| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Health Services and Economics Branch, Applied Research Program, National
Cancer Institute, Executive Plaza North, Rm. 4005, 6130 Executive Blvd.,
Bethesda, MD 20892-7344.
2 Diagnostic Radiology Department, Clinical Center, National Institutes of
Health, Bldg. 10, Rm. 1C-660, Bethesda, MD 20892-1182.
3 Statistical Research and Applications Branch, Surveillance Research Program,
National Cancer Institute, Rm. 5047, 6116 Executive Blvd., Bethesda, MD
20892.
Received March 7, 2002;
accepted after revision May 30, 2002.
Supported by the National Cancer Institute (contract N01-PC-85169) and the
Centers for Disease Control and Prevention (interagency agreement
99FED06571).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We surveyed a nationally representative sample of diagnostic radiologists. Of 381 eligible radiologists, 312 (82%) responded. Descriptive statistics and chi-square tests were used to assess radiologists' opinions about double-contrast barium enema volume and capacity and to compare radiologists' beliefs about colorectal cancer screening with those of primary care physicians. Logistic regression was used to identify characteristics of radiologists who receive referrals for or perform a higher volume of screening double-contrast barium enema and of those who expect the volume of double-contrast barium enemas to increase.
RESULTS. Seventy-five percent of radiologists said that double-contrast barium enema is a "very effective" colorectal cancer screening procedure compared with 33% of primary care physicians. Although 86% of radiologists reported performing one or more screening double-contrast barium enema procedures during a typical month, only 27% indicated that they did so 11 or more times. Fifteen percent of radiologists said that their double-contrast barium enema volume had increased over the past 3 years, and 50% expect an increase over the next 3 years. Only 8% said that the capacity of facilities and personnel to meet the demand for double-contrast barium enemas in their geographic area of practice is inadequate. Geographic region and belief in double-contrast barium enema efficacy were predictors of double-contrast barium enema volume and referrals.
CONCLUSION. Most diagnostic radiologists perform colorectal cancer screening with double-contrast barium enema, but procedure volumes are modest. Because primary care physicians view double-contrast barium enema less positively than do radiologists, radiologists' expectations for an increased volume of double-contrast barium enemas over the next few years may not be realized.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Unlike sigmoidoscopy, double-contrast barium enema has the advantage of permitting examination of the entire colon. Compared with sigmoidoscopy and colonoscopy, double-contrast barium enema is believed to have lower rates of such major complications as bowel perforation, hemorrhage, and death [3]. Simulations using mathematic models also have shown double-contrast barium enema to be a cost-effective colorectal cancer screening procedure [4,5,6]. Data on the performance characteristics, effectiveness, and adverse events associated with double-contrast barium enema are sparse, however. No controlled trial has been conducted to directly assess the impact of double-contrast barium enema on colorectal cancer mortality. Support for double-contrast barium enema as a colorectal cancer screening modality is predicated on indirect evidence from other modalities showing that detection and removal of polyps and early-stage cancers reduce colorectal cancer incidence and mortality. The sensitivity of double-contrast barium enema for detecting polyps 1 cm or larger has been estimated to range from 50% to 90% [3, 7, 8]. Sensitivity declines with decreasing lesion size, however, and double-contrast barium enema appears to be less sensitive than colonoscopy for identifying small lesions [9]. Although a multidisciplinary panel charged with establishing a set of clinical guidelines for colorectal cancer screening concluded that "performance of double-contrast barium enema is sufficient to detect the majority of clinically important lesions" [3], and Medicare covers double-contrast barium enema as a colorectal cancer screening modality [10], the sparse and inconsistent data about this procedure may limit its clinical appeal as a screening strategy [11].
Despite growing recognition of the importance of screening asymptomatic average-risk patients for colorectal cancer and documentation of low rates of screening for this disease among adults 50 years old and older in the United States [12, 13], few nationally representative analyses of health care providers' colorectal cancer screening knowledge, attitudes, and practices have been conducted. Furthermore, no prior study of diagnostic radiologists' colorectal cancer screening opinions and practices has been undertaken.
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 survey is a comprehensive study of primary care and specialty physicians and health plan medical directors that is designed to obtain nationally representative data on how colorectal cancer screening is being conducted in the United States and to identify barriers to delivery of screening in community practice [14]. In this report, we describe diagnostic radiologists' beliefs about the efficacy of colorectal cancer screening modalities in current use, their self-reported colorectal cancer screening activities in their clinical practices, and their opinions about current and future capacity for colorectal cancer screening with double-contrast barium enema. In addition, because primary care physicians are an important referral source for colorectal cancer screening procedures, we compare radiologists' opinions about colorectal cancer test effectiveness, recommended screening strategy, and published guidelines with those of the primary care physicians who participated in this national survey.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Survey Methodology
Sampled radiologists 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 several medical societies and the United States Surgeon General,
and a postcard with a stamped return envelope on which they were asked to
verify their specialty and status as a practicing physician and to indicate
their preferred mode of response to the survey (i.e., mail, fax, telephone, or
secure Internet Web site). Radiologists 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, and a $50 honorarium check.
Approximately 6 weeks later, radiologists who did not respond to the advance
mailing were sent a follow-up mailing containing the mail version of the
questionnaire and a $50 honorarium check. A second follow-up questionnaire
mailing was sent by express mail to eligible nonrespondents in February 2000.
Telephone follow-up of nonrespondents was undertaken in March and early April
2000.
The survey took approximately 15 min to complete and was organized into two sections: cancer screening beliefs and practices, and practice and other characteristics. Items inquiring about diagnostic radiologists' colorectal cancer screening beliefs, opinions, and practices were specific to asymptomatic, average-risk patients and encompassed the four colorectal cancer screening modalities most commonly mentioned in published guidelines: fecal occult blood test, sigmoidoscopy, colonoscopy, and double-contrast barium enema, although the focus of the questionnaire was primarily on double-contrast barium enema. Definitions of screening and diagnostic follow-up were provided. Radiologists were asked to rate the effectiveness of each screening modality using a three-point Likert scale that included the categories "very effective," "somewhat effective," and "not effective." Radiologists also were asked to identify the test or test combination with which they believe most average-risk adults should be screened for colorectal cancer and to rate the influence of published guidelines on their practice of colorectal cancer screening using the categories "very influential," "somewhat influential," and "not influential." A separate set of items inquired about whether radiologists routinely receive patient referrals from other providers to perform colorectal cancer screening procedures and the frequencies with which radiologists order or perform fecal occult blood testing and perform or supervise screening and diagnostic double-contrast barium enemas. Finally, radiologists were asked whether the volume of colorectal cancer screening or diagnostic procedures they perform or supervise had increased, stayed the same, or decreased over the past 3 years; whether they expect demand for double-contrast barium enema in the geographic area in which they practice to increase, stay the same, or decrease over the next 3 years; and whether they believe the capacity of facilities and personnel for performing double-contrast barium enemas in their geographic area of practice is more than enough, just about right, or inadequate to meet anticipated demand over the next 3 years.
Linkage to the Area Resource File
To obtain socioeconomic and health personnel data for each respondent's
county of medical practice, we extracted several variables from the February
1999 version of the Area Resource File
[16] and merged them with the
diagnostic radiologists' survey data set using the state and county Federal
Information Processing Standards code
[17] as the linkage variable.
Variables obtained from the Area Resource File included whether the
respondent's county of practice had been designated in whole or in part a
primary care health professions shortage area; the proportion of the county
population self-identified as white; the proportion of the county's population
of adults 25 years old and older holding a college degree; the number of
patient care gastroenterologists; the number of patient care diagnostic
radiologists; and an estimate of the county population for 1997.
Survey of Primary Care Physicians
The sampling and survey methodologies for the survey of primary care
physicians were analogous to those described for the diagnostic radiologists.
Eligible respondents for the primary care physician survey were general
practitioners, family practitioners, general internists, and
obstetrician—gynecologists. We selected a systematic, stratified random
sample of 2098 primary care physicians, with physician specialties as the
sampling strata and selection proportional to the specialty's representation
in the total United States physician population. This sample size was selected
to provide point estimates of population proportions within ±3% at a
95% confidence interval. Of the initial sample, 380 primary care physicians
subsequently were determined to be ineligible, leaving a final sample of 1718
primary care physicians.
A separate questionnaire was developed for primary care physicians. This questionnaire was lengthier than the diagnostic radiologists' questionnaire and was organized into four sections: cancer screening beliefs and practices, attitudes toward and training in colorectal cancer screening, recommendations for and use of colorectal cancer screening modalities, and practice and other characteristics. Items asking primary care physicians to rate the effectiveness of colorectal cancer screening modalities and the influence of published guidelines on their practice of colorectal cancer screening were identical to those previously described for the diagnostic radiologists' survey. Primary care physicians also were asked to identify the test or test combination that they most often recommend to their asymptomatic, average-risk patients for colorectal cancer screening. Additional detail on the primary care physician survey is available elsewhere [18].
Statistical Analysis
A sample weight that accounts for the probability of selection into the
sample and nonresponse was assigned to each respondent. We used descriptive
statistics to assess diagnostic radiologists' opinions about current and
future double-contrast barium enema volume and capacity. We used descriptive
statistics and contingency tables with chisquare tests to compare diagnostic
radiologists' opinions about colorectal cancer screening strategies, test
effectiveness, and the influence of published guidelines with those of primary
care physicians. We used multivariate logistic regression analysis to identify
characteristics of diagnostic radiologists who reported performing a high
volume of screening double-contrast barium enemas, expectations for performing
a greater volume of double-contrast barium enemas, and receiving referrals
from other providers to perform screening double-contrast barium enemas.
Because of sample size limitations, we used a sequential approach to fitting each of the logistic regression models. Independent variables were grouped into three categories: physician characteristics, practice characteristics, and environmental characteristics. In each of these categories, the bivariate association of each independent variable with the dependent variable of interest was assessed using the chi-square test for categoric variables and t tests of differences in means for continuous variables. Independent variables that showed a statistically significant association at a p value of less than 0.20 were then entered into separate logistic regression models examining physician characteristics, practice characteristics, and environmental characteristics with the dependent variable of interest. Independent variables that were statistically significant in these models at a p value of less than 0.20 were included in three final logistic regression models to identify important predictors of diagnostic radiologists who reported a high volume of screening double-contrast barium enemas, expectations for performing an increased volume of double-contrast barium enemas, and receiving referrals to perform screening double-contrast barium enemas. Sampling weights were applied in all statistical analyses to permit generalization of the results to the United States population of practicing diagnostic radiologists.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Perceived Test Effectiveness and Recommended Screening Strategy
Three quarters or more of diagnostic radiologists reported the belief that
colonoscopy and double-contrast barium enema are "very effective"
screening procedures for reducing colorectal cancer mortality in patients 50
years old and older (Fig. 1).
Fewer than 25% of radiologists, however, rated flexible sigmoidoscopy and
fecal occult blood testing as "very effective." Primary care
physicians also were most likely to rate colonoscopy as "very
effective." Although their assessment of this modality was similar to
that of radiologists, primary care physicians differed significantly from
radiologists in their ratings of the effectiveness of the other three
modalities. Differences between radiologists and primary care physicians were
most striking for double-contrast barium enema, with 75% of radiologists but
only 33% of primary care physicians rating this modality as "very
effective." Differences between the two physician groups were smaller
but still statistically significant for flexible sigmoidoscopy and fecal
occult blood testing. Primary care physicians were more likely than
radiologists to rate these modalities as "very effective."
|
When asked to specify the testing approach with which most average-risk adults should be screened for colorectal cancer, nearly half the radiologists indicated that double-contrast barium enema, either alone or combined with flexible sigmoidoscopy, should be used (Table 2). In contrast, 85% of primary care physicians reported that they most often recommend fecal occult blood testing and flexible sigmoidoscopy, either alone or in combination, to their average-risk patients, and only about 2% said that they most often recommend double-contrast barium enema, either alone or combined with flexible sigmoidoscopy. Furthermore, despite the high ratings of colonoscopy effectiveness by both physician groups, fewer than 10% of radiologists responded that colonoscopy was the testing approach with which most average-risk adults should be screened, and fewer than 10% of primary care physicians reported that colonoscopy was the screening approach that they most often recommend to their average-risk patients.
|
Influence of Published Guidelines on Screening Practice
When asked to rate the influence of published guidelines on their
colorectal cancer screening practices, 30% of diagnostic radiologists reported
that the 1997 American Cancer Society guidelines
[1] were "very
influential" compared with other specialty society guidelines (25%), the
1997 G. I. (Gastrointestinal) Consortium guidelines
[3] (10%), or the 1996 United
States Preventive Services Task Force guidelines
[2] (5%)
(Fig. 2). Primary care
physicians also were more likely to note American Cancer Society guidelines as
being "very influential" compared with these other guidelines.
However, a greater proportion of primary care physicians rated each type of
guideline as "very influential" than did radiologists, with the
exception of the guidelines of other specialty societies. For all guideline
types, differences between the two physician groups were statistically
significant. Between one quarter and one half of radiologists and between one
quarter and one third of primary care physicians indicated that they were
unfamiliar with the United States Preventive Services Task Force, G. I.
(Gastrointestinal) Consortium, or other specialty society guidelines. In
contrast, only 17% of radiologists and 8% of primary care physicians reported
that they were unfamiliar with American Cancer Society guidelines.
|
Opinions About Current and Future Double-Contrast Barium Enema Volume
and Capacity
Radiologists reported performing or supervising a smaller volume of
screening than diagnostic double-contrast barium enema procedures
(Fig. 3). Nearly half the
respondents indicated that they perform or supervise five or fewer screening
double-contrast barium enema procedures during a typical month, whereas 40%
said that they perform or supervise five or fewer diagnostic double-contrast
barium enema procedures during a typical month. Furthermore, 45% reported that
they typically perform or supervise six to 20 diagnostic double-contrast
barium enema procedures per month, compared with 39% who said they perform or
supervise this volume of screening procedures. Only for the highest volume
category were the proportions of radiologists equivalent, with 11% reporting
that they perform or supervise more than 20 screening double-contrast barium
enema procedures per month and 10% indicating that they perform or supervise
more than 20 diagnostic procedures per month. Although 15% of radiologists
indicated that the volume of double-contrast barium enema procedures they
performed or supervised over the past 3 years had increased, 54% reported that
their procedure volume had stayed the same, and 23% said that it had decreased
(Fig. 4). Of the minority of
radiologists who said that the volume of double-contrast barium enema
procedures they performed or supervised had increased, a greater volume of
referrals for diagnostic follow-up procedures was mentioned more often as the
reason for the increase (91%) than were more referrals for screening (84%) or
for surveillance (79%) procedures.
|
|
One half of radiologists, however, reported that they expect the demand for colorectal cancer screening and diagnostic procedures with double-contrast barium enema in their geographic area of practice to increase during the next 3 years, with 9% indicating that they expect an increase in procedure volume of more than 20% per year (Fig. 5). Forty-four percent said they expect demand to stay about the same, and 5% reported the belief that it would decrease. Finally, more than 90% of respondents said that the capacity of facilities and personnel for performing colorectal cancer screening and diagnostic procedures with double-contrast barium enema in their geographic area was either just about right (44%) or more than adequate (47%) to meet anticipated demand; only 8% reported the belief that capacity was inadequate (Fig. 6).
|
|
Predictors of Screening Double-Contrast Barium Enema Volume and
Referrals
With the exception of board certification and employment in a health
maintenance organization, each of the physician, practice, and environmental
characteristics displayed in Table
1 was examined as a potential predictor of performing a high
volume of screening double-contrast barium enemas, the expectation that the
volume will increase in the radiologist's geographic area of practice, and
receiving referrals from other providers to perform screening double-contrast
barium enema. Relatively few of these characteristics were statistically
significant in multivariate modeling, however. In the final logistic
regression model comparing diagnostic radiologists who reported performing 11
or more screening double-contrast barium enema procedures per month with
radiologists reporting a lower volume of this procedure, having referral
arrangements with other providers to perform screening examinations, having a
high proportion of patients covered by managed care, and practicing in the
North Central census region were significant predictors
(Table 3) at a p value
of less than 0.05. In the final model comparing radiologists who indicated
that they expect demand for double-contrast barium enemas to increase in their
geographic area of practice during the next 3 years with those not expecting
an increase in demand, being a graduate of a United States medical school and
believing that most average-risk adults should be screened for colorectal
cancer with double-contrast barium enema were significant predictors of the
opinion that demand will increase (Table
4). Finally, being a full or part owner of a physician practice,
practicing in the North Central census region, and believing that most
average-risk adults should be screened for colorectal cancer with
double-contrast barium enema were significant predictors of diagnostic
radiologists who reported receiving referrals from other providers to perform
screening double-contrast barium enemas
(Table 5).
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The results of our study show that most diagnostic radiologists in the United States view colorectal cancer screening with double-contrast barium enema as a very effective strategy for reducing colorectal cancer mortality in average-risk older adults, and most reported performing this procedure in their clinical practices. However, only slightly more than 25% said that they perform the procedure 11 or more times during a typical month (i.e., roughly three or more procedures per week). The low double-contrast barium enema volumes reported by many diagnostic radiologists in this survey may have implications for procedure quality, especially given the findings of some studies suggesting that physicians who perform low volumes of a particular procedure have less favorable outcomes than do those who perform high volumes [28,29,30,31,32]. The volume—outcome relationship for radiologic procedures has received little attention and is an area that may merit further investigation.
Although the double-contrast barium enema procedure volumes reported in our study were modest, one half of diagnostic radiologists expressed their belief that the volume of double-contrast barium enemas will increase in their geographic area of practice over the next few years. This expectation may derive from recent and ongoing public health efforts to improve colorectal cancer screening rates, which are considerably lower than screening rates for cervical, breast, and prostate cancer [33]. Whether these expectations will be realized is debatable, given that primary care physicians view double-contrast barium enema less positively than do radiologists, and primary care physicians indicated that they most often recommend fecal occult blood testing and sigmoidoscopy as colorectal cancer screening modalities to their average-risk patients. However, even though radiologists reported primary care physicians to be their main referral source for screening double-contrast barium enema procedures, limited sigmoidoscopy and colonoscopy capacity in certain geographic areas may result in greater double-contrast barium enema volume for some radiologists. In addition, as previously mentioned, some patients will choose double-contrast barium enema when presented with a menu of options for colorectal cancer screening [26]. Also, increased use of screening colonoscopy may lead to an increase in the number of double-contrast barium enema procedures because not all colonoscopic examinations successfully reach the cecum, and patients with incomplete colonoscopy are often referred for a barium enema procedure [34]. The results of our logistic regression analyses suggest that screening double-contrast barium enema may be more sustainable when there are existing referral relationships for diagnostic double-contrast barium enema, when health care payers place a premium on cost, and in locations in which access to alternative modalities may be difficult.
Fewer than 10% of radiologists indicated the belief that the capacity of facilities and personnel to meet the demand for double-contrast barium enema procedures over the next few years in their geographic area of practice was inadequate. This finding, coupled with the low and static or decreasing double-contrast barium enema procedure volumes reported in our study, lends support to prior work showing that colonoscopy has been replacing barium enema as an initial colorectal examination since the mid 1980s [35]. Our survey did not ascertain diagnostic radiologists' levels of training or interest in performing an increased volume of screening double-contrast barium enema examinations, however. With new colorectal cancer screening technologies such as CT colonography on the horizon [36], radiologists who view double-contrast barium enema as an older and potentially soon-to-be-obsolete technology may have little interest in investing, or incentive to invest, more time and resources in the procedure.
To date, fecal occult blood testing is the only colorectal cancer screening modality that is supported by evidence from randomized, controlled trials [3]. The effectiveness of double-contrast barium enema as a colorectal cancer screening strategy has received very little study. Moreover, data on the use of double-contrast barium enema at the population level is extremely limited because the surveys such as the National Health Interview Survey [37] that are the basis for monitoring screening utilization nationally do not include double-contrast barium enema as a screening modality. Although it is possible to study colorectal cancer screening with double-contrast barium enema using Medicare claims, this type of analysis would exclude individuals not covered under the Medicare program, as well as individuals who are enrolled in health maintenance organizations.
Our study is limited by the self-reported nature of its data. However, to our knowledge, it is the first nationally representative assessment of diagnostic radiologists' colorectal cancer screening opinions and practices. With increasing emphasis on colorectal cancer control as a public health priority, the need for comprehensive data to monitor and understand colorectal cancer screening practices is acute. Diagnostic radiologists potentially have an important role to play in colorectal cancer control. Their challenge is to obtain the data necessary to document that the technologies within their scope of practice—double-contrast barium enema and, in the future, CT colonography—can be appropriately applied in a screening context.
Acknowledgments
We thank Lorayn Olson, project director for Abt Associates (Chicago, IL),
for survey research work and Timothy McNeel of Information Management Services
for data preparation assistance.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  W. Ajaj, T. C. Lauenstein, G. Pelster, G. Holtmann, S. G. Ruehm, J. F. Debatin, and S. C. Goehde MR Colonography in Patients with Incomplete Conventional Colonoscopy Radiology, February 1, 2005; 234(2): 452 - 459. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
