Gender and the professional career of primary care physicians in Andalusia (Spain)

The number of women in medicine is progressively increasing in numerous countries [1], but they remain unequally distributed among medical specialties [1‐3]. Female physicians predominate in general specialties (e.g., pediatrics, psychiatry and family medicine) that are characterized by a much lower salary and prestige [2], a more easily planned work schedule, and a greater interaction with patients in comparison to specialties in which males prevail (e.g., surgery) [1].

Women are underrepresented in healthcare management. In our regional health system, 68% of the health personnel were female in 2007 but only occupied 34% of management posts were held by women (37% in the primary care [PC] sector) [4].

In 2006, 33.8% of male physicians in Catalonia held a management post compared with 18.7% of the females (in the PC sector: 16.9% of males vs. 6.2% of females) [5].

Female physicians are also poorly represented on the governing bodies of scientific societies [6‐8] and any presence is frequently in a secondary position [7]. They also appear to face barriers in the leadership of research projects [9] and the publication of scientific papers [7, 10‐12]. Hence, despite the increased number of female physicians, they do not appear able to develop career-enhancing activities under conditions of equality with their male counterparts [9, 13].

The objectives of this study were to determine and compare the professional activities of female and male PC physicians in Andalusian cities and to evaluate the effect of the health center on the performance of these activities. This study is part of a wider project designed to determine the healthcare activities, objective professional achievements and achievement perceptions of male and female physicians in Andalusia.

Responses were received from 368 PC physicians (73.6%): 182 (71.7%) of the females and 186 (75.6%) of the males. The mean age of the respondents was 49.1 ± 4.3 yrs for the females versus 51.3 ± 4.9 yrs for the males (p < 0.005). After exclusion of the non-respondents, the statistical power of the study was reduced to 60.3%.

Comparison of the four personal variables obtained in all enrolled physicians showed no differences between respondents and non-respondents except that health center managers were more likely than non-managers to be respondents (Table 1).

As shown in Table 2, the female physicians were younger, more frequently possessed the family medicine specialty, devoted more hours to housework, and were more likely to live alone in comparison to the males. There were no gender differences in work load variables.

Table 3 depicts the crude and adjusted ORs for the relationships between gender and professional activities. The adjusted OR was significant for health center manager, recycling courses, clinical sessions led at the health center, collaborating investigator, author of original article, author of book/chapter, co-author of congress paper, member of governing body of scientific society, member of governing body of medical association, and member of public health service working group (p < 0.05), while it was at the limit of significance for participation as teacher, congress participant, and congress speaker (p < 0.10). 10 of the 24 professional activities studied and at the limit of significance (0.05 < p < 0.10) in a further 3: all of these 13 activities were performed less frequently by female physicians after adjusting for family responsibilities (3 variables), family situation, and work load (2 variables), among others.

The ICC was virtually 0% for health center management and course attendance. The ICC was >0.50 for assignment of resident, authorship of original articles, and resident tutorship.

The response rate in this study was very acceptable for a self-administered questionnaire, and respondents only differed from non-respondents in the higher proportion of center managers responding; these represented only 7.6% of the sample and had been personally contacted about the questionnaire.

The female physicians were younger and were more frequently specialized in family medicine, while there were no gender differences in patient list size or mean n° patients/day, as also found in a study of 32 European countries [15]. Although there was no difference in cohabitation with dependent individuals, the females devoted more than double the time to housework and were more than three times more likely to live alone with children in comparison to the males, reproducing the situation in the general population [16, 17]. Hence, the higher professional qualifications of female physicians in comparison to the general population do not appear to have improved their conditions of inequality in domestic tasks. Interestingly, females comprised 68% of our regional health service healthcare professionals in 2007, as noted in the Introduction, but made 74% of the requests for permission to care for a relative (excluding birth-related requests) [4].

After adjusting for age, family medicine specialty, care load, family load, and family situation, the participation of female physicians was significantly less frequent in 13 of the 24 activities in comparison to the males. In all activities, except for congress participation as a speaker, the adjusted odds ratio was higher than the crude odds ratio, i.e., the gender inequality in the professional activities was more marked after adjustment for work and family loads.

Only 5% of female physicians were health center managers in comparison to 15% of male physicians; mirroring the situation observed elsewhere in Spain [5, 9] and other countries [1, 18].

The higher proportion of female physicians with the family medicine specialty does not translate into differences in accreditation as resident tutor, suggesting that women do not take full advantage of their superior training. After postgraduate training, male and female physicians attend the same number of courses but the females are less likely to participate in recycling activities that entail leaving their surgery for weeks at a time. Moreover, female physicians less frequently conduct sessions for their teams or take on a teaching role.

There was no gender difference in participation in scientific societies or medical associations (compulsory), but female physicians are less likely to sit on the governing body of these organizations, as found by other studies [1, 6, 7]. According to one report [9], although 45% of physicians are female, they only occupy 33% of management posts. In the present study, the participation of female physicians was similar to that of males in working groups of scientific societies but lower in national and regional health authority working groups.

There was no significant gender difference in work as principal investigator, likely due to the small numbers involved, but the female physicians were less frequently collaborative investigators. Likewise, there was no difference in being the first author of scientific papers, but female physicians were less frequently co-authors. It is possible that the females focused more on leading projects and authoring papers than on collaborating with colleagues. An alternative interpretation is that they are less frequently included in research led by others, as claimed by female physicians in a previous study [19]. The male physicians attended more scientific congresses and more frequently presented papers, as observed in the congresses of other scientific societies [8], although was no gender difference in congress committee participation.

The female physicians published fewer original articles and books, but there was no gender difference in other types of publication. A review of four Spanish medical journals found that 71% of the first authors were male [11]. The proportion of female authors in the publication of original articles is increasing and varies according to the specialty, but data published by two reviews, one of six US journals [10] and another of six UK journals [20], suggest that a gender gap persists, more markedly for the last (senior) author. The lesser career progress of female physicians has been attributed to their inferior productivity, but adjusting for publications and work load, they continue to occupy lower professional positions [21, 22].

Jiménez-Rodrigo [12] summarizes the complex causes of gender inequality in the publication process, including: the underrepresentation of females on editorial committees; the role of influence networks in citations and peer reviews, more favorable for men; and the selection of "softer" research subjects by women, with different methodological criteria, which are frequently less readily accepted by the scientific community.

There were no significant gender differences in the possession of a PhD or university position, although a small number of the physicians fulfilled this condition. Nationally, only 26% of physicians with university teaching positions were female in the academic year 2007-2008 [23].

According to the ICC findings, the health center had a significant impact on the tutorship and assignment of residents, which took place in 41 health centers (containing 236 of the physicians). The health center also influenced the authorship of original articles (including co-authorship), with the 55 authors grouped in a few centers (0.70 ICC). Nevertheless, a gender difference in this activity remained after controlling for the effects of the health center. However, due to our cross-sectional study design, we cannot know whether this high ICC implies cause or effect, i.e., whether physicians who publish as a group together in certain health centers or whether conditions in some centers favor research activity.

The health center had no influence on differences observed in the remaining activities studied (low ICC values).

It has been proposed that professional gender inequalities will diminish with the growing proportion of females in the medical profession (cohort or generation effect) [24]. However, although all categories of the professional hierarchy are feminizing, there is evidence that women remain disadvantaged in their access to the highest levels [24, 25]. Thus, it has been reported the leadership levels now achieved by female physicians are no higher than they were some decades ago, and that this phenomenon cannot be explained by a so-called "period" or "environmental" effect [18].

Riska [26] brings together two mechanisms by which gender inequalities in medical careers are maintained. One relates to the variation in decision-making due to gender differences in socialization, based on sex role theory. The other involves structural barriers, based on the concept of the "glass-ceiling" [27, 28], i.e., an invisible barrier to the advancement of women (in general) to the highest posts, formed by external obstacles (organization structure, organizational culture, and gender stereotypes), internal obstacles (gender identity, achievement motivation, and personality aspects), and interactive obstacles.

Studies have identified different categories of obstacle for female physicians, including rigidity of the organization and professional career, sexual discrimination, psychological barriers, career motivation and family responsibilities [29, 13, 30]. These responsibilities and the difficult balance between family tasks and work were found to be the greatest impediment to career advancement in a North America study [19] and were reported to diminish the interest of female physicians in their career [31]. Aspects of the interaction between family and work roles were reported to be causes of dissatisfaction among female medical faculty in the USA [30]. There is some evidence that male and female expectations of the work-life balance are changing, with both genders desiring its improvement [32]. One US medical school [33] found that career difficulties for female physicians augmented with the larger number of children they had, although this finding was not replicated in a subsequent study in the same setting [25] or in the United Kingdom [34]. It has been claimed that gender inequalities will no longer pose a problem when female physicians do not have to choose between family responsibilities and professional status [35]. The Association of American Medical Colleges has published the most effective measures for improving the success and leadership of female physicians and reducing inequalities [36].

Study limitations include the cross-sectional design, preventing study of the direction of the relationships found, and the restriction of health centers and PC physicians to the urban setting, preventing extrapolation of results to PC physicians as a whole. In addition, the statistical power of the study was reduced to 60% after excluding non-respondents. However, the study has internal validity, since the main objective was to compare professional activities by gender. There may also have been information bias, since all except for six of the analyzed activities were self-reported, although this risk is limited by the use of the same data source (questionnaire) for both genders.

Two major study strengths are the inclusion of some family variables and the multi-level analysis. As a result, we were able to determine the association of GP gender with professional activities after controlling for the effects of the family and health center settings.

Longitudinal studies are required to improve our knowledge of this issue, analyzing the professional performance of each gender, identifying barriers, enhancing the development of female leaders and supporting their needs [1, 13, 30].

There is also a need for qualitative studies that address the meaning and complexity of this phenomenon for female and male physicians, integrating this information with quantitative data.

These results strongly indicate that female physicians are disadvantaged in Andalusia. Gender inequalities were found in the development of professional activities of physicians in urban health centers, even after controlling for family responsibilities and the effect of the center, which was important in only a few of the activities studied. This situation of disadvantage is likely to have a negative effect on the professional career of female physicians.