JAMA & ARCHIVES
Arch Fam Med
SEARCH
GO TO ADVANCED SEARCH
HOME  PAST ISSUES  TOPIC COLLECTIONS  CME  PHYSICIAN JOBS  CONTACT US  HELP
Institution: STANFORD Univ Med Center  | My Account | E-mail Alerts | Access Rights | Sign In
  Vol. 9 No. 4, April 2000 TABLE OF CONTENTS
  Archives
  •  Online Features
  Original Contribution
 This Article
 •Abstract
 •PDF
 •Send to a friend
 • Save in My Folder
 •Save to citation manager
 •Permissions
 Citing Articles
 •Citation map
 •Citing articles on HighWire
 •Citing articles on Web of Science (60)
 •Contact me when this article is cited
 Related Content
 •Related article
 •Similar articles in this journal

Correlates of Physicians' Prevention-Related Practices

Findings From the Women Physicians' Health Study

Erica Frank, MD, MPH; Richard Rothenberg, MD, MPH; Charles Lewis, MD; Brooke F. Belodoff, MS

Arch Fam Med. 2000;9:359-367.

ABSTRACT

Background  Determinants of physicians' prevention-related counseling and screening practices are not well understood. Such determinants are worth knowing because we can then intervene on malleable variables and produce physicians with stronger prevention-related skills. Of the few such variables that have been examined, they have typically only been studied in univariate analyses or in small or otherwise limited populations and have been especially sparsely studied in women physicians.

Objective  To explore the effect of potential counseling- and screening-related variables in 4501 respondents to the Women Physicians' Health Study, a questionnaire-based study of a representative sample of US women MDs.

Results  Being a primary care practitioner and practicing a related health habit oneself were significantly correlated with reporting counseling and screening for all prevention-related variables examined. Current attempts to improve a related habit oneself, ethnicity, region, practice site, and amount of continuing medical education were sometimes significantly correlated with counseling and screening; most other variables studied (ie, personal health status, a personal or family history of disease, control of work environment, and career satisfaction) were rarely significantly correlated.

Conclusions  Being a primary care practitioner and having related healthy habits oneself were the most significant correlates of US women physicians' self-reported prevention-related counseling and screening practices. These findings suggest potential new directions for physician training.



INTRODUCTION
 Jump to Section
 •Top
 •Introduction
 •Materials and methods
 •Results
 •Comment
 •Author information
 •References

ALTHOUGH SOME contributors to physicians' prevention-related counseling and screening practices have been studied, it is not well known what determines whether a physician practices clinical prevention. It is worth understanding these determinants because knowing which characteristics are associated with such habits can guide interventions on malleable variables and help produce better preventionists.1

The few published studies2-9 of correlates of physicians' prevention-related counseling and screening practices have found that variables such as personal health behaviors, attempts to change personal health behaviors, perceived personal health, age, specialty, training as a subspecialist, confidence in counseling skills, and attitudes toward counseling may be correlated with likelihood to counsel patients about prevention. However, these variables have only been examined in small or otherwise limited populations (eg, in single specialties or small geographic regions), have typically only been studied in univariate analyses, and have been especially sparsely studied in women physicians. In addition, many variables that may affect counseling and screening have not been examined in either male or female physicians, such as a personal or family history of a relevant disease (ie, the effect of a family history of cardiovascular disease or lung cancer on smoking counseling), the reported amount of counseling training, practice location (rural, suburban, or urban), ethnicity, hours worked, and control of work environment. This article explores directly, and through models, the effect of such variables on 14 self-reported prevention-related counseling and screening variables in 4501 respondents to the Women Physicians' Health Study (WPHS), a questionnaire-based study of a representative sample of US women MDs.


MATERIALS AND METHODS
 Jump to Section
 •Top
 •Introduction
 •Materials and methods
 •Results
 •Comment
 •Author information
 •References

GENERAL WPHS METHODS

The WPHS design and the fundamental characteristics of the WPHS population have been more fully described elsewhere.10-12 WPHS is a cross-sectional survey of women physicians aged 30 to 70 years, not in residency training, and residing in the United States in 1993. A probability sample was selected from the American Medical Association Physician Masterfile, stratified by decade of graduation (1950-1989), with oversampling of earlier decades. Of potential respondents, an estimated 23% were ineligible to participate because their addresses were wrong or they were men, deceased, living out of the country, or interns or residents. Our response rate is 59% of physicians eligible to participate; nonrespondents were less likely than were respondents to be board certified, but they did not consistently or substantively differ on other tested measures. Based on these findings, we weighted the data by decade of graduation (to adjust for our stratified sampling scheme) and by decade-specific response rate and board certification status (to adjust for our identified response bias). Using these weights allows us to make inference to the entire population of women physicians who graduated from medical school between 1950 and 1989.

COUNSELING AND SCREENING QUESTIONS

Our primary outcome variables are responses to questions about screening and counseling. Physicians were asked how often, "considering your typical patient," they discussed or performed screening for cholesterol; blood pressure; colorectal cancer (if patient is >50 years old); skin cancer and sunscreen use; human immunodeficiency virus (HIV) risks/testing; influenza vaccine (if >65 years old); nutrition; weight; exercise; smoking cessation (if a smoker); alcohol use; and (for female patients) clinical breast examinations, mammograms (if 50-75 years old), and hormone replacement therapy (HRT) (if postmenopausal). These criteria were based on national recommendations at the time of the survey.13 Response options included "every visit," "every <=1 year," "every >1 to 2 years," "every >2 to 3 years," "every >3 to 5 years," "only at initial visit," "only if clinically indicated," or "never." The outcome variable for each individual counseling item was dichotomized. Physicians who reported counseling at every visit or at least once a year were considered "high counselors" and are described as "counseling at least once a year." Also, a 14-point summary counseling score was developed for which physicians received 1 point for each counseling question on which she reported counseling at every visit or at least once a year.

Physicians were asked the "relevance to your practice" and "self-confidence in counseling" for each of the screening practices considered. We compared responses of "highly" vs "somewhat," "not very," or "not at all." Physicians were also asked the "amount of training in counseling" regarding the dependent variables in question, and we compared responses of "extensive" vs "some," "little," or "none." Family medicine, general practice, general internal medicine, and public health physicians were considered primary care, and obstetricians/gynecologists were considered a separate category. All other specialties were considered non–primary care physicians. Because of their relatively minimal clinical contact with adult patients for whom these prevention-related behaviors are recommended, pediatricians, pathologists, radiologists, and anyone spending less than 5 hours per week in clinical practice were excluded from these analyses.

A physician's family history of a high cholesterol level (yes or no) was tested for relationships with the patient cholesterol screening question, a family history of hypertension was compared with the patient blood pressure screening question, colon cancer was compared with the colorectal cancer question, skin malignancy with skin cancer and sunscreen, HIV with HIV, obesity with weight, cigarette smoking with cigarette counseling, alcohol abuse or dependence with alcohol use counseling, and breast cancer or other breast disorder with clinical breast examination and mammogram. The same relationships were also tested for physicians' personal histories of disease vs the related patient screening or counseling question. Respondents were also asked whether they were currently trying to lose weight, change their eating habits, exercise more, or drink less alcohol. Other independent variables include age, ethnicity, general health rating, region, practice site (urban, suburban, or rural), practice type, control of work environment, career satisfaction, and continuing medical education (CME).

DATA ANALYSIS

The SUDAAN statistical program (Research Triangle Institute, Research Triangle Park, NC) was used to perform {chi}2 and t tests (unpaired, 2-tailed) to determine whether counseling was related to personal and professional characteristics. Logistic regression in SUDAAN was used to model counseling practices as a function of several personal and professional characteristics. A modified version of backwards selection for logistic regression was used, including goodness-of-fit tests for the final models using a modification of the Hosmer and Lemeshow technique.14

All analyses were weighted to make inference to the entire population, and SEs and significance testing were performed using SUDAAN analyses that recognized the sample design. To help control for type I errors, for univariate tests, P<.01 was considered significant; unless otherwise noted, only those variables significant at P<.01 are discussed. To determine which variables remained in the final logistic regression model, the criterion was P<.10 for the Wald F test. A 95% confidence interval on the odds ratio is given for any regression coefficient with P<.05.


RESULTS
 Jump to Section
 •Top
 •Introduction
 •Materials and methods
 •Results
 •Comment
 •Author information
 •References

PERSONAL AND PROFESSIONAL CHARACTERISTICS

Overall, physicians reported intervening (ie, screening or counseling) at least once a year for just under half of the 14 clinical outcomes (a total score of 5.6 of a possible total of 14 in our summary score) (Table 1). There was no significant difference by age in the summary score (although age was positively related to the cholesterol (P<.001) and nutrition (P<.01) variables (data not shown). Differences by ethnicity (data not shown) were most pronounced for cholesterol, nutrition, weight, and alcohol-related clinical outcomes, with Asians (42.1%) having the highest frequency of counseling yearly regarding cholesterol, Hispanics and blacks counseling more regarding nutrition (54.6% and 50.2%, respectively) and weight (59.3% and 59.6%, respectively), and Hispanics and Asians (52.9% and 53.6%, respectively) counseling more about alcohol (P<.001 for all) (data not shown). No differences significant at P<.01 were found for any individual outcome variables in stratifications by personal health status or personal or family histories of related diseases (data not shown). However, for 4 variables there was a modest (P<.10) positive relationship: a personal history of obesity for patient weight counseling (P=.03) and family histories of hypertension for blood pressure counseling or screening (P=.03), cigarette smoking for smoking cessation counseling (P=.04), and osteoporosis for HRT counseling (P=.08).


View this table:
[in this window]
[in a new window]
Table 1. Personal Characteristics and Their Relation With Physicians' General Likelihood to Provide Prevention-Related Counseling or Screening at Least Once a Year


Primary care practitioners (PCPs) and obstetricians/gynecologists were more likely than other specialists to report counseling or screening (Table 2); PCPs were most likely to do so (P=.009 vs obstetricians/gynecologists; data not shown). As judged by our summary score, PCPs were most likely to perform these practices when PCPs were considered in aggregate and also when each primary care specialty was compared separately with other non–primary care specialties for all 14 prevention practices (Table 2). However, when individual types of prevention-related counseling and screening practices were considered, different specialties' performances varied substantially; this is shown in more detail in a later table.


View this table:
[in this window]
[in a new window]
Table 2. Professional Characteristics and Their Relation With Physicians' General Likelihood to Provide Prevention-Related Counseling or Screening at Least Once a Year


Physicians working less than 40 clinical hours per week were significantly less likely to report discussing or performing prevention-related screening overall (Table 2) and specifically less likely to report discussing or screening for blood pressure, weight, exercise, smoking, clinician breast examination, or postmenopausal HRT (data not shown). Physicians working in groups and in government counseled or screened most frequently, and hospital-based physicians counseled or screened least frequently (Table 2). All 14 outcomes were highly significantly related to perceived practice relevance, self-confidence in counseling, and amount of counseling training about the specific topic (P<.001 for all; data not shown). No significant (P<.01) differences for any individual variable were seen by practice site (urban, suburban, or rural), control of work environment, or career satisfaction (data not shown).

PERSONAL PREVENTION HABITS

There was a strong and consistent relationship between personal prevention habits and related reported patient counseling and screening practices (Table 3). Physicians who ate less fat were more likely to counsel about or screen their patients for cholesterol; those who drank less were more likely to counsel regarding alcohol use; those more likely to receive flu vaccinations were more likely to administer them as well; frequent performers of breast self-examination, to perform patient breast examinations; sunscreen users, to discuss sunscreen use and skin cancer prevention; and postmenopausal HRT users, to discuss HRT. Nonsmokers were more likely to report counseling patients regarding smoking cessation (though the P value was only .02, likely due to the small number, 4%, of smoking physicians). Those who ate less fat (P=.08) or at least 5 fruits and vegetables per day (P=.046) were modestly more likely to discuss nutrition with their patients (data not shown).


View this table:
[in this window]
[in a new window]
Table 3. Physicians' Personal Habits and the Percentage Counseling or Screening Patients at Least Once a Year About Related Prevention Items*


Those who were themselves screened more recently for cholesterol were more likely to report screening patients for cholesterol, and those who had more recent clinician-performed skin examinations were more likely to report promoting skin cancer prevention to patients (Table 3). Personal blood pressure measurement, colon and breast cancer screening, and HIV testing habits had no significant effect on related patient practices. Physicians who were trying to change their personal eating habits were more likely to discuss cholesterol or colon cancer (although not nutrition in general); those trying to exercise more were more likely to discuss exercise and weight (data not shown). We also tested (data not shown) whether there was a relationship between a physician's general likelihood to check herself for the screening items in Table 2 and her general likelihood to report counseling or screening on the related items (with summary scores given for all self-screening and patient screening). The correlation between the 2 summary scores was only 5% (P=.06).

PRACTICE TYPE

In general, PCPs and medical subspecialists were most likely to report performing patient prevention counseling and screening (Table 4). Family physicians, general practitioners, general internists (for all variables except HIV, where general internists performed at the average), and public health physicians (for all variables except HRT and cholesterol) were more likely than the average woman physician to discuss or screen for all preventive practices examined; medical subspecialists performed better than average for most preventive practices surveyed. Dermatologists and ophthalmologists were generally least likely to counsel or screen. Anesthesiologists reported intervening less than average for all preventive practices examined, as did dermatologists (for all practices except skin cancer), emergency physicians, neurologists, ophthalmologists, psychiatrists, general surgeons (for all practices except breast and colorectal cancer), and surgical subspecialists. Obstetricians/gynecologists were especially likely to report intervening on breast cancer and to discuss HRT; they also performed more counseling or screening than average for blood pressure, colorectal cancer, HIV, tobacco, nutrition, weight, and exercise and less than average for cholesterol, skin cancer, and flu vaccinations.


View this table:
[in this window]
[in a new window]
Table 4. Physicians' Specialty Type, Specialty, and Practice Type and the Percentage Counseling or Screening at Least Once a Year by Prevention Type*


MULTIVARIATE MODELS

We found that being a PCP was a significant predictor of reporting counseling or screening for all 14 practices examined (Table 5). Primary care practitioners also had more training in and self-confidence about performing counseling and screening and perceived these interventions as being more relevant to their practices. Self-confidence about, perceived relevance of, and training regarding counseling and screening also typically remained significant correlates of counseling and screening when added to the models including all physicians (data not shown). Being an obstetrician/gynecologist was a significant positive predictor of intervention for 11 of 14 models (all except cholesterol, skin cancer, and flu vaccine). Practicing a particular health habit oneself, by doing primary prevention or screening for that behavior, was a frequent correlate of counseling and screening. Ethnicity, region, changing a related personal habit, CME, and practice site were occasionally significant correlates of intervention. Most other variables, including control of work environment and career satisfaction, were rarely significantly correlated with our intervention outcomes.


View this table:
[in this window]
[in a new window]
Table 5. Models for Significant Correlates of Physicians' Counseling at Least Once a Year About Prevention, by Prevention Type*



COMMENT
 Jump to Section
 •Top
 •Introduction
 •Materials and methods
 •Results
 •Comment
 •Author information
 •References

We sought characteristics of physicians who were more frequent preventionists and found a few personal and professional characteristics that consistently predicted reporting counseling and screening.

PERSONAL CHARACTERISTICS

Our findings confirm intuition: health care providers report that they preach what they practice. Our univariate analyses showed that personal related health habits affected physicians' likelihood to report counseling or screening patients regarding cholesterol level, alcohol use, smoking cessation, flu vaccination, breast and skin cancer prevention and detection, and HRT. In most cases these relationships were statistically significant (P<.01), and they retained significance when adjusted for all variables in our models. We also found that physicians who were trying to improve their personal health habits were somewhat more likely to discuss related habits with patients. There was, however, only a weak relationship between a physician's general likelihood to have been recently screened for diseases and her general likelihood to report counseling or screening. This suggests that practicing a particular habit oneself may only increase the likelihood that a physician will discuss that particular healthy habit with her patients, not that she will be more likely to counsel about prevention generally, although physicians' attitudes toward personal screening habits may be different from their attitudes toward other personal health practices.12

Previous analyses of physician behavior have also shown that physicians who themselves have healthy personal habits are more likely to encourage their patients to adopt such habits; however, previous studies have typically been limited by size, specialty, region, and nonadjustment for the multiple variables that may confound the effect of personal health practices. They have also, like this study, almost all concentrated on one gender (most previous work has studied primarily men physicians). However, their findings have been in the same direction as ours: physicians with good personal health habits or a desire to improve their own health are more likely to report counseling or screening patients regarding prevention.3-5,9, 15 Specifically, these smaller trials have shown that regular exercisers were more likely to counsel their patients on exercise habits,5, 7 nonsmokers to counsel on smoking,5, 16-17 seat belt users to counsel on seat belt use,5 low-fat food eaters and those who have been screened for cholesterol to counsel on cholesterol,18 and physicians who were attempting weight loss to counsel their patients on weight loss.15

Why might a physician with healthy personal practices be more likely to report counseling or screening patients about related practices? Such physicians could be more interested in prevention and may therefore extend their personal interests into the clinical context. They might feel more comfortable addressing barriers to, benefits of, and costs of behavior change or be more knowledgeable about the physiological and other effects of behavioral change. Physicians with high personal investments in healthy habits may not only be more active preventionists1 but may also be more effective.19 Personal health practices also may interact with perceived counseling relevance, self-confidence, and training, 3 areas that were highly related to counseling frequency. Alternatively, these results may reflect a reporting bias; physicians who report healthier personal practices may also report more counseling, regardless of actual personal or clinical practices.

No other personal characteristics correlated frequently with reporting screening and counseling, although nonwhite ethnicity and region of residence were correlated with a few outcomes. There are social values related to nutrition and alcohol use, and it is not surprising that differing societies, both ethnic and geographic, might differ in considering these appropriate clinical topics. We hypothesized that age would be a negative predictor of counseling and screening, as has been demonstrated in several previous studies.4, 6-7,16, 20 However, these studies were primarily univariate analyses and mainly studied men; Wells and colleagues,2 in a study including 50 women and 405 men internists, also found that age is not a significant indicator of counseling, especially when controlling for specialty and other variables. We also tested (for the first time that we know of in the literature) whether having a personal or family history of a particular disease would encourage intervening on related risk factors. This was often true, although the effects were usually not large or statistically significant and were inconsistent when examined univariately and when modeled. Having had such a history may affect one's clinical practice, but its effect does not seem to be large.

There are several important limitations to this study, especially our use of self-reported outcomes21 and the fact that more counseling (ie, counseling or screening at every visit or at least once a year, our major outcome) may not always mean better or more effective counseling. In addition, there is considerable and consistent evidence that women physicians are more likely to offer preventive services than are men. This is true for counseling and screening for female-specific disorders,22-26 but it is also so for scheduling general prevention visits27 and for providing assorted types of prevention-related8, 28 and other29 counseling and testing, including smoking counseling.30 Even when data are adjusted for age and specialty, women are more likely to counsel regarding prevention than are men.8 Our other major study limitation is response rate. However, the few other large (N>500) studies of US physicians (primarily or exclusively men) conducted in the past 20 years have used similar methods for determining eligibility and have reported similar response rates: 43%,31 47%,32 59%,33 63%,34 and 75%.4

PROFESSIONAL CHARACTERISTICS

We found that one of the most consistent predictors of physicians' reporting screening and counseling is being a PCP or obstetrician/gynecologist; this is also true of the previous literature.9 Spending more time on CME also increased the likelihood that one would screen or counsel on some variables; no other professional characteristics were consistently significant influences.

It is understandable that some specialties would screen and counsel more than others; PCPs have clinical prompts and patient relationships that promote such interventions, although most specialists have opportunities to intervene on selected risk. It would be useful for individual physicians to select a few areas on which they will consistently intervene; collectively this could have powerful positive effects.

We also found that amount of training on a topic and self-confidence about and perceived relevance of a topic are significant predictors of reported prevention-related clinical practices. This is intuitive and supported by previous literature: others have also found that physicians who believe in the importance of counseling,3 in their personal effectiveness in patient behavior change,5, 16 and in their having a greater knowledge of risk factors16 are more likely to counsel and screen their patients. Physicians without prevention-related self-confidence are also less optimistic about patients' abilities to change.35 This has important implications for educating the physician workforce because it suggests that training programs should attend not only to the science presented but also to the attitudes that are created.

We found other professional characteristics not to be significantly correlated with reporting counseling and screening. Some of these, such as board certification3-4,7 and practice type,4, 6 have been found in other analyses to be positive, although not powerful, predictors. We also found that control of work environment and career satisfaction were neither univariately significant for any screening or counseling type nor often significant in our models. We hypothesized that physicians would derive career satisfaction from prevention-related activities and that those with more work control would be more active preventionists. However, perhaps those with less work control, such as those working for health maintenance organizations, are expected to perform some prevention-related activities as part of their jobs.

There are other barriers to counseling and screening that we have not studied. Some are practical,36 such as a lack of physician time7, 35, 37 or reimbursement.7, 25, 35, 37 There may also be cognitive barriers: Gemson and Elinson7 reported that 58% of physicians cited unclear recommendations as a reason for not doing more preventive counseling.

VARIATIONS WITHIN PREVENTION-RELATED OUTCOMES

Rosso and colleagues16 reported that nonsmoking physicians are more likely to counsel patients regarding cigarette use. Other previously described predictors include physicians' being younger,19 being female,5, 8 being PCPs,16 having a positive perception of counseling's efficacy,16, 25 and being salaried.25 We found that being a nonsmoker, a PCP or an obstetrician/gynecologist, or participating in more CME were the most significant modeled predictors of advising patients not to smoke.

Regarding alcohol, previously determined screening and counseling predictors include a higher knowledge level and fewer years since medical school graduation.16 However, drinking less alcohol oneself or a personal or family history of alcoholism had not been shown to increase counseling or had not yet been examined. We found that being a PCP, being Asian vs white, residing in the Pacific or Mountain region vs the Central states, and trying to change one's alcohol drinking habits were correlated with being more likely to counsel regarding drinking alcohol.

Regarding weight, Lewis and colleagues15 reported that physicians who were trying to lose weight counseled more on weight loss to their patients; we did not find this to be a significant contributor to counseling about weight. Lewis and colleagues also reported that physicians' thinking that they are overweight negatively affects counseling about smoking, weight, exercise, and alcohol.15

Previous studies include a personal exercise habit,7 female gender,5 and younger age as predictors of exercise counseling.38 We found that being a PCP and trying to improve one's own exercise habits predicted exercise counseling.

Physicians have previously been found to be more likely to counsel or screen for breast cancer if they are female,22-25 younger,16 or salaried; engage in more CME; have fewer practice barriers25; or perceive mammograms to be effective.16 Women physicians are also more likely to teach patients breast self-examination techniques.4, 16 We found that patient mammography counseling was correlated with being a PCP or an obstetrician/gynecologist, having a personal history of breast cancer, and having mammograms more frequently oneself. Counseling patients about breast self-examination was correlated with being a PCP or an obstetrician/gynecologist and more frequently performing breast self-examination oneself.

In conclusion, we found, analyzed for the first time with a large population and in a relatively comprehensive model, that being a PCP and having related healthy habits oneself were the most important correlates of US women physicians' self-reported counseling of patients about prevention. This was consistently true across a broad array of clinical prevention-related habits. These findings have substantial policy implications, as they could provide important guidance for creating more active preventionists.1

Although the relationship between being a PCP and an active preventionist has been demonstrated in several previous univariate and a few multivariate analyses,9 this is one of the first demonstrations that physicians' personal health habits are more strongly and consistently correlated with related prevention activities than are many other personal or professional variables. Such a finding suggests action: if we value disease prevention, and if physicians' personal health practices are consistent predictors of their likelihood to be more active preventionists, we ought to try to cultivate healthy physicians (in undergraduate and graduate education and in CME). We are now testing in a multi–medical school study called "Healthy Doc—Healthy Patient" whether we can actively cultivate healthy physicians and whether such deliberately cultivated healthy physicians are also more active preventionists. In the interim, however, those trying to encourage physicians to provide more clinical preventive services should consider adding messages, experiences, and environments that promote physicians' personal health.


AUTHOR INFORMATION
 Jump to Section
 •Top
 •Introduction
 •Materials and methods
 •Results
 •Comment
 •Author information
 •References

Accepted for publication October 26, 1999.

This work was supported by Wyeth-Ayerst Laboratories, Philadelphia, Pa; the American Medical Association Foundation, Chicago, Ill; the American Heart Association, Dallas, Tex; institutional national research service award 5T32-HL-07034 from the National Institutes of Health, National Heart, Lung, and Blood Institute, Bethesda, Md; the Emory Medical Care Foundation, Atlanta, Ga; and the Ulrich and Ruth Frank Foundation for International Health, Newtown, Pa.

We wish to thank Sally McNagny, MD, MPH, and Lisa Elon, MPH, for their expertise regarding hormone replacement therapy analyses.

Reprints: Erica Frank, MD, MPH, Department of Family and Preventive Medicine, Emory University School of Medicine, 69 Butler St, Atlanta, GA 30303 (e-mail: efrank{at}fpm.eushc.org).

From the Department of Family and Preventive Medicine, School of Medicine (Drs Frank and Rothenberg), and Rollins School of Public Health (Drs Frank and Rothenberg and Ms Belodoff), Emory University, Atlanta, Ga; and the Center for Health Promotion and Disease Prevention, University of California, Los Angeles (Dr Lewis).


REFERENCES
 Jump to Section
 •Top
 •Introduction
 •Materials and methods
 •Results
 •Comment
 •Author information
 •References

1. Frank E. Osler was wrong: you are a preventionist [see comments]. Am J Prev Med. 1991;7:128. ISI | PUBMED
2. Wells KB, Lewis CE, Leake B, Schleiter MK, Brook RH. The practices of general and subspecialty internists in counseling about smoking and exercise. Am J Public Health. 1986;76:1009-1013. FREE FULL TEXT
3. Wells KB, Lewis CE, Leake B, Ware JE Jr. Do physicians preach what they practice? a study of physicians' health habits and counseling practices. JAMA. 1984;252:2846-2848. FREE FULL TEXT
4. Schwartz JS, Lewis CE, Clancy C, Kinosian MS, Radany MH, Koplan JP. Internists' practices in health promotion and disease prevention: a survey. Ann Intern Med. 1991;114:46-53.
5. Lewis CE, Clancy C, Leake B, Schwartz JS. The counseling practices of internists [see comments]. Ann Intern Med. 1991;114:54-58.
6. Radecki SE, Mendenhall RC. Patient counseling by primary care physicians: results of a nationwide survey. Patient Educ Counsel. 1986;8:165-177. FULL TEXT | ISI | PUBMED
7. Gemson DH, Elinson J. Prevention in primary care: variability in physician practice patterns in New York City. Am J Prev Med. 1986;2:226-234. ISI | PUBMED
8. Frank E, Harvey L. Prevention advice rates of women and men physicians. Arch Fam Med. 1996;5:215-219. FREE FULL TEXT
9. Frank E, Kunovich-Frieze T. Physicians' prevention counseling behaviors: current status and future directions. Prev Med. 1995;24:543-545. FULL TEXT | ISI | PUBMED
10. Frank E. The Women Physicians' Health Study: background, objectives, and methods. J Am Med Womens Assoc. 1995;50:64-66.
11. Frank E, Rothenberg R, Brown WV, Maibach H. Basic demographic and professional characteristics of U.S. women physicians. West J Med. 1997;166:179-184. ISI | PUBMED
12. Frank E, Brogan DJ, Mokdad AH, Simoes EJ, Kahn HS, Greenberg RS. Health-related behaviors of US women physicians versus other US women. Arch Intern Med. 1998;158:342-348. FREE FULL TEXT
13. US Preventive Services Task Force. Guide to Clinical Preventive Services. Baltimore, Md: Williams & Wilkins; 1989.
14. Hosmer DW Jr, Lemeshow S. Applied Logistic Regression. New York, NY: John Wiley & Sons Inc; 1989.
15. Lewis CE, Wells KB, Ware J. A model for predicting the counseling practices of physicians. J Gen Intern Med. 1986;1:14-19 [published correction appears in J Gen Intern Med. 1986;1:154].
16. Rosso S, Battista RN, Segnan N, Williams JI, Suissa S, Ponti A. Determinants of preventive practices of general practitioners in Torino, Italy. Am J Prev Med. 1992;8:339-344. PUBMED
17. Ockene JK, Kristeller J, Goldberg R, et al. Increasing the efficacy of physician-delivered smoking interventions: a randomized clinical trial [see comments]. J Gen Intern Med. 1991;6:1-8. ISI | PUBMED
18. Hyman DJ, Maibach EW, Flora JA, Fortmann SP. Cholesterol treatment practices of primary care physicians. Public Health Rep. 1992;107:441-448. ISI | PUBMED
19. Frank E, Breyan J, Elon L. Physician disclosure of healthy personal behaviors improves credibility and ability to motivate. Arch Fam Med. 2000;9:287-290. FREE FULL TEXT
20. Fortmann SP, Sallis JF, Magnus PM, Farquhar JW. Attitudes and practices of physicians regarding hypertension and smoking: the Stanford Five City Project. Prev Med. 1985;14:70-80. FULL TEXT | ISI | PUBMED
21. Stange K, Flocke S, Goodwin M. Opportunistic preventive services delivery: are time limitations and patient satisfaction barriers? J Fam Pract. 1998;46:419-424. ISI | PUBMED
22. Lurie N, Slater J, McGovern P, Ekstrum J, Quam L, Margolis K. Preventive care for women: does the sex of the physician matter [see comments]? N Engl J Med. 1994;329:478-482.
23. Hall JA, Palmer RH, Orav EJ, Hargraves JL, Wright EA, Louis TA. Performance quality, gender, and professional role: a study of physicians and nonphysicians in 16 ambulatory care practices. Med Care. 1990;28:489-501. ISI | PUBMED
24. Levy S, Dowling P, Boult L, Monroe A, McQuade W. The effect of physician and patient gender on preventive medicine practices in patients older than fifty [see comments]. Fam Med. 1992;24:58-61. PUBMED
25. Battista RN, Williams JI, MacFarlane LA. Determinants of primary medical practice in adult cancer prevention. Med Care. 1986;24:216-224. FULL TEXT | ISI | PUBMED
26. Margolis K, Lurie N, Mink P, McGovern P. Doctor-patient communication predicts breast and cervical cancer screening [abstract]. J Gen Intern Med. 1996;11(suppl 1):8126.
27. Osborn EH, Bird JA, McPhee SJ, Rodnick JE, Fordham D. Cancer screening by primary care physicians: can we explain the differences [see comments]? J Fam Pract. 1991;32:465-471. ISI | PUBMED
28. Sox C, Dietrich A, Tosteson T, Winchell C, Labaree C. Periodic health examinations and the provision of cancer prevention services. Arch Fam Med. 1997;6:223-230. FREE FULL TEXT
29. Roter D, Lipkin M Jr, Korsgaard A. Sex differences on patients' and physicians' communication during primary care medical visits. Med Care. 1991;29:1083-1093. ISI | PUBMED
30. Counseling practices of primary-care physicians—North Carolina, 1991. MMWR Morb Mortal Wkly Rep. 1992;41:562-569.
31. Buring JE, Hennekens CH. Cost and efficiency in clinical trials: the U.S. Physicians' Health Study. Stat Med. 1990;9:29-33. FULL TEXT | ISI | PUBMED
32. Bensing J. Doctor-patient communication and the quality of care. Soc Sci Med. 1991;32:1301-1310.
33. Hughes PH, Brandenburg N, Baldwin DC Jr, Storr CL, Williams KM, Sheehan DV. Prevalence of substance use among US physicians . JAMA. 1992;267:2333-2339 [published correction appears in JAMA. 1992;268:2518] [see comments]. FREE FULL TEXT
34. Shea S, Gemson DH, Mossel P. Management of high blood cholesterol by primary care physicians: diffusion of the National Cholesterol Education Program Adult Treatment Panel guidelines. J Gen Intern Med. 1990;5:327-334. ISI | PUBMED
35. Orleans CT, George LK, Houpt JL, Brodie KH. Health promotion in primary care: a survey of US family practitioners. Prev Med. 1985;14:636-647. FULL TEXT | ISI | PUBMED
36. Kushner RF. Barriers to providing nutrition counseling by physicians: a survey of primary care practitioners. Prev Med. 1995;24:546-552. FULL TEXT | ISI | PUBMED
37. Morris PD, Morris ER. Family practice residents' compliance with preventive medicine recommendations. Am J Prev Med. 1988;4:161-165. ISI | PUBMED
38. Wechsler H, Levine S, Idelson RK, Rohman M, Taylor JO. The physician's role in health promotion: a survey of primary-care practitioners. N Engl J Med. 1983;308:97-100. ISI | PUBMED

RELATED ARTICLE

The Archives of Family Medicine Continuing Medical Education Program
Arch Fam Med. 2000;9(4):375-377.
FULL TEXT  


THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES

Preventive, Lifestyle, and Personal Health Behaviors Among Physicians
Bazargan et al.
Acad. Psychiatry 2009;33:289-295.
ABSTRACT | FULL TEXT  

Physical activity habits of doctors and medical students influence their counselling practices
Lobelo et al.
Br. J. Sports. Med. 2009;43:89-92.
ABSTRACT | FULL TEXT  

Medical Student Health Promotion: The Increasing Role of Medical Schools
Estabrook
Acad. Psychiatry 2008;32:65-68.
ABSTRACT | FULL TEXT  

Themed Review: Clinical Interventions to Promote Physical Activity in Youth
Meriwether et al.
AMERICAN JOURNAL OF LIFESTYLE MEDICINE 2008;2:7-25.
ABSTRACT  

Predictors of nutrition counseling behaviors and attitudes in US medical students.
Spencer et al.
Am. J. Clin. Nutr. 2006;84:655-662.
ABSTRACT | FULL TEXT  

Physician health and wellness.
Taub et al.
Occup Med (Lond) 2006;56:77-82.
ABSTRACT | FULL TEXT  

Antibiotic Prescribing for Children With Nasopharyngitis (Common Colds), Upper Respiratory Infections, and Bronchitis Who Have Health-Professional Parents
Huang et al.
Pediatrics 2005;116:826-832.
ABSTRACT | FULL TEXT  

Counseling Parents and Children on Sun Protection: A National Survey of Pediatricians
Balk et al.
Pediatrics 2004;114:1056-1064.
ABSTRACT | FULL TEXT  

Physician Health and Patient Care
Frank
JAMA 2004;291:637-637.
FULL TEXT  

Breastfeeding Education, Treatment, and Referrals by Female Physicians
Arthur et al.
J Hum Lact 2003;19:303-309.
ABSTRACT  

Smoking Cessation Counseling: Training and Practice Among Women Pediatricians
Malarcher et al.
CLIN PEDIATR 2002;41:341-349.
ABSTRACT  

Self-reported Health, Illness, and Self-care Among Finnish Physicians: A National Survey
Toyry et al.
Arch Fam Med 2000;9:1079-1085.
ABSTRACT | FULL TEXT  

Personal and Clinical Skin Cancer Prevention Practices of US Women Physicians
Saraiya et al.
Arch Dermatol 2000;136:633-642.
ABSTRACT | FULL TEXT  




HOME | CURRENT ISSUE | PAST ISSUES | TOPIC COLLECTIONS | CME | PHYSICIAN JOBS | HELP
CONDITIONS OF USE | PRIVACY POLICY | CONTACT US | SITE MAP
 
© 2000 American Medical Association. All Rights Reserved.