Introduction

Emergency medicine has traditionally been a male specialty, and remains so in spite of increasing female representation.₁ As of June 1996, the total number of (self-designated) emergency physicians in the AMA Masterfile database was 19,112, and of those, the total number of women was 3,297 (17%). ₂ However, women are now about a quarter of applicants to Emergency Medicine programs, still less than their prevalence among medical school graduates, but increasing.₁

There have been few studies of personal health choices or other characteristics of either women or men emergency physicians, although some studies (usually involving disciplines other than emergency medicine) have found that physician characteristics are important predictors of patient outcomes, ₃, ₄, ₅, and may also be inherently of interest. An a priori goal of the Women Physicians’ Health Study (WPHS) was to examine the extent to which the personal and professional characteristics of women physicians varied by specialty.

Methods

Women emergency physicians are a subset of WPHS. The fundamental population characteristics, the design and the methods of WPHS have been more fully described elsewhere.₆ In brief, WPHS surveyed a stratified random sample of women medical doctors (MDs) in the United States. The sampling frame was the 1993 Physician Masterfile of the American Medical Association (AMA). This database lists all MDs residing in the U.S. and its possessions. A random sample of 2500 women, stratified by decade of graduation from medical school, was drawn from each of the last four decades' graduating classes (1950 through 1989). Women who graduated in the 1950s and 1960s were thereby over-sampled to correct for their otherwise sparse proportional allocation.

The WPHS sample included physicians in active practice, part-time
practice, professionally inactive, and retired physicians, ranging in
age from 30 to 70 years, who were not in a residency training program
in September 1993 when the sampling frame was constructed.

After receiving IRB approval, the first of four mailings of the WPHS
survey was sent out in September 1993. Each mailing contained a cover
letter and a self-administered four-page questionnaire. Three
additional mailings were sent out before enrollment was closed in
October 1994. Of the initial sample, an estimated 23% were ineligible
to participate because their addresses were wrong, they were men, were
deceased, were living out of the country, or were interns or
residents. The overall response rate was 59% of eligible
physicians. The final number of respondents was 4,501.

Respondents were compared to non-respondents in three ways. First, a random sample of 200 non-respondents were telephoned and compared to all written survey respondents. Second, the AMA Physician Masterfile data was used to compare characteristics of all respondents to all non-respondents. Third, an examination by mailing waves compared respondents to non-respondents, from wave 1 through 4 for a large number of key variables. Non-respondents were less likely than were respondents to be board-certified, however, they did not consistently or substantively differ on any other tested measure, including age, ethnicity, marital status, number of children, alcohol consumption, fat intake, exercise, smoking status, hours worked per week, primary care practice, personal income, or percentage actively practicing medicine.Data were weighted to adjust for our decade-specific response rate, and for board-certification status. This weighting permits us to generalize our results to the entire population of women physicians who graduated from medical school between 1950 and 1989. All analyses were run in SUDAAN ₇ to account for our weighting strategy. Due to multiple testing, unless otherwise noted, only characteristics significant at p<0.01 are discussed.

Results

Of the 4501 women physicians included in WPHS, 90 were women
emergency physicians (WEPs), 2.5% of the weighted sample. These WEPs
were compared to other women physicians (OWPs).Table 1 presents the
demographic and personal characteristics of women emergency
physicians. WEPs were younger than OWPs, and were more likely to be
single or part of an unmarried couple, and less likely to be
married. They were more likely to be white and somewhat more likely to
be born in the United States (p=.02).

Figure 1

Table 1

Figure 2

¹chi-square test for prevalences

²includes married and unmarried couples

*≤
.05, **≤ .01, ***≤.001, ****≤.0001

Table 2 describes professional
characteristics of WEPs and OWPs. Young WEPs were more likely than
young OWPs to be board certified in another specialty and less likely
to be board certified in their primary specialty (p=.06). Fewer WEPs
were residency trained.

Figure 3

Table 2: Professional Characteristics of Women Emergency Physicians by Decade of Graduation

¹chi-square test for prevalences

*≤
.05, **≤ .01, ***≤.001, ****≤.0001

Table 3 shows the practice characteristics
of WEPs. Nearly all WEPs practiced in a hospital or group setting.
WEPs were more likely than OWPs to practice in a rural area, and less
likely to practice in an urban area. Total clinical hours were similar
(although WEPs work fewer non-clinical and fewer total hours), but
their personal income and hourly wage were substantially higher. CME
hours per month were also similar, however, WEPs used audiotapes more
and textbooks less than did OWPs.

Figure 4

Table 3: Practice Characteristics of Women Emergency Physicians

Figure 5

¹chi-square test for prevalences, t-test for means, and medican split tests for medians

²Income per hour was estimated for each individual by dividing the income catagory midpoint by her total work hours per year (inactives were excluded).

³Categorical choices for personal and household incomefdere: $0; $1-<25,000; $25,000-<50,000; $50,000-<100,000; $100,000-<150,000; $150,000-<200,000; $200,000-<250,000; $>250,000. Medians were calculated using midpoints of the catagories.

*≤
.05, **≤ .01, ***≤.001, ****≤.0001

Table 4 illustrates the
professional satisfaction of WEPs. WEPs felt a similar level of
professional satisfaction as OWPs, with the exception of work stress;
more women emergency physicians reported severe work stress.

Figure 6

Table 4: Professional Satisfaction of Women Emergency Physicians

¹chi-square test for prevalences

²<2% of the sample responded “too little”

*≤
.05, **≤ .01, ***≤.001, ****≤.0001

Table 5
reports on the personal health habits of WEPs. WEPs and OWPs did not
differ on any measure of health status.

Figure 7

Table 5: Personal Health Habits of Women Emergency Physicians

¹Chi-square test for prevalences, t-test for means, and median split tests for medians

²% complying with all screening recommeendations from the U.S. Preventive Services Task Force regarding cholerterol (check every ≤ 5 years), blood pressure (every ≤ 2 years), Pap smears (every≤3 years if uterus present), clinical breast exam (every ≤3 years if age 30-39, every≤1 year if ≥40 years old), and mammography (every ≤2 years if 50-70 years old).

Table 6 shows WEPs
reported patient counseling practices. This is where the largest
differences between WEPs and OWPs lie (with the exception of blood
pressure counseling and HIV risk). The WEPs did less counseling, less
screening and felt less confidant of their ability to do so than did
other women physicians. They ascribed less clinical relevance to the
majority of counseling practices, and their training and
self-confidence in performing these practices was significantly less
than that of other practitioners. For most of these counseling
practices, the significance of the difference between emergency
physicians and others was p £ 0.0001.

Figure 8

Table 6: Health Counseling Practices of Women Emergency Physicians1,2

Figure 10

{image:10}

*p≤.05
**p≤.01 ***p≤.001 ****p≤.0001

¹ n varies slightly by question; minimum n were: 58 (emergency physicians) and 1818 (others)

² Chi-square test

³Percent of physicians responding “every visit” or “every ≤1 year” to “Considering you typical patients, how often do you usually discuss or perform screening: every visit, every ≤1 year, every >1-2 years, every >2-3 years, every >3-5 years, only at initial visit, only if clinically indicated, never.”

⁴Percent of physicians responding “highly” to query regarding physician’s perception of the column risk factors’ “relevance to your practice: highly, somewhat, not very, not at all.”

⁵Percent of physicians responding “highly” to query regarding physician’s “self-confidence in counseling: highly, somewhat, not very, not at all.”

⁶Percent of physicians responding “extensive” to query regarding physician’s “amount of training in counseling: extensive, some little, none.”

⁷All other physicians are defined here as general practitioners, general internists, obstetricians/gynecologists, specialists in public health/preventionists, dermatologists, medicine specialists, neurologists, opthamologists, psychiatrists, surgeons and others working at least 5 clinical hours per week.

⁸ Includes only those patients >50 years old (per USPSTF guidelines).

⁹ Includes only those patients >65 years old (per USPSTF guidelines).

¹⁰ Includes only smokers.

¹¹Includes all women patients

¹²For all women patients≥50-≤75 years old.

¹³For women patients who are post-menopausal.

Discussion

WEPs differed from OWPs in several interesting ways, although there were also many similarities. Inherent characteristics of the individuals choosing this specialty explain some of our findings; others are a result of the nature of the practice coupled with the history of the specialty. Demographic Characteristics of Women Emergency Physicians Emergency physicians’ demographic (and other) characteristics are in part due to self-selection into the specialty,₈ and in part developed in response to the nature of the specialty.₉, ₁₀ Emergency departments must be constantly staffed, physicians see unselected patient populations, and off-time is usually completely free of professional duties. EM may therefore be more appealing to younger practitioners who better tolerate chronostress,₁₁, ₁₂ and those who are willing to accept risk.₉ Others have noted that emergency physicians like excitement, change and challenge, are impatient, want rapid gratification and tolerate a high degree of uncertainty.₁₃ Such persons may also be more hesitant to make a commitment. Consonant with these prior findings, we found that WEPs were younger, and less likely to be married than OW Ps.

The percentages of WEPs in rural practice (23.9% of WEPs vs 9.9% of
OWPs) is noteworthy. WEPs might have chosen their specialty because
their off-time allows more active, rural, outdoor
activities. Similarly, since emergency physicians need not be
available when not on duty, they can more easily live at considerable
distances from their practice sites. WEPs’ higher use of audiotape
hours could possibly also be ascribed to their rural residence, which
might decrease access to other CME sources or increase commuting time.

Stress and career satisfaction

We found that WEPs reported higher perceived stress than did others. This was true although they reported fewer work hours, similar work control, similar satisfaction with amount of work hours, and similar amounts of stress at home. These findings would suggest attitudinal, patient- or system-related explanatory factors. Much has been written about stress and career longevity in emergency medicine.₁₄, ₁₅ Previous studies have hypothesized that high stress is inherent to emergency medicine: shift work and chronostress; unknown patients; acute presentations; and the need for immediate decisions with inadequate information.₁₃ Others ₁₆ have suggested that a lack of control could be an issue for EPs: EPs usually cannot hire or fire the staff with whom they work, design or change the emergency charting system, or affect lab turn-around time, and are often not part of the hospital power structure.₁₀ And still others have conjectured that women physicians may also sense competing personal and professional roles more keenly, and may therefore be subject to more stress than are men.₁₇

High stress levels might also be reflected in a desire by practitioners to change specialty. ₁₆ Yet we found that WEPs were as satisfied with their careers as were OWPs and were no more likely to change their specialty. One possible explanation for this seeming paradox might be found in the WEPs’ higher average income. Another explanation for the general satisfaction WEPs expressed with their career might be the personality characteristics summarized above, or the lack of non-clinical duties. WEPs averaged 2 non-clinical hours per week compared to 5.0 hours for OWPs.

Counseling Practices of Women Emergency Physicians

It has been shown that in general, women are more avid preventionists than are men physicians ₅,₆, ₇,₁₈,₁₉, and that primary care practitioners counsel more than specialists ₂₀. Emergency medicine has long struggled with whether it is primary care or a specialty. We found that WEPs performed significantly less counseling and screening than did OWPs, and felt much less confident than OWPs in performing these activities; this would tend to support the ‘specialty’ hypothesis. Emergency Departments have been called society’s ‘safety net.’ ₂₁ Many have suggested that the ED is an ideal place to perform health screenings ₂₂, vaccinations ₂₂,₂₃,₂₄, ₂₅, or health counseling activities ₂₆. In many states, EDs are required by law to screen children for vaccination compliance, and tetanus screening is on most triage forms. EDs often provide the physician of last resort or most anonymity for the poor, the disenfranchised, or those with sexually transmissible infections. However, EDs are also chaotic, under time pressure, and chronically understaffed. As a result, although ED-based child vaccination programs have been shown to temporarily improve immunization rates, these improvements are not maintained. ₂₄ Furthermore, more than half of emergency physicians themselves are reluctant to immunize even adults, with the exception of tetanus ₂₄. Their unwillingness stems mainly from: 1) the perception that ED physicians are not primary care providers, 2) inadequate time or personnel; and 3) concerns about adverse reactions or medicolegal liability. Confirming this, few WEPs discussed flu vaccination with their patients, and most WEPs felt that they had received less extensive training in performing health-related counseling or screening than other women physicians.

Screening differences
depended in part on which specific health issue was at question.
Human Immunodeficiency Virus (HIV) is a topic very relevant to the
practice of EM. Emergency physicians care for many patients at high
risk for HIV infection, but are usually unable to appropriately test
their patients for HIV. Here, the differences between WEPs’ and
OWPs’ counseling practices were not statistically significant,
although the diminished difference is largely a function of small
numbers of OWPs who counsel about HIV, and the number of EPs who felt
inadequately prepared by their training was still considerable.

EPs daily face the contribution alcohol makes to motor vehicle crashes and interpersonal violence, and many repeatedly see the same alcoholic patients. We found that WEPs were as likely to think alcohol counseling was relevant to their practice as OWPs, but we and others have found that EPs are much less likely to actually counsel about alcohol.₂₇ The negative effects of smoking are also often encountered in the ED. Again, there was no important difference in the percent of WEPs vs. OWPs who felt that smoking cessation counseling was relevant to their practice (42.1% vs 52.8%) but here again, the WEPs felt inadequately prepared to perform such counseling.

Conclusions

Women emergency physicians resemble other women physicians in many
respects, but tend to be younger, earn more, and to report much the
same job satisfaction in spite of higher reported job stresses. The
biggest difference between WEPs and OWPs is that WEPs in general felt
poorly trained and poorly prepared to do prevention-related
counseling. These findings have important implications for the future
development of emergency medicine training programs, and the type and
cost-effectiveness of care rendered in the ED. Study Group Authorship
and Acknowledgements: We are grateful for our financial support for
this research, which has come from the American Medical
Association’s Education and Research Foundation, the American Heart
Association, an NIH (NHLBI) institutional National Research Service
Award (#5T32-HL-07034), the Emory Medical Care Foundation, and the
Ulrich and Ruth Frank Foundation for International Health.