0
0.2
0.0–0.6
How screening double contrast barium enema is conducted
100
28
129
Refers within Practice
11.7
5.4–18.0
7.5
0.0–18.1
21.6
14.5–28.8
Refers outside of Practice
84.3
77.1–91.5
82.0
66.8–97.2
76.1
68.7–83.5
Did not specify
4.1
0.1–8.1
10.5
0.0–22.5
2.3
0.0–4.9
a
b

213
5.2

2.2–8.1

86.6
9.4
0.4
0.4
0.4

82.1–91.1
5.5–13.2
0.0–1.3
0.0–1.3
0.0–1.3

20.9
74.9
4.1

3.4–38.4
56.3–93.6
0.0–12.7

11

224

24

CI ⫽ confidence interval.
P ⬍ 0.001.

kits should be used, other guidelines [11,30,31,34] do not
provide this specific guidance.
Another potential quality issue is the low volume of
screening sigmoidoscopy procedures reported by the 36%
of FPs and IMs and the 23% of GPs who said that they
perform or supervise this procedure in their practices.

Twenty percent of fewer of these physicians indicated that
they perform or supervise screening sigmoidoscopy 11 or
more times during a typical month (i.e., three or more times
per week). One-half of IMs and two-thirds of FPs and GPs
noted that they perform or supervise the procedure only one
to five times per month. Such low volumes may be inadequate for the conduct of an efficient, thorough, safe, accurate, and comfortable screening examination. We are un-

Fig. 2. Self-reported influence of published guidelines on primary care
physicians’ colorectal cancer screening practices, Survey of Colorectal
Cancer Screening Practices, 1999 –2000

Fig. 3. Estimated proportion of patients 50⫹ who are up-to-date with the
CRC screening recommended by the primary care physician, Survey of
Colorectal Cancer Screening Practices, 1999 –2000

360

C.N. Klabunde et al. / Preventive Medicine 36 (2003) 352–362

aware of any published standards for the conduct of
screening sigmoidoscopy in primary care practice. A small
but growing literature on the relationship between physician
procedure volume and patient outcomes suggests that physicians who perform a lower volume of a given procedure
have poorer outcomes than do those who perform high
volumes [36 –38]. This relationship also is reflected in the
federal Mammography Quality Standards Act, which mandates a minimum volume threshold for physicians who
interpret mammograms [39].
Although nearly all PCPs (98%) reported recommending
CRC screening to average-risk patients, the relatively low
volumes of ordering, performing, or referring that they
noted indicate that the majority are not providing screening
to all eligible patients in their practices. Under the assumption that the typical FP cares for 800 patients aged 50 and
older in his or her practice and works 45 weeks per year, this
physician on a weekly basis would need to order 18 FOBTs
(1-year screening interval), perform or refer for four sigmoidoscopies (5-year screening interval), or refer for two
colonoscopies or two DCBE exams (10-year screening interval) to screen most of these patients. Although we recognize that some eligible patients may receive screening
through other providers and the categorical nature of our
data precludes precise estimates, we can infer that at best
one-fourth of PCPs are generating this level of screening
volume. Consistent with these estimates, the majority of
PCPs acknowledged in this study that many of their older,
average-risk patients are not up to date with the CRC
screening recommended by the physician. There are many
possible reasons for the low levels of screening delivery.
Some physicians may not be knowledgeable about current
CRC screening guidelines, and the multiple modalities
available may create uncertainty about how best to screen.
The reluctance of many patients to undergo CRC screening
has been well documented [40], and insurance coverage for
CRC screening may be lacking or inadequate to cover costs
for some modalities [41]. PCPs also may lack office systems
to ensure that CRC screening is systematically offered to
eligible patients [42] or may encounter barriers to referring
patients for more involved screening procedures such as
sigmoidoscopy, colonoscopy, or DCBE.
Strengths of this study include its large sample size, high
response rate, and nationally representative sample. A limitation is that it is based on physician self-reports of screening practice, which may be higher than actual practice
[43,44]. However, as noted, estimates of CRC screening
volume in this study were low, and comparable to national
surveys of the public that document low screening rates
[13–15]. Our data help to elucidate findings from studies of
patients by demonstrating that most PCPs are not systematically recommending or providing CRC screening to a
high proportion of eligible patients. Patient report of a
physician recommendation for screening consistently has
been shown to be one of the strongest predictors of screening utilization [40,45– 47].

This study demonstrates high awareness of CRC screening among PCPs in the United States but also knowledge
gaps about the timing and frequency of screening and suboptimal screening delivery. Results suggest several opportunities for improvement. Efforts to educate the public and
PCPs about the appropriateness and importance of regular
screening for CRC must accelerate. Practice guidelines that
are explicit in providing physicians with information about
when to begin screening, how frequently to screen with a
given modality or combination of modalities, and the techniques to employ to optimize the benefits of screening are
needed. In addition, detailed assessment of the complex
factors that influence physicians’ adherence to screening
guidelines is an important area for future work [32]. Practice
enhancements such as office systems to systematically identify eligible patients and track screening uptake, as well as
involving other members of the primary care practice team
(e.g., physician assistants or nurses) in certain aspects of
CRC screening delivery, also might be beneficial in many
primary care settings [48 –52]. Finally, further research to
investigate the patient, provider, and health system barriers
that contribute to low rates of CRC screening and to identify
innovative approaches to overcoming these barriers is
needed. Ongoing analysis of data from this survey will
enhance our understanding of these barriers and help to
foster improvements in screening delivery.

Acknowledgments
We thank Dr. Lorayn Olson, project director for Abt
Associates, Inc. (Chicago, IL), for survey research work and
Timothy McNeel of Information Management Services,
Inc., for data preparation assistance. Funding support for
this study was provided by the National Cancer Institute
(contract number N01-PC-85169) and the Centers for Disease Control and Prevention (inter-agency agreement number 99FED06571).

References
[1] Mandel JS, Bond JH, Church TR, Snover DC, Bradley GM, Schuman
LM, et al. Reducing mortality from colorectal cancer by screening for
fecal occult blood. Minnesota colon cancer study. N Engl J Med
1993;328:1365–71.
[2] Hardcastle JD, Chamberlain JO, Robinson MH, Moss SM, Amar SS,
Balfour TW, et al. Randomised controlled trial of faecal-occult-blood
screening for colorectal cancer. Lancet 1996;348:1472–7.
[3] Kronborg O, Fenger C, Olsen J, Jorgensen OD, Sondergaard O.
Randomised study of screening for colorectal cancer with faecaloccult-blood test. Lancet 1996;348:1467–71.
[4] Newcomb PA, Norfleet RG, Storer BE, Surawicz TS, Marcus PM.
Screening sigmoidoscopy and colorectal cancer mortality. J Natl
Cancer Inst 1992;84:1572–5.
[5] Selby JV, Friedman GD, Quesenberry CP Jr, Weiss NS. A casecontrol study of screening sigmoidoscopy and mortality from colorectal cancer. N Engl J Med 1992;326:653–7.

C.N. Klabunde et al. / Preventive Medicine 36 (2003) 352–362
[6] Winawer S, Zauber A, O’Brien M, Ho M, Gottlieb L, Sternberg S, et
al. Prevention of colorectal cancer by colonoscopic polypectomy.
N Engl J Med 1993;329:1977– 81.
[7] Muller AD, Sonnenberg A. Prevention of colorectal cancer by flexible endoscopy and polypectomy. A case-control study of 32,702
veterans. Ann Intern Med 1995;123:904 –910.
[8] Hixson LJ, Fennerty MB, Sampliner RE, Garewal HS. Prospective
blinded trial of the colonoscopic miss-rate of large colorectal polyps.
Gastrointest Endosc 1991;37:125–7.
[9] Steine S, Stordahl A, Lunde OC, Loken K, Laerun E. Double-contrast
barium enema versus colonoscopy in the diagnosis of neoplastic
disorders: aspects of decision-making in general practice. Fam Pract
1993;10:288 –91.
[10] Byers T, Levin B, Rothenberger D, Dodd GD, Smith RA. American
Cancer Society guidelines for screening and surveillance for early
detection of colorectal polyps and cancer; update 1997. American
Cancer Society Detection and Treatment Advisory Group on Colorectal Cancer. CA Cancer J Clin 1997;47:154 – 60.
[11] U.S. Preventive Services Task Force, DiGuiseppi C, Atkins D, Woolf
H, editors. Guide to clinical preventive services. second ed. Baltimore, MD: Williams & Wilkins, 1996.
[12] Winawer SJ, Fletcher RH, Miller L, Godlee F, Stolar M, Mulrow C,
et al. Colorectal cancer screening: clinical guidelines and rationale.
Gastoenterology 1997;112:594 – 642.
[13] Ries LAG, Wingo PA, Miller DS, Howe HL, Weir HK, Rosenberg
HM, et al. The annual report to the nation on the status of cancer,
1973–1997, with a special section on colorectal cancer. Cancer 2000;
88:2398 – 424.
[14] Trends in screening for colorectal cancer-United States, 1997 and
1999. Morb Mort Wkly Rep 2001;50:162– 6.
[15] Anderson LM, May DS. Has the use of cervical, breast, and colorectal
cancer screening increased in the United States? Am J Public Health
1995;85:840 –2.
[16] Clasen CM, Vernon SW, Mullen PD, Jackson GL. A survey of
physician beliefs and self-reported practices concerning screening for
early detection of cancer. Soc Sci Med 1994;39:841–9.
[17] Cooper GS, Fortinsky RH, Hapke R, Landefeld CS. Primary
care physician recommendations for colorectal cancer screening:
patient and practitioner factors. Arch Intern Med 1997;157:1946 –
50.
[18] Cooper GS, Fortinsky RH, Hapke R, Landefeld CS. Factors associated with the use of flexible sigmoidoscopy as a screening test for the
detection of colorectal carcinoma by primary care physicians. Cancer
1998;82:1476 – 81.
[19] Czaja R, McFall SL, Warnecke RB, Ford L, Kaluzny AD. Preferences
of community physicians for cancer screening guidelines. Ann Intern
Med 1994;120:602– 8.
[20] Hawley ST, Levin B, Vernon SW. Colorectal cancer screening by
primary care physicians in two medical care organizations. Cancer
Detect Prev 2001;25:309 –18.
[21] Herman CJ, Hoffman RM, Altobelli KK. Variation in recommendations for cancer screening among primary care physicians in New
Mexico. J Community Health 1999;24:253– 67.
[22] Lewis JD, Asch DA, Ginsburg GG, Hoops TC, Kochman ML, Bilker
WB, et al. Primary care physicians’ decisions to perform flexible
sigmoidoscopy. J Gen Intern Med 1999;14:297–302.
[23] Montano DE, Phillips WR, Kasprzyk D. Explaining physician rates of
providing flexible sigmoidoscopy. Cancer Epidemiol Biomarkers
Prev 2000;9:665–9.
[24] Richards C, Klabunde C, O’Malley M. Physicians’ recommendations
for colon cancer screening in women: too much of a good thing?
Am J Prev Med 1998;15:246 –9.
[25] Schoen RE, Weissfeld JL, Kuller LH. Sigmoidoscopy use among
primary care physicians. Prev Med 1995;24:249 –54.

361

[26] Schwarz JS, Lewis CE, Clancy C, Kinosian MS, Radany MH, Koplan
JP. Internists’ practice in health promotion and disease prevention.
Ann Intern Med 1991;114:46 –53.
[27] Sharma VK, Vasudeva R, Howden CW. Colorectal cancer screening
and surveillance practices by primary care physicians: results of a
national survey. Am J Gastroenterol 2000;95:1551– 6.
[28] Survey of Colorectal Cancer Screening Practices in Health Care
Organizations. Federal Register 63, no. 229 (November 30, 1998):
65796.
[29] Starfield B. Primary care: balancing health needs, services, and technology. New York: Oxford University Press, 1998, p. 154.
[30] American College of Obstetricians and Gynecologists. Revised
recommendations on colorectal cancer screening. Http://www.acog.com.
[31] American Academy of Family Physicians. Periodic health examinations summary of AAFP policy recommendations and age charts.
Http://www.aafp.com.
[32] Cabana MD, Rand CS, Powe NR, Wu AW, Wilson MH, Abboud PA,
et al. Why don’t physicians follow clinical practice guidelines? A
framework for improvement. JAMA 1999;283:1458 – 65.
[33] Levin B, Murphy G. Revision in American Cancer Society recommendations for the early detection of colorectal cancer. CA Cancer
J Clin 1992;42:296 –9.
[34] American College of Physicians—American Society of Internal Medicine. Screening for breast, colorectal, and prostate cancer. Http://
www.acponline.org.
[35] Rex DK, Lierberman DA. Feasibility of colonoscopy screening: discussion of issues and recommendations regarding implementation.
Gastrointest Endosc 2001;54:662–7.
[36] Hughes RG, Hunt SS, Luft HS. Effects of surgeon volume and
hospital volume on quality of care in hospitals. Med Care 1987;25:
489 –503.
[37] Jollis JG, Peterson ED, Nelson CL, Stafford JA, Delong ER, Muhlbaier LH, et al. Relationship between physician and hospital coronary
angioplasty volume and outcome in elderly patients. Circulation
1997;95:2485–91.
[38] Harmon JW, Tang DG, Gordon TA, Bowman HM, Choti MA, Kaufman HS, et al. Hospital volume can serve as a surrogate for surgeon
volume for achieving excellent outcomes in colorectal resection. Ann
Surg 1999;230:404 –13.
[39] Food and Drug Administration. Mammography Quality Standards
Act (MQSA) final regulations. Http://www.fda.gov/cdrh/mammography.
[40] Vernon SW. Participation in colorectal cancer screening: a review.
J Natl Cancer Inst 1997;89:1406 –22.
[41] Lewis JD, Asch DA. Barriers to office-based screening sigmoidoscopy: does reimbursement cover costs? Ann Intern Med 1999;130:
525–30.
[42] Frame PS. Implementing clinical preventive medicine: time to fish or
cut bait. J Am Board Fam Pract 2000;13:84 –5.
[43] Montano DE, Phillips WR. Cancer screening by primary care physicians: a comparison of rates obtained from physician self-report,
patient survey, and chart audit. Am J Public Health 1995;85:795–
800.
[44] Peterson SK, Vernon SW. A review of patient and physician adherence to colorectal cancer screening guidelines. Semin Colon Rectal
Surg 2000;11:58 –72.
[45] National Cancer Institute. Screening mammography: a missed clinical opportunity? Results of the NCI Breast Cancer Screening Consortium and National Health Interview Survey studies. JAMA 1990;
264:54 – 8.
[46] Mandelblatt J, Kanetsy PA. Effectiveness of interventions to enhance physician screening for breast cancer. J Fam Pract 1995;40:
162–71.

362

C.N. Klabunde et al. / Preventive Medicine 36 (2003) 352–362

[47] Taplin SH, Urban N, Taylor VM, Savarino J. Conflicting national
recommendations and the use of screening mammography: does the
physician’s recommendation matter? J Am Board Fam Pract 1997;
10:88 –95.
[48] Leininger LS, Finn L, Dickey L, Dietrich AJ, Foxhall L, Garr D, et al.
An office system for organizing preventive services: a report by the
American Cancer Society Advisory Group on Preventive Health Care
Reminder Systems. Arch Fam Med 1996;5:108 –15.
[49] Lieberman DA, Ghormley JM. Physician assistants in gastroenterology: should they perform endoscopy? Am J Gastroenterol 1992;87:
940 – 43.

[50] Cash B, Schoenfeld PS, Ransohoff D. Licensure, use, and training of
paramedical personnel to perform screening flexible sigmoidoscopy.
Gastrointest Endosc 1999;49:163–9.
[51] Shaheen NJ, Crosby MA, O’Malley MS, Murray SC, Sandler RS,
Galanko JA, et al. The practices and attitudes of primary care nurse
practitioners and physician assistants with respect to colorectal cancer
screening. Am J Gastroenterol 2000;95:3259 – 65.
[52] Thompson NJ, Boyko EJ, Dominitz JA, Belcher DW, Chesebro BB,
Stephens LM, et al. A randomized controlled trial of a clinic-based
support staff intervention to increase the rate of fecal occult blood test
ordering. Prev Med 2000;30:244 –51.