Albina Rodrigues TorresI; Tânia RuizII; Sérgio Swain MüllerIII; Maria Cristina Pereira LimaI
IDepartment of Neurology, Psychology and Psychiatry of the Botucatu Medical School at the Universidade Estadual de São Paulo - UNESP
IIDepartment of Public Health of the Botucatu Medical School at the Universidade Estadual de São Paulo - UNESP
IIIDepartment of Surgery and Orthopedics of the Botucatu Medical School, at the Universidade Estadual de São Paulo
OBJECTIVE: This paper aimed to analyze self-evaluations in terms of quality of life (QoL), physical health (PH) and mental health (MH) of ex-medical students from a Brazilian public university, correlating these outcomes with demographic data and several professional aspects.
METHODS: a cross-sectional study with a target population of all students graduated from the Botucatu School of Medicine (UNIFESP - São Paulo State University) between 1968 and 2005. A self-administered questionnaire, which could be answered by regular mail or internet, was used.
RESULTS: From the 2,864 questionnaires that were sent by mail, 1,224 (45%) were answered and sent back. Good or very good QoL, PH and MH were reported by 67.8%, 78.8% and 84.5% of participants, respectively. In the final logistic regression model, positive QoL was associated with good PH and MH, regular attendance to scientific meetings, enough leisure time, and professional satisfaction. Good or very good PH was independently associated with positive QoL and MH, higher income level, regular physical activities, and never having smoked. Positive MH remained associated with professional satisfaction, enough leisure time, and positive evaluation of both QoL and PH.
CONCLUSIONS: Among medical doctors graduated from São Paulo State University, PH and MH were inseparable aspects, which were also related to the self-evaluation of QoL. Good habits, such as regular practice of physical exercise, enough leisure time, and not smoking were associated with positive health in general and should be encouraged. Professional satisfaction had an important impact on the emotional well-being of participants.
Keywords: Quality of life. Health. Mental health. Self-evaluation. Physicians.
The medical profession has highly demanding aspects, such as the requirement of great dedication of time, involvement with much social responsibility, and constant contact with patients' and family members' suffering1. Recent research performed by the Conselho Regional de Medicina do Estado de São Paulo (CREMESP - State of São Paulo Regional Council of Medicine) revealed that the mean number of working hours of interviewed doctors was 52 hours per week, that these had three different jobs on average, and that approximately one third of them surpassed this average. In addition to the number of employment relationships and long working hours, unsatisfactory working conditions (such as lack of recognition, adequate pay, autonomy, stability, infrastructure and safety) can further overload doctors, causing stress and negatively affecting their family life and quality of life (QoL)2.
According to Firth-Cozens3, the prevalence of doctors being stressed above the usual cut-off points is approximately 28%. Nonetheless, few studies have analyzed the mental (MH) and/or physical health (PH) of Brazilian doctors. More recent studies have been conducted with specific groups, such as doctors who work in intensive care units4,5, doctors on call in pediatric emergency departments6 or oncologists7-9. In general, poor working conditions are important predictors of psychological distress. Nascimento-Sobrinho et al.10 studied the working conditions of 350 doctors in the city of Salvador, BA, Brazil, and observed a high frequency of time burden, multiple jobs, low pays and professional instability, in addition to a prevalence of 26% of "minor" psychological disorders. By using the demand/control model, authors found that doctors who had high demand and low control over work (high strain) were three times more likely to have psychological disorders, when compared to colleagues who worked under low strain conditions. In an international literature review conducted in 2010, Lourenção11 approached resident doctors' health and QoL particularly and observed high incidences of stress, fatigue, sleepiness and depression in this population, suggesting the implementation of specific health care programs.
In a review published about 20 years ago, which still remains current, Martins12 emphasizes that stress during medical qualification and practice would be an etiological factor for the onset of mental health problems, including psychoactive substance abuse and dependence, work overload and occupational stress syndrome.
The present study was part of a survey conducted by the Faculdade de Medicina de Botucatu (FMB-UNESP - Botucatu Medical School of the Universidade Estadual de São Paulo), which aimed to evaluate the former medical students' opinion about their graduate course, their inclusion in the job market and professional satisfaction, in addition to their self-evaluation of QoL, MH and PH. The present study specifically approaches the three last aspects of this survey.
A cross-sectional study was conducted, in which all doctors graduated from the FMB - UNESP between 1968 (1st class) and 2005 (38th class) were considered potential participants. A questionnaire was especially developed and sent by mail in February 2007 to the target population of this study, i.e. a total of 2,864 former medical students. The questionnaire included questions about socio-demographic data (such as sex, age, marital status and place of residence), and data on graduation and continuing education (first and last year of course, medical residence, specialization, post-graduation, title of specialist, participation in scientific events, reading of periodicals), professional activity (specialization, employment relationships, work in different types of services, management or coordination jobs, income, level of stress due to the medical practice), QoL, leisure, MH, PH, and smoking habit, in addition to their opinion about their graduate course at the FMB (how prepared they felt to work in certain areas and to perform certain procedures). Sealed envelopes with a questionnaire, an explanatory letter and a stamped envelope were sent, which should be then sent back to the FMB with the anonymously completed questionnaire. An informed consent form was also included, to be returned in a separate envelope, thus enabling the control of refusals. The questionnaire was sent a second time in October 2007, in an attempt to reduce the number of these refusals. The returned questionnaires were subsequently typed. There was still the option of responding to the online questionnaire through the FMB website using a password, when former students preferred this. Online completion followed the same procedure of the written questionnaire, although including an adaptation to the virtual environment and having the signed consent form as a prerequisite for its completion. Self-evaluation of QoL, in addition to that of MH and PH, was performed, with five possible responses as parameter: very good, good, fair, poor or very poor, which were subsequently grouped.
The analysis was performed using quantitative methods in the Stata 10.0 software 13. First, a descriptive analysis of variables was made, followed by univariate analyses to identify possible associations, selecting the following as dependent variables: self-evaluation of QoL, PH and MH. Pearson's chi-square test was used (or Fisher's exact test, when recommended) for categorical variables, while Student's t-test was used for continuous variables. Next, a multivariate analysis was performed using stepwise logistic regression, thus obtaining the adjusted odds ratios (OR)14. A total of three logistic models were constructed, one for each dependent variable, where the explanatory variables that showed p<0.10 in the univariate analysis were included in the logistic regression models. Regardless of their statistical significance, age, sex and income were included in the logistic models, because the objective was to obtain adjusted OR not confounded with these variables. Age is a variable which is closely associated with physical health, whereas the female sex and income show a significant association with more common mental disorders15. The statistical significance level adopted was p < 0.01, due to the large sample size and to the high number of tabulations performed, which could result in mathematically significant results exclusively..
Questionnaires were not identified and their data were analyzed together, guaranteeing participants' anonymity, who signed an informed consent form to be included. The present study was approved by the Comitê de Ética em Pesquisa da FMB (Botucatu Medical School Research Ethics Committee), in November 2006.
Of all 2,864 questionnaires sent, 1,224 (44.9%) were completed. A total of 72 (2.5%) former students had addresses which had not been updated and thus were not found, and 64 deaths were reported (2.2%). The remaining questionnaires were not responded in the interval of one year, although a second questionnaire was sent by mail. As each class had 90 students, the response rate varied from 12.2% to 51.1% of medical graduates, and all classes were represented in the study.
Of all 1,224 former students who responded to the questionnaire, 791 (64.6%) were males and 411 (33.6%) were females, and there were no differences between the sample and universe of former students (p=0.24), evaluated by the Chi-square test. A total of 22 questionnaires (1.8%) did not include information about sex. Mean age of former students was 46.5 years ± 10.9, ranging from 24 to 72 years, of which 5.7% were aged up to 30 years; 22.3%, between 31 and 40 years; 28.4%, between 41 and 50 years; 26.2%, between 51 and 60 years; and 17.3%, more than 60 years. The majority of former students were either married or cohabiting (77.5%) and reported having at least one child (76.0%).
In addition, most of them (96.4%) lived in the state of São Paulo, of which 844 (70.4%) lived in cities in the countryside and 728 (61.1%) in cities with less than 500,000 inhabitants. A total of 98% reported practicing medicine and 66.1% affirmed they were professionally satisfied. With regard to the level of income, 10.5% reported earning less than R$ 5,000.00 per month (approximately US$ 2,500); 34.2%, between R$ 5,000.00 and R$ 10,000.00 (between US$ 2,500 and US$ 5,000); 28.6%, between R$ 10,000.00 and R$ 15,000.00 (between US$ 5,000 and US$ 7,500); 41.1%, between R$ 15,000.00 and R$ 20,000.00 (between US$ 7,500 and US$ 10,000); and 12.5%, more than this value.
Table 1 shows the frequencies of the three main indicators (QoL, PH and MH), self-assessed by participants. It was observed that 68% of former students considered their QoL as "very good or good" and that, likewise, 79% and 85% considered their PH and MH, respectively.
"Average, high, or very high" levels of stress were reported by 656 (56.3%) participants when dealing with situations of death of patients; 644 (54.7%), when dealing with seriously ill patients; 330 (27.7%), when communicating with patients and family members; and 363 (31.1%), when dealing with civil law suits. No or little stress in such situations was reported by the remaining participants, except for 29 who did not respond to these items in the questionnaire (data not shown).
Table 2 shows data on former students' QoL, leisure, physical activity practice, smoking habit, MH and PH, according to sex. Among men, ex-smokers and smokers predominated (60.7%), whereas "never smoked" was the most frequent response among women (78.6%). There were no significant differences between men and women in the remaining aspects. Only 52.3% of participants reported having sufficient leisure time. Moreover, with regard to lifestyle, 44.4% practiced physical activities at least three times per week, whereas 28.6% did not practice any regular activities (Table 2).
The QoL variable was categorized into "very good or good" and "fair, poor or very poor" for the multivariate analysis, whose results are shown in Table 3. In the final logistic regression model, "very good or good" QoL was found to be independently associated with: regular participation in congresses or scientific events, considering one's MH and PH as "good or very good", being satisfied with one's profession, reporting that one would take up medicine again, and having sufficient leisure time (Table 3).
The "self-evaluation of PH" variable was also categorized into "very good or good" and "fair, poor or very poor" for the multivariate analysis. "Good or very good" PH remained associated with the following independent variables: having a higher income, having never smoked, considering one's QoL and MH as "good or very good" and performing regular physical activity (Table 4).
When the "good and very good" levels of MH were grouped for the multivariate analysis, the following variables remained independently associated: being older, considering one's QoL and PH as "good or very good", having sufficient leisure time and being satisfied with one's profession (Table 5).
This is a pioneer study in terms of coverage, because its target population was the entire universe of former students who had graduated from the FMB-UNESP in the first 38 classes. It evaluated socio-demographic aspects and those related to professional satisfaction and inclusion in the job market, opinion about the course, occupational stress, QoL, MH and PH, these last three indicators being the objects of the present study. Thus, differently from the 2002 CREMESP survey2, which interviewed a sample of 400 doctors of this state exclusively, this study aimed to evaluate the entire population of former FMB students.
It was observed that 80% of former students evaluated their own PH and MH positively, despite all stressful factors and demands of their professional practice. Although this may have occurred as a result of response bias, i.e. former students with better health conditions have higher representativeness among the study participants, it is known that poor or very poor MH and PH conditions are practically incompatible with the professional practice. This result was similar to that found by the CREMESP 2, where such satisfaction was attributed to positive aspects of the medical practice, such as doing a relevant job that results in fulfillment or gratification; helping people; being valued; earning a reasonable income; and achieving status, respect or social recognition. The fact that a higher number of working hours did not remain associated with worse MH supports this hypothesis, suggesting that, although this may be important, aspects such as job satisfaction are relevant to determine the health status of doctors.
It should be emphasized that favorable MH and PH were positive and statistically associated with each other, as well as with "good or very good" QoL. In a recent article entitled "No health without mental health", published in The Lancet16, the authors stated again the long known fact that there is a close association between psychological problems or, strictly speaking, mental disorders and other general health conditions. Emotional problems increase the risk of communicable (such as AIDS and tuberculosis) or non-communicable diseases (such as cardiovascular diseases, diabetes and neoplasias) and contribute to the occurrence of injuries, whether intentional or not, such as self- or hetero-aggressions, and occupational and traffic accidents. Mental disorders are also associated with risk factors for chronic diseases, such as smoking, alcohol and drug use, little physical activity, unhealthy dietary habits and arterial hypertension. On the other hand, many health conditions increase the risk of mental disorders and this co-occurrence can hinder the search for services, diagnosis and treatment, with a negative impact on the prognosis of cases16. Thus, it was observed that mental disorders contribute to mortality and that they are important causes of incapacitation and dependence. In addition, in the 2005 World Health Organization (WHO) report17, 32% of all years lived with disability were attributed to neuropsychiatric conditions. The above mentioned publication has recently launched a movement18 that warns about the importance of MH in one's general health, emphasizing that MH care is still a neglected issue that, however, should be integrated into the remaining aspects of planning and offering of health services. Thus, greater attention should be paid to the emotional aspects of well-being and health, whose care must be fully included in the general public health policies, because psychosocial interventions can improve several physical health outcomes16, including that of professionals working in this area.
With regard to the positive self-evaluation of PH in particular, it could be observed that this was also independently associated with higher level of income, having never smoked and practicing physical activity at least three times per week. As a result, the association between healthy habits and lifestyle and good PH was confirmed, even when controlling for variables such as age and income. Although the cross-sectional design of the study does not enable the direction of causality to be inferred, it is possible that professionals with higher levels of income can focus on physical activities more regularly, which would have a positive impact on their PH.
With regard to MH, it is interesting to observe that there was an independent association between favorable assessment and older age, having leisure time and being satisfied with one's profession. Consequently, it is possible that more experienced professionals have a more stable and safer professional condition, including their financial condition, which has a repercussion on greater professional satisfaction and enables one to have more leisure time. In the survey conducted by the CREMESP in 2007, younger professionals worked for more hours, had a higher number of jobs and received lower salaries, when compared to the older ones2. A study conducted in Turkey by Uncu et al.19 found a significant association between negative emotional perceptions of work and the occurrence of depression, anxiety and stress among doctors working in primary care services. In a nationwide study conducted by Gouveia et al.20, older age was also associated with doctors' greater satisfaction - in this case, with life in general - in five Brazilian regions.
According to Firth-Cozens 3, the prevalence of doctors whose level of stress was above average totaled approximately 28%, in both longitudinal and cross-sectional studies. It is not this prevalence that has changed throughout the years, but rather the awareness of such professionals, who began to report their suffering more easily than before. In the present study, the prevalence of MH rated as "fair, poor or very poor" was only 18%. However, it should be noted that this survey did not use standardized instruments to identify psychological suffering.
According to Martins 21,22, there are few studies that analyzed Brazilian doctors' MH or PH, although the theme has deserved more attention in the literature. In this author's view22, psychologically unhealthy conditions inherent to the medical practice can cause psychological and psychiatric disorders in more vulnerable individuals. The development of new diagnostic and therapeutic technologies, the influence of the pharmaceutical industry and companies that provide medical services have considerably changed the practice of medicine. These changes have repercussions, such as the decrease of autonomy and income, increase in work demand, change in lifestyle and relationship with patients and colleagues, and effect on professionals' health23. Thus, despite the innumerable psychological gratifications and the possibility of material, emotional and intellectual fulfillment, the medical profession is usually highly stressful and anxiogenic23.
Certain recent studies evaluated psychological aspects in specific groups, such as doctors who work in emergency services, intensive care units (ICUs)4,5 and first-aid rooms; those on call in pediatric emergency units6; or oncologists7-9. In general, working conditions appear as important predictors of psychological suffering. In a study conducted by Cabana et al.4, although the difference did not reach statistical significance, the prevalence of common mental disorders (which include anxiety, somatoform and "minor" depressive disorders) was higher among professionals who worked in emergency services (32%), when compared to those who worked in ICUs or first-aid rooms (approximately 17%). In addition, the former showed less favorable working conditions, including a higher number of jobs and weekly working hours, greater feeling of work overload, and lower level of income. In a study conducted by Tironi et al.5, the prevalence of burnout syndrome among doctors working in ICUs was 7.4%, being closely associated with high psychological demand of work, including a high number of weekly on call hours, younger age, fewer years of professional practice, lower income, lack of physical activity practice or hobbies.
In a qualitative study conducted with professionals of pediatric emergency services, Feliciano et al.6 identified unequal feelings such as tiredness, anguish, fear of making mistakes and outrage caused by work overload and a salary incompatible with the responsibility and effort required and, on the other hand, satisfaction for enjoying the work and feeling useful. For this reason, authors concluded that there is a great need for health promotion actions in the work environment of professionals of this area.
An English study that evaluated the possible impact of working conditions on the health of young doctors24 found a positive correlation between the feeling of work overload and several long term measures for work performance, and MH and PH. In a different study,25 the same authors reported that young doctors had several health complaints and that 30% of them were classified as "positive cases" in relation to the cut-off point for psychiatric symptoms. Moreover, many showed maladaptive patterns of self-care, such as continuing to work when one is not feeling well, not going on sick leaves, and self-medicating oneself, in addition to the tendency of informally consulting friends instead of setting up regular consultations. Gardner and Ogden 26, while studying English general practitioners, reported that many do not do for themselves what they recommend their patients should do. Martins23 reports the same type of posture in Brazil, where the difficulty to seek a colleague as a patient causes many doctors to have a tendency towards self-diagnosing and self-medicating themselves.
In a study conducted by Glasberg et al.9, moderate or severe levels of burnout syndrome (which involves emotional exhaustion, depersonalization and low personal fulfillment) occurred in 16% of the oncologists studied. Lower levels of the syndrome were associated with the following: older age, being in a stable marital relationship, participating in a religious group, practicing physical activities or, having a hobby, and sufficient vacation time. Lack of personal time was the main factor associated with the burnout syndrome among oncologists as well7. It is interesting to observe that, in the present study, leisure remained associated with positive evaluation of MH, but not of PH or QoL.
Certain limitations to this study must be taken into consideration. The first one is the response rate, a key issue in cross-sectional studies. The percentage of responses obtained (45%) can be considered satisfactory, because a rate varying from 30% to 40%27 is what is expected of a mail survey and because all classes were represented in the sample. In addition, even if a higher rate were ideally expected, the rate obtained was higher than those of similar studies conducted in Brazil, which varied from 25.0% to 32.1%, in the Faculdade de Medicina de Ribeirão Preto (Ribeirão Preto School of Medicine)28 and in the Universidade Luterana do Brasil (Brazilian Lutheran University)29, respectively. Moreover, the number of deaths (only 2.2%) is probably underestimated. Another aspect to be emphasized is that the cross-sectional design of the study only indicates associations between the outcomes of interest and possible explanatory variables, not enabling conclusions about causality to be made. Another question which cannot be disregarded is the possibility of response bias, because former students with better self-evaluation of their PH and/or MH may be more likely to participate in the survey. The question of demand/control30 and that of participants' specialty were not evaluated, which could also be associated with health and QoL31. Finally, this survey did not use standardized instruments to evaluate health problems, psychological suffering or QoL, only direct questions about these aspects, with five response choices.
Despite professional practice demands, approximately 80% of FMB-UNESP former students reported good levels of MH or PH. A direct and significant association between good PH and good MH emphasizes the inseparability of these two aspects, which should be considered in an integrated way, when planning not only therapeutic actions, but also possible health promotion and illness prevention actions among doctors. It should be noted that these two aspects, like the two sides of a coin, are also significantly associated with participants' better evaluation of QoL. In addition, good life habits, such as practicing physical activities regularly, having leisure time and not smoking, were associated with better health evaluation in general and should be promoted, whenever possible. On the other hand, favorable MH was associated with professional satisfaction. Considering the fact that the majority of doctors spends a great part of their time with professional activities, job satisfaction - whose positive aspects cannot be ignored - seem to have an important influence on their emotional well-being.
In conclusion, the present study identified certain relevant factors that were independently associated with each of the outcomes evaluated. It is essential for professionals in the medical area to adequately care for their own PH and MH, because these are directly associated with one another and also with self-evaluation of QoL. It is necessary that doctors pay attention to their own needs to better perform the important task of caring for the health of others.
Authors would like to thank FMB management professionals André Franco Pagnin and Cristina de Almeida, responsible for the organization of the printed material sent to the former students and for the distribution and receiving of questionnaires; the FMB Technical Academic Management team, coordinated by Mrs. Eliane Sako, who helped to obtain former students' addresses with the CREMESP and CFM; and Denise M. Zornoff, chief of the FMB Núcleo de Ensino à Distância (NEAD -Distance Learning Center), who created the virtual interface for questionnaire responses, thus helping all process stages.
1. Pitta A. Hospital, dor e morte como ofício. São Paulo: Editora Hucitec; 1990.
2. CREMESP. Conselho Regional de Medicina do Estado de São Paulo. O trabalho médico no Estado de São Paulo. 2007. Disponível em http://www.cremesp.org.br/library/modulos/ centro_de_dados/arquivos/mercado_de_trabalho.pdf. [Acessado em outubro de 2010]
3. Firth-Cozens J. Doctors, their wellbeing, and their stress: it's time to be protective about stress and prevent it (editorials). Brit Med J 2003; 326: 670.
4. Cabana MCFL, Ludermir AB, Silva ER, Ferreira MLL, Pinto MER. Transtornos mentais comuns em médicos e seu cotidiano de trabalho. J Bras Psiquiatr 2007; 56(1): 33-40.
5. Tironi MOS, Nascimento Sobrinho CL, Barros DS, Reis EJF, Marques Filho ES, Almeida A, et al. Trabalho e síndrome da estafa profissional (burnout) em médicos intensivistas de Salvador Rev Assoc Med Brasil 2009; 55(6): 656-62.
6. Feliciano KVO, Kovacs MH, Sarinho SW. Sentimentos de profissionais dos serviços de pronto-socorro pediátrico: reflexões sobre o burnout. Rev Bras Saúde Matern Infant 2005; 5(3): 319-28.
7. Whippen DA, Zuckerman EL, Anderson JW. Burnout in the practice of oncology: results of follow-up survey. J Clin Oncol 2004; 22(S14): 605-13.
8. Tucunduva LTCM, Garcia AP, Prudente FVB, Centofanti G, Souza CM, Monteiro TA et al. A síndrome de estafa profissional em médicos cancerologistas brasileiros. Rev Assoc Med Brasil 2006; 52(2): 108-12.
9. Glasberg J, Horiuti L, Novais MAB, Cavanezzi AZ, Miranda VC, Chicoli FA, et al. Prevalence of burnout syndrome among Brazilian medical oncologists. Rev Assoc Med Brasil 2007; 53(1): 85-9.
10. Nascimento-Sobrinho CL, Carvalho FM, Bonfim TAS, Cirino CAS, Ferreira IS. Condições de trabalho e saúde mental dos médicos de Salvador, Bahia, Brasil. Cad Saúde Publ 2006; 22(1): 131-40.
11. Lourenção LG, Moscardini AC, Soler ZASG. Saúde e qualidade de vida de médicos residentes. Rev Assoc Med Brasil 2010; 56(1): 81-91.
12. Martins LAN. Morbidade psicológica e psiquiátrica na população médica. Bol Psiquiatr 1990; 23: 9-15.
13. StataCorp. Stata Statistical Software: Release 10. College Station, Texas: StataCorp LP; 2007.
14. Kleinbaum DG. Logistic Regression: A self-learning text. New York: Springer; 1994.
15. Patel V, Araya R, Lima M, Ludermir A, Todd C. Women, poverty and common mental disorders in four restructuring societes. Soc Sci Med 1999; 49: 1461-71.
16. Prince M, Patel V, Saxena S, Maj M, Maselko J, Phillips MR, et al. No health without mental health. Lancet 2007; 4: 13-31.
17. World Health Organization (WHO). Mental health: facing the challenges, building solutions. Report from the WHO European Ministerial Conference. Copenhagen, Denmark: WHO Regional Office for Europe; 2005.
18. Horton R. Launching a new movement for mental health. Lancet 2007; 4: 1.
19. Uncu Y, Bayram N, Bilgel N. Job related affective well-being among primary health care physicians. Eur J Publ Health 2007; 17(5): 514-9.
20. Gouveia VV, Barbosa GA, Andrade EO, Carneiro MB. Medindo a satisfação com a vida dos médicos no Brasil. J Bras Psiquiatr 2005; 54(4): 298-305.
21. Martins LA. Atividade médica: fatores de risco para a saúde mental do médico. Rev Bras Clin Ter 1991; 20(9): 355-64.
22. Martins LAN. Vicissitudes do exercício da medicina e saúde psicológica do médico. Arq Bras Endocrinol Metab 1995; 39(3/4): 188-93.
23. Martins LAN. Saúde mental dos profissionais de saúde. Rev Bras Med Trab 2003; 1(1): 56-68.
24. Baldwin PJ, Dodd M, Wrate RM. Young doctors' health - I. How do working conditions affect attitudes, health and performance? Soc Sci Med 1997; 45(1): 35-40.
25. Baldwin PJ, Dodd M, Wrate RM. Young doctors' health - II. Health and health behaviours. Soc Sci Med 1997; 45(1): 41-4.
26. Gardner M, Ogden J. Do GPs practice what they preach? A questionnaire study of GPs' treatments for themselves and their patients. Patient Educ Couns 2005; 56(1): 112-5.
27. Prince M. Cross-sectional surveys. In: Prince M, Steward R, Ford T, Hotopf M. Practical Psychiatric Epidemiology. Oxford: Oxford University Press; 2003. p. 111-29.
28. Souza GMB, Cruz EMTN, Cordeiro JA. Perfil do egresso da Faculdade de Medicina de São José do Rio Preto. Rev Bras Educ Med 2002; 26(2): 105-14.
29. Caovilla F, Leitzke L, Menezes HS, Martinez PF. Perfil do médico egresso do Curso de Medicina da Universidade Luterana do Brasil (Ulbra). Rev AMRIGS 2008; 52(2): 103-9.
30. De Jong J, Dollard MF, Dormann C, Le Blanc PM, Houtman ILD. The Demand-Control Model: specific demands, specific control, and well-defined groups. Int J Stress Management 2000; 7(4): 269-87.
31. Braga LC. Condições de trabalho e saúde dos profissionais da rede básica de saúde de Botucatu, SP [dissertação de mestrado]. Programa de Saúde Coletiva da Faculdade de Medicina de Botucatu: UNESP; 2007. Received: 04/05/10 Conflitos de Interesses: inexistentes
Final version: 21/10/10
Conflitos de Interesses: inexistentes