Physician Gender and Lifestyle Counselling to Prevent Cardiovascular Disease: A Nationwide Representative Study

Abstract

Background

Primary care physicians (PCPs) have a key role in the prevention of cardiovascular diseases (CVD). However, it is not clear whether lifestyle counselling behaviour differs between female and male PCPs. Nonetheless, this information might be helpful to develop need-based advanced training for female and male PCPs. Therefore, our aim was to identify potential gender differences in the implementation of health promotion and the prevention of CVD in primary care.

Design and Methods

In a Germany-wide survey called the ÄSP-kardio Study, we collected data from 4074 PCPs (40% female; from October 2011 to March 2012). We compared the provision of prevention measures, the attitude towards counselling, and the potential barriers in counselling among female and male German PCPs. We used chi² tests, Mann-Whitney U tests, and logistic regression analysis.

Results

We found differences in all of the above-mentioned aspects. Female PCPs were less likely to perceive barriers than male and more likely to ask patients about lifestyle, for example, nutrition (OR=1.62, P≤0.001). Additionally, female PCPs were more likely to feel well prepared (84.2% vs. 76.0%, P≤0.001) and successful (75.6% vs. 68.0%, P≤0.001). Male PCPs were more likely to mention barriers in daily practice that hinder lifestyle counselling.

Conclusions

Overall, both female and male PCPs had a positive attitude towards lifestyle counselling. Nevertheless, in view of the barriers that they indicated, incentives such as better reimbursement may help output-oriented PCPs to translate their positive attitude into action. Moreover, awareness of gender differences may help PCPs to acquire the specific advanced training that they need for effective lifestyle counselling in CVD.

Keywords

Primary health care general practice gender prevention cardiovascular disease

Introduction

Lifestyle counselling is an important instrument to reduce the burden of cardiovascular diseases (CVD).¹ In this respect, primary care physicians (PCPs) play an important role as health advisors since they are often the patient's first person of contact in questions of cardiovascular care.² Following the idea of social equity, characteristics of the PCPs should not influence lifestyle counselling.

Significance for public health

Lifestyle counselling is an important instrument to reduce the burden of cardiovascular disease. Here, primary care physicians (PCPs) play an important role as health advisors. Our study was able to identify deficits in the health promotion behaviour of PCPs. Because of the gender differences revealed in our study, male PCPs in particular should be sensitized to the importance and the potentials of prevention and health promotion. Overcoming the barriers of prevention and health promotion identified by the PCPs may be an important starting point. If, for instance, PCPs were better financially compensated for offering lifestyle counselling, which was rated as the most important barrier, it is conceivable that more PCPs would start to incorporate such measures into their daily routine. Additionally, a stronger focus on prevention and health promotion during advanced training programs for PCPs could increase the use of lifestyle counselling.

However, one PCP is not like another. Therefore, it seems to be important to identify core characteristics that might lead to different counselling behaviours. One example could be gender. Identifying potential differences in lifestyle counselling between female and male PCPs could be a first step towards the development of gender-specific and needs-based advanced training.

Previous studies have shown first evidence of a link between the gender of PCPs and the prevention measures they offer.^3,4 For instance, female physicians were more likely to offer dietary treatment,⁵ control risk factors for CVD,⁶ and reach treatment goals.⁷ A review by Jefferson et al.⁸ revealed that female physicians spent on average more than two minutes more per patient consultation than male physicians. Additionally, they were more engaged in positive talk and the consultation was more patient-cantered.⁸

Despite the important results of previous studies about gender influences on lifestyle counselling, PCP gender has often been neglected, especially in larger quantitative studies. Clark et al.⁹ ascribe this neglect to a traditional lack of variation in the medical profession, which was prevalent for a long time. However, nowadays about 40% of the PCPs in Germany are female.¹⁰

The lack of research on PCP gender in lifestyle counselling, combined with today's high percentage of female PCPs, makes it necessary to re-evaluate the extent and the impact of gender differences in lifestyle counselling. Ramirez et al.¹¹ came to the conclusion that future studies should investigate the greater likelihood of female physicians to discuss general prevention practices. Jefferson et al.⁸ call for future studies that use larger samples, show clear sample frameworks, and are from other health care settings than the United States.

Additionally, many initiatives and associations in Germany and other countries call for more gender-sensitive research. We want to take these requests into account. Therefore, this manuscript focuses on PCP gender in CVD prevention rather than just treating it as a covariate. In this manuscript we chose the term gender rather than sex to express not only the biological and physiological characteristics of female and male PCPs but also social, cultural, and psychological norms and roles.

Our aim was to identify potential gender differences in the provision of lifestyle counselling. Since identifying risk factors and potentially unhealthy behaviours is the first step towards a comprehensive prevention strategy,¹² we analysed if gender differences exist. We were able to confirm or disconfirm the link between PCP gender and prevention measures by using a large representative database. Additionally, we aimed at identifying potential gender differences and determining their importance both in PCP attitudes towards providing lifestyle counselling and in perceived barriers to the provision of lifestyle counselling. Female and male PCPs may differ in their perception of the importance of lifestyle counselling and its potential difficulties. As a result, it is essential to recognize these potential differences in female and male PCPs in order to supply them with gender-specific advanced training for lifestyle counselling.

Design and Methods

The data for this manuscript were drawn from the Germany-wide representative physician survey ÄSP-kardio (Physician Survey on Cardiovascular Disease Prevention), which included 4074 PCPs (general and medical practitioners as well as general internists with practices in Germany) and was conducted by the authors.¹³ The study was approved by the ethics committee of the Medical Faculty Mannheim, Heidelberg University (2008-272E-MA).

Data collection

Data of the ÄSP-kardio Study were collected from October 2011 to March 2012. Altogether, 13,294 PCPs were randomly selected by gender, medical specialty, and region from the largest PCP register existing for Germany, which was provided by ArztData GmbH, Hamburg. These PCPs were asked to fill in a four-page standardized questionnaire. The questionnaire included 145 items on PCP, patient, and practice characteristics; attitudes towards prevention and health promotion; and potential barriers to prevention and health promotion. The PCPs were given a compensation of 20 € for the time they needed to fill in the questionnaire (about 20 minutes). The questionnaire was carefully evaluated in a regional pilot study^14,15 and in in-depth cognitive interviews¹⁶ with female and male PCPs.

Before the questionnaire was sent to the PCPs, information about the ÄSP-kardio Study was published in relevant medical journals and all of the randomly selected 13,294 PCPs received personal study information. The questionnaire was sent out one week after the study announcement, together with a personalized letter, a data protection statement, and a prepaid self-addressed envelope. One week later we sent out a postcard reminder. Four weeks later we sent out the questionnaire with all supplementary documents for a second time to all PCPs who had not yet answered the questionnaire. In addition, we offered the possibility of filling in an online questionnaire. The response rate was 33.9% (n=4074). Our response rate was higher compared to most other German physician surveys (e.g., 7% or 15%).^17,18 Only surveys with a smaller sample size reached a higher response (e.g., 48%).¹⁹ There were no significant differences between the participating PCPs and the total German PCP population with regard to gender, medical specialty, and region (P-values: 0.902, 0.792 and 0.928).

Measures

The PCPs indicated their gender by ticking female or male. Since different socially constructed roles, behaviours, activities, and attributes that a given society considers appropriate for men and women are attached to this simple answer, we use the term gender instead of sex.

To measure the provision of prevention measures, we used several items (I measure their height and weight, I ask them about their dietary habits/their physical activity/their tobacco consumption/their alcohol consumption/their stress, I measure their blood pressure at rest, I ask them about their family history of CVD, I ask them about their job and/or education).^20,21 Here, we recoded the original answer categories [No patient, About a third, About half, About two thirds, (Almost) all] into measures provided to (almost) all patients vs. not provided to (almost) all patients.

To measure the attitude towards health promotion and prevention, we used eight statements (e.g., I can offer a wide range of lifestyle advice to my patients);^22,23 that the PCPs could rate (Completely true, Rather true, Rather not true, Not true at all). For the analysis we distinguish between PCPs that rated the statements as completely or rather true and those who rated them as rather not true or not true at all.

To measure potential barriers in lifestyle counselling, we asked the PCPs to what extent predefined barriers (e.g., Insufficient reimbursement) hamper them.^24-27 For the analysis, we combined the answer categories Completely true and Rather true, as well as the categories Rather not true and Not true at all.

Statistics

In order to compare individual and practice characteristics between female and male PCPs, we used chi² tests and Mann-Whitney U tests. We used chi² statistics and logistic regression models (crude OR and OR adjusted for PCPs’ age, years since residence, medical specialty, and number of patient contacts per week) to analyse the relationship between PCP gender and provision of prevention measures.

Additionally, we used chi² statistics to analyse the attitude towards prevention and health promotion (agreement in females vs. in males). We also analysed (predefined) potential barriers for prevention and health promotion in female and male PCPs (agreement in females vs. in males) using chi² statistics. P-values <0.05 were considered to be significant. All analyses were conducted with IBM SPSS Statistics Version 21 (IBM Corporation, Armonk, USA).

Results

In the ÄSP-kardio Study, 40.0% of the PCPs were female. Our analysis showed that female PCPs were on average younger than their male colleagues (49.9 vs. 52.4 years of age, P<0.001). While about one quarter of the female PCPs were 55 years of age and older (24.9%), 39.8% of male PCPs fell into this category. Female PCPs were less likely to work in a single practice compared to male PCPs (49.9% vs. 53.7%, P<0.001) and had a lower number of patient contacts per week (203.8 vs. 236.3, P<0.001).

Provision of prevention measures

Female PCPs were more likely to check the blood pressure of (almost) all of their patients compared to male PCPs (75.9% vs. 71.8%, P=0.004; OR=1.24, P=0.004). In addition, they were more likely to ask patients about their family history of CVD (67.3% vs. 59.3%, P≤0.001; OR=1.41, P≤0.001) and about their occupation and/or educational status than their male counterparts (55.8% vs. 47.0%, P≤0.001; OR=1.43, P≤0.001; Table 1).

Table 1

Prevention measures provided to (almost) all patients by female and male primary care physicians in Germany (ÄSP-kardio Study 2012).

Prevention measures provided to (almost) all patients	Bivariate results (chi²)		OR [CI]	Logistic regressions		P-value
Prevention measures provided to (almost) all patients	Proportion (%)	P-value	OR [CI]	P-value	OR_adj [CI]	P-value
Measuring weight and height		0.003		0.003		0.001
Female PCPs	35.4		1.23 [1.07-1.41]		1.27 [1.10-1.47]
Male PCPs	30.8		1.00 [Ref.]		1.00 [Ref.]
Asking about dietary habits		≤0.001		≤0.001		≤0.001
Female PCPs	28.4		1.62 [1.39-1.89]		1.63 [1.39-1.92]
Male PCPs	19.7		1.00 [Ref.]		1.00 [Ref.]
Asking about physical activity		≤0.001		≤0.001		≤0.001
Female PCPs	39.1		1.39 [1.21-1.59]		1.30 [1.13-1.50]
Male PCPs	31.7		1.00 [Ref.]		1.00 [Ref.]
Asking about tobacco consumption		0.002		0.002		0.008
Female PCPs	66.1		1.24 [1.08-1.42]		1.21 [1.05-1.40]
Male PCPs	61.2		1.00 [Ref.]		1.00 [Ref.]
Asking about alcohol consumption		≤0.001		≤0.001		≤0.001
Female PCPs	50.6		1.38 [1.21-1.57]		1.39 [1.21-1.60]
Male PCPs	42.7		1.00 [Ref.]		1.00 [Ref.]
Asking about stress		≤0.001		≤0.001		≤0.001
Female PCPs	47.2		1.78 [1.56-2.03]		1.75 [1.52-2.01]
Male PCPs	33.5		1.00 [Ref.]		1.00 [Ref.]
Measuring blood pressure		0.004		0.004		≤0.001
Female PCPs	75.9		1.24 [1.07-1.43]		1.33 [1.14-1.55]
Male PCPs	71.8		1.00 [Ref.]		1.00 [Ref.]
Asking about family history of CVD		≤0.001		≤0.001		≤0.001
Female PCPs	67.3		1.41 [1.24-1.61]		1.41 [1.22-1.63]
Male PCPs	59.3		1.00 [Ref.]		1.00 [Ref.]
Asking about job and/or education		≤0.001		≤0.001		≤0.001
Female PCPs	55.8		1.43 [1.26-1.62]		1.39 [1.22-1.60]
Male PCPs	47.0		1.00 [Ref.]		1.00 [Ref.]

PCPs, primary care physicians; OR, odds ratio; CI, confidence interval; OR_adj, odds ratio adjusted for PCPs’ age, years since residence, medical specialty, and number of patient contacts per week. n=4074 German PCPs.

We found that female PCPs were significantly more likely to ask their patients about dietary habits (OR=1.62, P≤0.001), physical activity (OR=1.39, P≤0.001), alcohol consumption (OR=1.38, P≤0.001), and tobacco consumption (OR=1.24, P=0.002). The highest OR was found for asking about stress (OR=1.78, P≤0.001), with 47.2% of female PCPs and 33.5% of male PCPs asking (almost) all patients about their perceived stress. The results remained stable after adjusting for age, medical speciality, years since residence, and number of patient contacts per week (Table 1).

Attitudes towards prevention and health promotion

There was no significant difference between male PCPs and female PCPs in their attitudes towards prevention and health promotion (Figure 1). Both groups said that they are not only responsible for the treatment of a disease but are also a kind of health advisor. However, female PCPs were more likely to feel well prepared for and successful in prevention. In contrast, male PCPs were more likely to state that it is difficult to give good advice to the patients.

Figure 1

Attitude towards prevention and health promotion of cardiovascular diseases among male and female primary care physicians in Germany (ÄSP-kardio Study 2012).

Potential barriers in prevention and health promotion

Female and male PCPs also differed in the perception of barriers that hamper the provision of prevention and health promotion in their daily business. Male PCPs were more likely to perceive the nine given barriers as relevant to their daily practice (Figure 2). The biggest difference in percentage points was found for the item too many prevention guidelines (8.8 percentage points), followed by unclear recommendations (7.5 percentage points). However, for female and male PCPs the most frequently mentioned barrier was insufficient reimbursement (89.0% vs. 92.1%, P=0.001).

Figure 2

Perceived barriers by male and female female primary care physicians in Germany (ÄSP-kardio Study 2012).

Discussion

Differences between female and male PCPs were prevalent in lifestyle counselling, attitudes towards prevention, and individual (e.g., age) and practice characteristics (e.g., patient contacts per week). Gender differences in the prevention measures remained stable after controlling for age, years since residence, medical specialty, and patient contacts per week. This result implies that the differences are not merely a result of the objective characteristics of the PCPs and their practices. Explanations why female PCPs are more likely to engage in prevention and health promotion are still lacking.²⁸ However, findings from gender research, psychology, and sociology might help to provide reasons for the differences in the attitude and practice of female and male PCPs.

A reason for the differences in lifestyle counselling between female and male PCPs might be the different subjective values between female and male PCPs.²⁹ These differences could result from the internalisation of gender roles during socialisation,³⁰ with parents, peers, teachers, and media determining what constitutes a woman and a man.³¹ It is important to note that today's socialisation process might be less gender stereotyped in western societies. However, when most of the included PCPs were being socialised, classic gender roles were more prevalent.³² Gender-role socialisation can lead to different hierarchies of personal values.²⁹ Female PCPs, for example, were more likely to be engaged in prevention and health promotion counselling,¹¹ Indeed, studies have shown that females tended to rate social values like helping others and caring for others higher compared with males.^29,33-35

Additionally, females are more likely to feel competent in social activities compared to males.³² Following Bandura,³⁶ these competence beliefs are a strong predictor for behaviour: If individuals believe that they are competent in a specific behaviour, they are more likely to behave like that.³² It is also possible that female PCPs are more likely to enjoy helping patients to modify their lifestyle to prevent CVD.³⁵ Previous studies showed that women in general are more likely to enjoy helping others and to show communal qualities than men.^37-39

However, besides these traditional gender roles, female PCPs need qualities for their non-traditional occupational roles.⁴⁰ Although female PCPs may internalize traditionally masculine characteristics (e.g., being active and ambitious), they often differ from men in their attitudes and behaviours.⁴¹ For example, female PCPs might invest more time in informal talks and be more patient oriented,⁴² while male PCPs may be more output oriented and focus on the number of patient contacts per week.

Indeed, the higher number of patient contacts per week for male PCPs could explain why male PCPs perceive more barriers in lifestyle counselling than their female counterparts. They have less time available per patient – and lifestyle counselling takes time, either directly (e.g., discussing lifestyle changes with non-compliant and undiscerning patients) or indirectly (e.g., acquiring knowledge on recommendations and guidelines). Taking more time with patients and establishing a more intimate relationship with them could lead to greater perceived success.

Strengths and limitations

When interpreting our results, several limitations should be considered. First of all, we cannot exclude a social desirability bias. It is possible that the PCPs gave particular answers to fit into their prescribed gender role. An alternative method would be direct observation; however, this approach can increase the risk of socially desirable behaviour. Second, the focus of the whole ÄSP-kardio Study was not primarily on gender differences but on lifestyle counselling in general. That is why we did not investigate the relationship between physician and patient gender.⁴³ Nonetheless, we followed the demand for investigating general prevention practices of female and male physicians in large samples.^8,11 Third, there might be a selection bias because PCPs that were more engaged in preventive measures might have been more likely to participate in this study. However, there were no significant differences between the participating PCPs and the total German PCP population with regard to gender, medical specialty, and region. Fourth, unfortunately, we cannot say how much training the PCPs received in prevention and health promotion. However, since the PCPs in this study had a mean age of 49.9 (women) and 52.4 (men), prevention and health promotion were not a core element in the medical curricula of their studies. Nowadays, a major focus is placed on these topics in medical studies in Germany.

Conclusions

Previous studies have shown first evidence of a link between PCP gender and prevention measures offered. Based on our results we could confirm findings of previous studies on gender differences in lifestyle counselling. Our study showed that female PCPs were more likely to be engaged in prevention, even after controlling for age, years since residence, medical specialty, and number of patient contacts per week.

Additionally, we were able to expand scientifically proven results to Europe by using a large sample size from a health care setting other than the United States, as called for by Jefferson et al.⁸ The perceived barriers to lifestyle counselling seem to be a topic of particular interest for further research, both with respect to gender differences and to the general significance of perceived barriers among female and male PCPs.

Overcoming the identified barriers may be an important starting point for enabling a larger portion of PCPs to offer preventive measures to all of their patients.¹⁶ Inadequate compensation for lifestyle counselling, for example, was rated as the most important barrier to offering such measures. Thus, if PCPs were financially better compensated for offering lifestyle counselling, it is conceivable that they would increasingly incorporate it into their daily routine. This might be an extrinsic motivation, particularly for output-oriented male PCPs.

Additionally, it is necessary to put a stronger focus on the prevention of CVD during both medical studies and postgraduate courses for PCPs. The gender differences revealed by our results imply that male PCPs in particular should be sensitized to the importance and the potential of such prevention. In general, a stronger focus on lifestyle counselling in needs-based advanced training for female and male PCPs could increase the use of lifestyle counselling in daily practice to prevent CVD.

Footnotes

Conflict of interests: the authors declare no potential conflict of interest.

Funding: we acknowledge the financial support of the Deutsche Forschungsgemeinschaft and Ruprecht-Karls-Universität Heidelberg within the funding programme Open Access Publishing

Acknowledgments

The ÄSP-kardio Study was funded by the German Research Foundation (reference number: SCHN 727/3-1). The funding institution had no influence on the study design, on the collection, analysis, and interpretation of data, on the writing of this article, and on the decision to submit this article for publication.

References

World Health Organization. Global health risks. Mortality and burden of disease attributable to selected major risks. Geneva: WHO; 2009.

Frijling

Lobo

Hulscher

. Provision of information and advice in cardiovascular care: clinical performance of general practitioners. Patient Educ Couns 2002;48:131–7.

Hølund

Thomassen

Boysen

. Importance of diet and sex in prevention of coronary artery disease, cancer, osteoporosis, and overweight or underweight: a study of attitudes and practices of Danish primary care physicians. Am J Clin Nutr 1997;65:2004S–6S.

Risha

. Women in the medical profession: international trends. In: Kuhlmann

Annandale

, eds. The Palgrave handbook of gender and healthcare. Basingstoke: Palgrave Macmillan; 2010. pp 423–428.

Ammerman

DeVellis

Carey

. Physician-based diet counseling for cholesterol reduction: current practices, determinants, and strategies for improvement. Prev Med 1993;22:96–109.

Journath

Hellenius

Carlsson

. Physicians’ gender is associated with risk factor control in patients on antihypertensive and lipid lowering treatment. Blood Press 2010;19:240–8.

Journath

Hellenius

Manhem

. Association of physician's sex with risk factor control in treated hypertensive patients from Swedish primary healthcare. J Hypertens 2008;26:2050–6.

Jefferson

Bloor

Birks

. Effect of physicians’ gender on communication and consultation length: a systematic review and meta-analysis. J Health Serv Res Policy 2013;18:242–8.

Clark

Potter

McKinlay

. Bringing social structure back into clinical decision making. Soc Sci Med 1991;32:853–66.

10.

National Association of Statutory Health Insurance Physicians. [Statistical information from the Federal physician registry]. Berlin: KBV; 2012. [In German]

11.

Ramirez

Wildes

Napoles-Springer

. Physician gender differences in general and cancer-specific prevention attitudes and practices. J Cancer Educ 2009;24:85–93.

12.

Whitlock

Orleans

Pender

Allan

Evaluating primary care behavioral counseling interventions: an evidence-based approach. Am J Prev Med 2002;22:267–84.

13.

Schneider

Diehl

Bock

. Modifying health behavior to prevent cardiovascular diseases: a nationwide survey among German primary care physicians. Int J Environ Res Publ Health 2014;11:4218–32.

14.

Bock

Diehm

Schneider

Physical activity promotion in primary health care: results from a German physician survey. Eur J Gen Pract 2012;18:86–91.

15.

Huy

Diehm

Schneider

[Cardiovascular prevention at the general practitioner? First results of a study on attitudes, services, success and barriers in practice]. Dtsch Med Wochenschr 2012;137:17–22. [Article in German]

16.

Diehl

Mayer

. Physical activity counseling by primary care physicians: attitudes, knowledge, implementation, and perceived success. J Phys Act Health 2015;12:216–23.

17.

Oriol-Zerbe

Abholz

. Primary prevention of cardiovascular diseases by lipid-lowering treatment in German general practice: results from GPs ignoring guidelines and risk calculators. Eur J Gen Pract 2007;13:27–34.

18.

Wiesemann

. Nutritional counseling in German general practices: a holistic approach. Am J Clin Nutr 1997;65:1957S–62S.

19.

Twardella

Brenner

Lack of training as a central barrier to the promotion of smoking cessation: a survey among general practitioners in Germany. Eur J Public Health 2005;15:140–5.

20.

Calnan

Cant

Williams

Killoran

Involvement of the primary health care team in coronary heart disease prevention. Br J Gen Pract 1994;44:224–8.

21.

McAvoy

Kaner

Lock

. Our healthier nation: are general practitioners willing and able to deliver? A survey of attitudes to and involvement in health promotion and lifestyle counselling. Br J Gen Pract 1999;49:187–90.

22.

Hulscher

van Drenth

Mokkink

. Barriers to preventive care in general practice: the role of organizational and attitudinal factors. Br J Gen Pract 1997;47:711–4.

23.

Steptoe

Doherty

Kendrick

. Attitudes to cardiovascular health promotion among GPs and practice nurses. Fam Pract 1999;16:158–63.

24.

Castaldo

Nester

Wasser

. Physician attitudes regarding cardiovascular risk reduction: the gaps between clinical importance, knowledge, and effectiveness. Dis Manag 2005;8:93–105.

25.

Kushner

. Barriers to providing nutrition counseling by physicians: a survey of primary care practitioners. Prev Med 1995;24:546–52.

26.

Tsui

Dodson

Jacobson

. Cardiovascular disease prevention counseling in residency: resident and attending physician attitudes and practices. J Natl Med Assoc 2004;96:1080–91.

27.

Walsh

Swangard

Davis

McPhee

. Exercise counseling by primary care physicians in the era of managed care. Am J Prev Med 1999;16:307–13.

28.

Frank

Harvey

. Prevention advice rates of women and men physicians. Arch Fam Med 1996;5:215–9.

29.

Eccles

. Gender roles and women's achievement-related decisions. Psychol Women Quart 1987;11:135–72.

30.

Phillips

Imhoff

. Women and career development: a decade of research. Annu Rev Psychol 1997;48:31–59.

31.

Bussey

Bandura

Social cognitive theory of gender development and differentiation. Psychol Rev 1999;106:676–713.

32.

Wigfield

Battle

Keller

Eccles

. Sex differences in motivation, self-concept, career aspiration and career choice: implications for cognitive development. In: De Lisi

De Lisi

, eds. Biology, sociology, and behavior: the development of sex differences in cognition. Greenwich: Ablex; 2002. pp 93–124.

33.

Feather

. Masculinity, femininity, psychological androgyny, and the structure of values. J Pers Soc Psychol 1984;47:604–20.

34.

Lyson

. Sex differences in the choice of a male or a female career line. Work Occup 1984;11:131–46.

35.

Eccles

Hoffman

. Socialization and the maintenance of a sexsegregated labor market. In: Stevenson

Siegel

, eds. Research in Child development and social policy. Chicago: University Press; 1984. pp 367–420.

36.

Bandura

. Self-efficacy: the exercise of control. New York : Freeman; 1997.

37.

Bridges

. Sex differences in occupational values. Sex Roles 1989;20:205–11.

38.

Eagly

Steffen

. Gender stereotypes stem from the distribution of women and men into social roles. J Pers Soc Psychol 1984;46:735–54.

39.

Mullis

Gerwels

Stability of vocational interests among high school students. Adolescence 1998;33:699–707.

40.

Prentice

Carranza

What women and men should be, shouldn't be, are allowed to be, and don't have to be: the contents of prescriptive gender stereotypes. Psychol Women Quart 2002;26:269–81.

41.

Gutek

. Women and paid work. Psychol Women Quart 2001;25:379–93.

42.

Roter

Lipkin

Jr. Korsgaard

Sex differences in patients’ and physicians’ communication during primary care medical visits. Med Care 1991;29:1083–93.

43.

Franks

Bertakis

. Physician gender, patient gender, and primary care. J Womens Health 2003;12:73–80.