VI.   Female Physicians

The growing proportion of physicians who are female is having a profound impact on the physician workforce and delivery of care.  In this chapter we discuss supply trends and differences in productivity and compensation between male and female physicians.

A.   Supply Trends

During the past 3 decades the proportion of physicians who are female has risen from 8 percent to nearly one in four physicians.  Recent trends suggest that within the next 2 decades women will constitute nearly half the physician workforce.

The increase in number of new physicians who are female means that male physicians tend to be older, on average, than female physicians.  AMA (2006) reports that in 2004 approximately 36 percent of active male physicians were under the age of 45 as compared to approximately 61 percent of active female physicians.  Exhibit 63 illustrates the different age distribution of male and female physicians. 

Exhibit 63. Male and Female Physician Age Distribution, 2004

[D]

Source: Physician Characteristics and Distribution in the US: 2006 Edition (AMA, 2006).

Since 1970 the number of female medical school applicants and medical school graduates have risen sharply (Exhibit 64).  Today (2005), nearly half of all U.S. medical students are female.

Exhibit 64. Percentage of U.S. Medical School Applicants and Graduates who are Female

[D]

Source: AAMC (2001) and AMA Physician Characteristics and Distribution in the U.S.  (various years).

Specialty Choice

The proportion of physicians who are women varies substantially by medical specialty, with women more likely to choose primary care specialties over surgical or other subspecialties.  Among those specialties with more than 10,000 physicians, the two specialties with the highest proportion of female physicians are general pediatrics (52 percent) and obstetrics and gynecology (41 percent).  The two specialties with the smallest proportion of female physicians are orthopedic surgery (4 percent) and urology (5 percent) (Exhibit 65).

Novielli et al. (2001) report that differences between male and female physicians in choice of medical career path stem not from experience, but rather from personal preference.  The authors find that women starting medical school are more likely than men to express a desire to practice in a non-surgical specialty.  Furthermore, during medical school women are more likely than men to be dissuaded from entering a surgical specialty.  Of those new enrollees in medical school who expressed an initial preference for a surgical specialty, the proportion that eventually entered a non-surgical residency program was higher for women than men. Similarly, of those new enrollees in medical school who expressed an initial interest in a non-surgical specialty, the proportion that eventually entered a non-surgical residency program was higher for women than for men. 

Exhibit 65. Percent of Physicians who are Women: 2004

[D]

Source: Physician Characteristics and Distribution in the US, 2006 Edition (AMA, 2006)

Nonnemaker (2000) tracked the appointments of medical school graduates to medical school faculties from the years of 1979 to 1993 and found that female associate professors were significantly less likely than males to be promoted to full professor.  Although women are under represented in academic medicine, their ranks are growing.  In 1979, only 647 women were full professors; by 1997 that number had increased almost four-fold to 2,335. 

Geographic Location

There is considerable evidence to show that female physicians are less likely to practice in non-metropolitan areas compared to their male colleagues.  Randolph and Pathman (2001) find that women, who make up approximately two thirds of pediatric residents, are 50 percent less likely to practice in rural areas than are male pediatric residents. 

Ellsbury et al. (2002) describe reasons why female physicians may be more hesitant to practice in non-metropolitan areas compared to male physicians.  Female physicians considering practice in a non-metropolitan area typically have greater concern about

1. spousal employment opportunities (58 percent of women compared to 26 percent of men),
2. flexible hours (66 percent versus 25 percent),
3. availability of child care (33 percent versus 3 percent), and
4. opportunities for part-time employment (38 percent versus 14 percent). 

Physicians in non-metropolitan areas work longer hours and work in smaller practices, on average, compared to physicians in metropolitan areas.  These factors possibly have a greater disincentive effect on female physicians who tend to have greater preferences for flexibility in hours to bear children and raise families. 

Mitka’s (2001) study of physicians in rural communities in the Pacific Northwest finds that 52 percent of women and 24 percent of men expressed that they had a partner or spouse looking for work when considering the location for their own practice.  Ellsbury et al. report that 54 percent of respondents to a question about spousal assistance found that their non-metropolitan community provided no assistance to help a spouse or partner find employment when relocating to the area.  The rising proportion of women in medicine and the higher propensity of female physicians to practice in metropolitan areas could hinder the national goal of improving physician supply in rural areas.  Although women are less likely to work in rural areas, according to study by Bickel and Ruffin (1995) women are more likely than men to work in clinics providing health care to medically indigent patients.

Employment Status

Female physicians are more likely than their male counterparts to work in salaried, office-based settings.  AMA (2001) reports that approximately two thirds of female physicians and one third of male physicians are salaried (Exhibit 66).  Although salaried physicians tend to earn less than self-employed physicians, salaried physicians tend to have more predictable and flexible work hours, factors that studies have found appeal more to women than to men. 

Exhibit 66. Employment Status of Male and Female Physicians

[D]

Source: Physician Socioeconomic Statistics (AMA, 2001).

B.  Productivity

AMA (2002) reports that female physicians work 49 hours per week, on average, compared to 57 hours for male physicians.  Female physicians also tend to work fewer weeks per year. Estimates of patient care hours worked per week from an unpublished HRSA survey (2002 data for approximately 46,800 physicians) shows that even controlling for age and specialty women tend to work fewer hours per year than do men (Exhibit 67).  Perhaps resulting from differences in treatment styles, Roter et al. (2002) find that female physicians average about 2 minutes (10 percent) longer than male physicians in terms of the average length of a patient visit.

Exhibit 67. Mean, Patient Care Hours per Week, 2002

Source: Unpublished HRSA Survey of Physician Work Hours (total sample is approximately 46,800 physicians; hours for only selected specialties presented here).

Because female physicians provide care to fewer patients per year, on average, compared to male physicians, the supply of physician services is growing more slowly than the number of active physicians in percentage terms.  As discussed in Chapter II, between 2005 and 2020 the overall supply of physicians is projected to grow by 16 percent, while the FTE supply is projected to grow by 14 percent.  Part of this discrepancy is due to the increasing proportion of women in the workforce, while part is due to the aging of the physician workforce.

C.  Compensation

Novielli et al. (2001) find that female medical students have lower earnings expectations than do male medical students, even after controlling for whether the student plans to pursue a high-paying surgical specialty or a non-surgical specialty.  Ness et al. (2000) analyzed salary information for 455 internists in Pennsylvania and found that male internists earned 53 percent more than female internists.  However, the authors identify numerous systematic differences between men and women that help explain the disparity in earnings.  Compared to their male colleagues, on average, female physicians:

• Are more likely to practice in lower-paying medical specialties,
• Have fewer years in practice,
• Are less likely to be in a partnership,
• Work fewer hours per week in professional activities, and
• Are more likely to take time off or work part time.

After adjusting for age, training, and practice characteristics the authors report an unexplained 14 percent disparity in earnings.

Ross (2001) notes that payment for services rendered do not discriminate by physician gender and proposes that income differences between men and women likely reflect a voluntary tradeoff between earnings and lifestyle beyond those factors controlled for by Ness et al. Additional factors that might explain the difference in average earnings of male and female physicians are that, compared to their male colleagues, female physicians might see fewer patient per hour and be less likely to participate in night and weekend call activities.

Analysis of the AMA’s 1998 SMS file explains part of the difference in earnings between male and female physicians.  The SMS is a sample of approximately 3,000 office-based physicians who provide at least 20 hours of patient care per week.  As described in Chapter V, a regression equation was estimated to quantify the relationship between annual net earnings and its determinants.  Explanatory variables include physician characteristics and practice patterns, practice characteristics, medical specialty, and geographic location.  Two-thirds (n=2,055) of SMS respondents reported data on net earnings and the explanatory variables of interest. Approximately 17 percent (n= 341) of usable surveys were from female physicians.

Female physicians had average annual earnings of approximately $149,000 compared to $208,000 for male physicians (Exhibit 68).  This difference of over $59,000 (29 percent) per year can be partially explained by differences in average hours worked.  Female physicians in this sample worked 11 percent fewer hours per year, on average, compared to male physicians (2,412 versus 2,725 hours), and after adjusting for hours worked the difference in annual earnings falls to  $45,000 (or 21 percent).  Controlling for many of the systematic differences between male and female physicians in terms of practice patterns and medical specialty (as noted by Ross and by Ness et al.), it was found that female physicians still earn $38,000 (18 percent) less than male physicians.  Data limitations prevent us from controlling for systematic differences in additional factors that might explain even more of the differences in compensation.

Exhibit 68. 1998 SMS Physician Compensation Comparison

Source: Analysis of the 1998 AMA Socioeconomic Monitoring System file.