Physician Self-Report of Comfort and Skill in Providing Preventive Care to Patients of the Opposite Sex
Nicole Lurie, MD, MSPH;
Karen Margolis, MD, MPH;
Paul G. McGovern, PhD;
Pamela Mink, MPH
Arch Fam Med. 1998;7:134-137.
Background Cancer screening in adults is a fundamental responsibility of primary care physicians. Previous studies have reported that when the patient and the physician are of the same sex, screening rates are higher; previous studies have also reported that trainees believe that they are poorly prepared for and are uncomfortable while performing sex-sensitive examinations.
Objectives To compare the level of skill and comfort of male physicians with that of female physicians in conducting breast and prostate examinations, obtaining Papanicolaou smears, and obtaining a sexual history from men and women and to compare ratings of comfort and skill of internists with those of family physicians.
Methods We surveyed 389 internists and family physicians from a large health plan in Minnesota. All female physicians and a random sample of male physicians were surveyed. Respondents rated their level of skill and comfort in conducting breast and prostate examinations, obtaining Papanicolaou smears, and obtaining a sexual history from a man and a woman. We compared the responses of male and female internists with those of male and female family physicians and computed odds ratios (ORs), adjusting for physician age and specialty. We also compared the ratings of comfort and skill of internists with those of family physicians.
Results Compared with male physicians, female physicians were more likely to report being "very comfortable" performing breast examinations (OR, 7.55; 95% confidence interval [CI], 3.06-18.65), obtaining Papanicolaou smears (OR, 13.80; 95% CI, 3.16-60.20), and obtaining sexual histories from women (OR, 3.99; 95% CI, 2.33-6.84). Conversely, female physicians were less likely to report being very comfortable obtaining sexual histories from men (OR, 0.52; 95% CI, 0.33-0.82). Only 6% and 13% of female family physicians and internists, respectively, believed that their skill in performing a prostate examination was excellent compared with 49% and 37% of male family physicians and internists, respectively (OR, 0.12; 95% CI, 0.06-0.22).
Conclusions Internal medicine and family practice physicians report significantly less comfort and lower levels of skill when performing sex-related examinations or obtaining a sexual history from patients of the opposite sex. Interventions to improve skill and comfort level should be considered.
THE PROVISION of preventive care is widely recognized as a fundamental responsibility of primary care physicians, yet suboptimal rates continue to be documented for almost all preventive maneuvers. In the case of cancer screening in adults, several investigators have reported that when the physician and the patient are of the same sex, the rates of cancer screening are higher.1-5 The reasons for this phenomenon are poorly understood. Important factors may relate to differences between male and female physicians in their perceptions of the importance of, and their comfort and skill in, breast, pelvic, and rectal or prostate examinations. Differences in screening rates related to physician sex may also stem from the self-selection of patients who vary in their motivation to obtain preventive care from male or female providers or from patient reluctance to be examined by physicians of the opposite sex.6
Because most physicians care for male and female patients, the preparation of primary care providers for performing sex-sensitive examinations well is essential. In a recent report,7 most primary care physicians in Massachusetts reported being "very prepared" to counsel patients about smoking, alcohol use, and exercise; however, less than half were very prepared to counsel patients about illicit drug use, diet, or stress. Further, only a few physicians were positive about their potential to change patient behavior in these domains. The Massachusetts study did not report on other established elements of preventive care, such as cancer screening in adults.7
In several studies, trainees have reported poor preparation and resultant discomfort in conducting pelvic, breast, or prostate examinations.8-11 Whether such discomfort persists into practice is unknown. In this article, we describe the level of comfort reported by practicing internists and family physicians when performing breast, pelvic, and rectal or prostate examinations and when obtaining a sexual history from patients of the same and opposite sex.
PARTICIPANTS AND METHODS
This study is derived from a project designed to determine why patients of female physicians have higher rates of breast and cervical cancer screening than those of male physicians.6
We identified all internists, family physicians, and obstetrician-gynecologists who were providers in Medica Health Plan, a large, independent practice association model health plan in the MinneapolisSt Paul, Minn, area, and who had seen at least 5 female patients during 1992. We included all female physicians in our sample, and an age- (within 10 years) and specialty-matched sample of male physicians. Because there were few female physicians older than 55 years, we also sampled male physicians in this age group. We obtained information on physician age from the databases of the Minnesota Medical Association, Minneapolis, and determined their sex based on the first name. When the physicians' age or sex could not be easily ascertained by these methods, we called their offices to obtain the information.1 Data on physician specialty were obtained from records of the health plan and confirmed by physician self-report at the time of the survey. The overall response rate to this survey was 92%. This analysis is limited to responses from internists and family physicians.
We surveyed all physicians by telephone. In addition to obtaining demographic data, we asked physicians to rate their own skill in the following domains: interpersonal care, performing a clinical breast examination, obtaining a Papanicolaou smear, and performing a rectal or prostate examination. We also asked how "awkward or uncomfortable" they were while performing these examinations and while obtaining a sexual history from a man and a woman. Skill was rated on a 5-point Likert-type scale, ranging from "excellent" to "poor"; comfort was rated on a 5-point scale, ranging from "very comfortable" to "very uncomfortable."
The pattern of distribution of physicians' ratings of comfort and skill indicated that respondents did not use the entire range of response options in that few respondents rated their skills in the lowest category. Based on the distributions, we dichotomized the responses by ratings of skill as excellent vs all else and by ratings of comfort as very comfortable vs all else. We age adjusted the ratings of male and female physicians regarding their perceived comfort and skill and used 2 methods to compare them.
With the use of logistic regression, we computed odds ratios (ORs) for each rating of skill as excellent and comfort as very comfortable for female physicians compared with male physicians, controlling for age and physician specialty. Similarly, in the computation of ORs for specialty differences, we controlled for physician age and sex. In separate analyses, we also controlled for the proportion of time the physician spent doing primary care and whether the physician had a subspecialty. Because the results did not change substantively, we only provide analyses controlling for physician age and specialty.
We received responses from 389 internists and family physicians. The response rates were similar across physician sex and specialty. Forty internists and 5 family physicians reported doing no primary care and were excluded from further analyses, leaving 344 physician surveys for study. One hundred fifty-four female physicians remained: 57 were internists, and 97 were family physicians. Of the 190 remaining male physicians, 60 were internists and 130 were family physicians.
The distribution of the physician sample by age, sex, and specialty is as follows:
Family physicians were more likely to be women than were internists.
Table 1 provides the self-reported ratings of comfort and skill of male and female physicians, as well as ORs for female vs male physicians reporting skill as excellent or being very comfortable performing the preventive maneuver, controlling for physician specialty and age. Female physicians report substantially more comfort than male physicians in examining women or obtaining a sexual history from women and substantially less comfort in examining men or obtaining a sexual history from men. Less than half of the female physicians surveyed report being very comfortable obtaining a sexual history from a man. The findings for perceived skill in performing clinical breast and prostate examinations and in obtaining Papanicolaou smears mirror the ratings of comfort; in aggregate, male and female physicians similarly rated their interpersonal skill. No more than three quarters of physicians rated their skills in any domain as excellent. Only 6% of female family physicians and 13% of female internists rated their prostate examination skills as excellent, whereas 49% of male family physicians and 37% of male internists rated these skills as excellent. The interaction of physician sex and specialty is statistically significant for prostate examination skills.
Physician Ratings of Comfort and Skill by Sex and Specialty*
Table 1 also provides the age-adjusted, self-reported ratings of comfort and skill of internists and family physicians, as well as the ORs comparing them, controlling for physician sex and age. Compared with family physicians, internists reported similar levels of comfort in performing a clinical breast examination, performing a prostate examination, and obtaining a sexual history. Male internists reported less comfort than male family physicians in obtaining Papanicolaou smears, and female internists reported more comfort in performing a prostate examination than female family physicians.
Also, internists were significantly less likely than family physicians to rate their skill in obtaining a Papanicolaou smear as excellent, but they rated their skills in performing breast and prostate examinations similarly. As was the case when comparing male with female physicians, fewer than half of internists or family physicians rated their skills in performing a breast or prostate examination as excellent.
For differences by age, after controlling for physician sex, physicians in the oldest quartile are more likely than those in the youngest quartile to self-report excellent skills in performing a clinical breast examination (51% vs 36%; OR, 2.1; 95% confidence interval [CI], 1.2-3.7). They are correspondingly more likely to report being "not at all" awkward or uncomfortable in performing a breast examination (95% vs 87%; OR, 3.8; 95% CI, 1.4-6.8). While there were modest differences in ratings in other domains according to physician age, they did not remain statistically significant after controlling for physician specialty and sex.
There is widespread agreement that preventive care, and particularly cancer screening, in men and women is important. In this article, we report substantial differences between male and female physicians in their self-reported comfort and skill in obtaining a sexual history from and in examining patients of the same and opposite sex. We also find that many primary care physicians rate their skills as less than excellent in sex-related examinations, especially those involving the opposite sex. This was most profound in the case of the prostate examination, in which fewer than 10% of female internists and family physicians rated their skills as excellent.
We also found differences in physicians' rating of their skill and comfort according to specialty. Even controlling for physician sex, internists lag well behind their colleagues in family medicine in their comfort and skill in obtaining Papanicolaou smears, perhaps providing an explanation for why their screening rates are lower. Regarding the prostate examination, the interaction of specialty and sex suggests a bigger sex difference for family physicians than for internists (Table 1).
While these data may not be surprising, there is a public expectation that physicians are objective and neutral and that they are aware of and able to overcome their personal biases for the good of their patients. These data do not dispute the objectivity or neutrality of physicians, but they do suggest that physicians may have difficulty overcoming some types of personal feelings. Physicians readily admitted experiencing feelings of awkwardness or discomfort with aspects of the sexual history taking and physical examination, but their ratings of skill suggest that they may have difficulty overcoming their emotional discomfort to become more proficient.
Some might argue that our findings simply reflect aspects of "human nature" that cannot be altered by interventions aimed at improving comfort and skill. However, we found that for the clinical breast examination, self-ratings of comfort and skill are higher among older physicians, suggesting that perhaps providing more experience to trainees (in medical school and during residency) might improve comfort and skill. The need for such improvement seems particularly pressing in training internists to obtain Papanicolaou smears and in training female physicians to perform prostate examinations. Our findings suggest that it may also be prudent to improve the comfort and skill of practicing physicians as well as trainees, although this would require clearer knowledge about sources of discomfort and how skills need to be improved.
One could argue that the strategy of expecting internists and family physicians to perform examinations that they are uncomfortable with is unrealistic and that such physicians should simply refer their patients to another provider for certain examinations or procedures, such as obtaining Papanicolaou smears. Unfortunately, our data suggest that such practices are associated with lower screening rates, perhaps because of the additional step this creates, at least for Papanicolaou smears6; we would hypothesize similar findings for prostate examinations.
We cannot determine whether the discomfort and lower skill perceived by a physician is also sensed by patients. Our previous work suggests that female patients rate the skills of female physicians more highly than those of male physicians and that they believe that female physicians are more comfortable discussing sexual issues, menstruation, and menopause.12 We also cannot determine whether the discrepancies in comfort and skill have any direct relation to other patient outcomes, such as satisfaction, or to actual screening rates. Other data from this study of female patients12 indicate that patient reports of the quality of the physician-patient communication are independently related to higher rates of breast and cervical cancer screening6 and that physician comfort and skill may be related to the quality of such communication. We are unaware of any analogous data relating to male patients.
One could argue that our dichotomization of response choices is artificial and that less than excellent is more a reflection of an honest self-evaluation than actual skill level. However, the distribution responses of physicians regarding patients of the opposite sex were fundamentally different. The use of mean scale scores led to similar findings, but these scores are less readily interpretable.
Another limitation of this study is its focus on physicians in a single health plan in 1 geographic area, although we know of no reason that the results could not be generalized. We rely on physician self-report of skill, as did Wechsler et al7 in Massachusetts; there is no clear standard criterion with which to compare physician self-report of skill. Tannen13 suggests that women are more likely than men to downplay their competence; we do not know if male and female physicians, or physicians in different specialties, have tendencies to overreport or underreport their skill or to evaluate their capabilities with more or less humility. However, the fact that physicians of each sex reported less comfort and skill in caring for patients of the opposite sex than their opposite sex counterparts argues that the results of such a potential bias are unlikely to quantitatively change the results. The use of standardized patients would provide a more objective way to measure skill but was not feasible in our study.
We used a telephone survey to collect our data. While physicians may be freer to admit to their weaknesses using an anonymous written instrument, it is almost impossible to obtain high response rates such as ours from physicians with written surveys.
In sum, there is substantial room for improving the skills of practicing physicians in the performance of sex-sensitive examinations and in obtaining a sexual history from members of the opposite sex. Further efforts to understand reasons for physician discomfort and less than optimal skill should be explored. Interventions aimed at improving skill and the physician's comfort level in performing these examinations should then be implemented.
Accepted for publication April 25, 1997.
This study was supported by the Commonwealth Fund, New York City.
We thank Sherrie Kaplan, PhD, for her help in the development of some of the measures used in this study.
Reprints: Nicole Lurie, MD, MSPH, Department of Medicine, Box 741 UMHC, 420 Delaware St SE, Minneapolis, MN 55455.
From the Departments of Medicine, Hennepin County Medical Center (Drs Lurie and Margolis), the University of Minnesota Medical SchoolMinneapolis (Drs Lurie and Margolis), and the University of Minnesota School of Public Health (Drs Lurie and McGovern and Ms Mink), Minneapolis.
1. Lurie N, Slater J, McGovern P, Ekstrum J, Quam L, Margolis K. Preventive care for women: does the sex of the physician matter? N Engl J Med. 1993;329:478-482.
WEB OF SCIENCE
2. Franks P, Clancy CM. Physician gender bias in clinical decision making: screening for cancer in primary care. Med Care. 1992;31:213-218.
3. Krueter MW, Strecher VJ, Harris R, Kobrin SC, Skinner CS. Are patients of women physicians screened more aggressively? a prospective study of physician gender and screening. J Gen Intern Med. 1995;10:119-125.
WEB OF SCIENCE
4. Bertakis KD, Helmes LJ, Callahan EJ, Azari R, Robbins JA. The influence of gender on physician practice style. Med Care. 1995;33:407-416.
WEB OF SCIENCE
5. 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. Fam Med. 1992;24:58-61.
6. Lurie N, Margolis KL, McGovern PG, Mink PJ, Slater JS. Why do patients of female physicians have higher rates of breast and cervical cancer screening? a study of doctors and their patients. J Gen Intern Med. 1997;12:34-43.
WEB OF SCIENCE
7. Wechsler H, Lewin S, Idelson RK, Schor EL, Coakley E. The physician's role in health promotion revisited: a survey of primary care practitioners. N Engl J Med. 1996;334:996-998.
WEB OF SCIENCE
8. Abraham S. Vaginal and speculum examination in medical curricula. Aust N Z J Obstet Gynaecol. 1995;35:56-60.
WEB OF SCIENCE
9. Frye CA, Weisberg RB. Increasing the incidence of routine pelvic examinations: behavioral medicine's contribution. Womens Health. 1994;21:33-55.
10. Hennigan TW, Franks PJ, Hocken DB, Allen-Mersh TG. Influence of undergraduate teaching on medical students' attitudes to rectal examination. BMJ. 1991;302:829.
11. Wilt TJ, Cutler AF. Physician performance and patient perceptions during the rectal examination. J Gen Intern Med. 1991;6:514-517.
12. Margolis K, Lurie N, Mink PJ, McGovern P. Doctor-patient communication predicts breast and cervical cancer screening. J Gen Intern Med. 1996;11:S126.
13. Tannen D. Talking From 9 to 5. New York, NY: George Morrow Press; 1994.
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES
Physicians' opinions about partner notification methods: case reporting, patient referral, and provider referral
Hogben et al.
Sex. Transm. Infect. 2004;80:30-34.
| FULL TEXT
Intimacy and Women With Type 2 Diabetes: An Exploratory Study Using Focus Group Interviews
Sarkadi and Rosenqvist
The Diabetes Educator 2003;29:641-652.
Preparedness of Internal Medicine and Family Practice Residents for Treating Common Conditions
Wiest et al.
| FULL TEXT
Erectile dysfunction--the effect of sending a questionnaire to patients on consultations with their family doctor
Kitai et al.
Fam Pract 2002;19:247-250.
| FULL TEXT
Improving Quality Improvement Using Achievable Benchmarks For Physician Feedback: A Randomized Controlled Trial
Kiefe et al.
| FULL TEXT
Male Adolescents and Physician Sex Preference
Van Ness and Lynch
Arch Pediatr Adolesc Med 2000;154:49-53.
| FULL TEXT