Self-rated health among physicians

Summary. The aim of the study was to analyze self-rated health among physicians depending on their sex, age, workplace (hospital or polyclinic), and specialty. Material and methods. The studied group consisted of 377 26–70-year-old physicians randomly selected from various county hospitals and polyclinics of Lithuania. There were 85 men and 292 women. The inquiry was performed using the complemented (by the authors of the study) version of the WHO anonymous questionnaire of the quality of life ( 1 995). Responses were evaluated based on physicians’ evaluation of their own health, which was rated as very good, good, satisfactory, poor, and very poor. Results. Only 8.2% of males and 5.8% of females evaluated their health as very good (P>0.05). More men, compared to women, evaluated their health as good (62.3% and 53. 1 %, respectively; P<0.05), health satisfactory, males and 25.9%, respectively;


Introduction
A number of studies have been performed -both in Lithuania and elsewhere in the world -on selfrated health in various groups of population and on factors that might influence people's health. Health and its evaluation is one of the indicators of the quality of life, and therefore studies of such type are very common. Meanwhile, scientifically substantiated in-formation about self-rated health among physicians in relation to their sex, age, workplace (hospital or polyclinic), etc. is very scarce. This creates and impression that physicians as a separate occupational group has become undeservedly marginal and ignored by physicians themselves. Health, social and other issues of medical personnel have not been extensively analyzed in most countries, and thus information on this subject is insufficient if published at all, and it is mostly found in nonscientific literature. It is noteworthy that sparse studies performed in this field most frequently deal with factors affecting physical health rather than with the evaluation of one's health. The majority of studies analyze physicians' stress-related problems, social issues, and relationships with colleagues or in the family (1). Literature sources emphasize that physicians are negatively affected by the lack of time and physical and mental load during working hours (2)(3)(4). Although responsibility for patients' health makes physicians satisfied with their work, they frequently feel physically and mentally exhausted. The reason for this is frequent encounters with other people's (patients') pain, suffering, and death. In addition to that, physicians' work is associated with continuous competition and tension in working relationships not only with the colleagues, but also with the administration (5).
The majority of studies are limited to the evaluation of the mental health of students of medicine, resident physicians, or assisting physicians. Studies showed that as much as 70% of Australian students of medicine frequently felt exhausted (6); in the United States, the percentage of such students was 76%, and onehalf of them were diagnosed with both exhaustion and depression (7). Even 8.8% of German students of medicine had pronounced depression, and 5.1% felt fear (8). Cases of depression among students of medicine in the United States and Canada are threefold more common than in the general population (9). In the United States, 12.8% of male physicians and 19.5% of their female colleagues were diagnosed with pronounced depression (10). It is thought that stress of various origins has attracted so much attention because it has been evidently proven as one of the major risk factors for cardiovascular diseases (11). It must be emphasized that foreign scientific literature frequently presents only presumptions about physicians' health without providing any concrete data. In Lithuania, physicians' health on the population level has not been evaluated yet, and the available information about factors affecting physicians' health is scarce. The majority of studies have been oriented toward the evaluation of physicians' fatigue and nervous stress. Studies have shown that family physicians and ambulance crews frequently experience negative emotions; in particular, family physicians more frequently experience negative rather than positive emotions (12,13). The 2007 study of clinicians' level of happiness complemented previous studies and showed that 62.3% of physicians felt themselves happy and only 7.5%unhappy. The study found that happiness depended on marital status, satisfaction with marriage, and love relationships (14). This suggests that family is the main condition of mental health determining the relationships not only among family members, but also among colleagues.
The aim of this study was to analyze physicians' self-rated health with respect to their sex, age, length of service, and workplace.

Material and methods
A sociological inquiry of the physicians was performed during August-September 2006, using the WHO quality-of-life questionnaire (15) that we complemented with additional questions of interest.
At the beginning, we performed a pilot study involving 44 physicians. This study allowed for determining the validity of the questionnaire, i.e. whether the respondents equally understood the questions. After the evaluation of the obtained results, we adjusted six questions and repeatedly performed the pilot study with the same subjects. Thirty-six physicians were included in the latter study. We determined the kappa coefficient (16) that in this case was 0.71 and indicated that the degree of agreement was substantial. The first page of the questionnaire stated the aim of the study and provided the instruction for filling in the questionnaire and assurance of the anonymity and confidentiality of the data.
During the next stage, we sent applications (including the questionnaire for familiarization) to the heads of hospitals and polyclinics, asking permission to perform the inquiry of physicians in their healthcare institutions. Healthcare unites were randomly selected, taking into account the counties in which they were operating. However, some institutions refused to participate in studies of such character, and thus we could not strictly observe random selection of the healthcare units, limiting the selection to the units that agreed to participate in the study. The principle of the representation of counties was maintained.
The inquiry included physicians (who were at work on the day of the inquiry) of hospitals and polyclinics of major Lithuanian cities and counties (Vilnius, Kaunas, Klaipėda, Šiauliai, and Utena). In total, 425 anonymous questionnaires were distributed; 399 (93.9%) questionnaires were received, of which 22 (5.5%) were unsuitable for the analysis (not fully answered). In total, 377 of the questionnaires were analyzed, which comprised 88.7% of the total number of the distributed questionnaires.
The respondents were distributed into groups according to their age, sex, workplace (the healthcare Medicina (Kaunas) 2009; 45 (7) institution they were working in), and specialty. Age was divided into intervals following the variation analysis of the age (a continuous value) of the studied group. In total, the inquiry included 85 males and 292 females; 149 (39.5%) of the studied physicians worked in hospitals, and 218 (57.8%) -in polyclinics.
Statistical data processing was performed using SPSS v.13 software. The evaluation of the relationships between attributes was performed using Pearson's (for continuous values) and Spearman's (for discrete values) correlation coefficients. For the analysis of category data, we used χ² criterion and df (number of the degrees of freedom). The data were processed by calculating absolute values, percentage, and 95% confidence intervals. The interpretation of the association strength of correlation coefficient r that was applied for the evaluation of symmetrical continuous relationships was the following: up to 0.2, weak relationship; 0.3-0.5, moderate relationship; 0.6-0.7, strong relationship; and 0.8-1.0, very strong relationship. The results were considered as statistically significant if P value was <0.05.

Results
The study showed that men and women ( Fig. 1) presented different evaluations of their health (P=0.05); 88.2% of men and 89.1% of women (P>0.05) evaluated their health as good or satisfactory. Only 11.1% of the respondents provided other evaluations of their health.
However, statistically significantly more men and women evaluated their health as "good" rather than "satisfactory" (P<0.001) or other ("very good," "poor," or "very poor"). Only one out of the 377 respondents (0.3%) evaluated his health as "very poor." This respondent was a 65-year-old male with more than 30year experience of working at district hospital.
The analysis of the findings showed that physicians working in city and district hospitals and polyclinics presented different evaluations of their health ( Table 1).
The findings of the study showed that differing evaluations of their health were presented by physicians working in city and district hospitals (P=0.006), city hospitals and polyclinics (P<0.001), and in city hospitals and district polyclinics (P=0.002). Meanwhile, the comparison of physicians' evaluations of their health among district hospitals and city and district polyclinics, and among city and district polyclinics showed no differences (P>0.05).
A significantly higher percentage of physicians who evaluated their health as "very good" was observed among those working in city hospitals, compared to any other studied personal healthcare units (P<0.001). More than half (55.6%) of the studied physicians evaluated their health as "good." The percentage of physicians who presented such evaluations was similar in all studied healthcare units (P>0.05). Somewhat different situation was observed among physicians from different healthcare units who eva-  (7) Self-rated health among physicians % luated their health as "satisfactory." The percentage of such physicians was significantly higher (P<0.05) in city and district polyclinics and district hospitals than in city hospitals. The analysis of the physicians' evaluation of their health according to age (Table 2) showed that respondents of different age provided differing evaluations. Different evaluations were also found according to health categories, except for "poor health." In this category, the evaluations were similar in all age groups.
The percentage of physicians who evaluated their health as "very good" in the age groups of 26-37 and 38-43 years was similar, but significantly exceeded the respective percentage in the age groups of 44-48, 49-55, and 56-70 years (P<0.05). Compared to other age groups, the lowest percentage of physicians who evaluated their health as "good" was in the age group of 56-70 years (P=0.05); no differences between other age groups in this respect were detected.
The percentage of physicians who evaluated their health as "satisfactory" was similar in all age groups except for the age groups of 56-70 years where the percentage of such respondents was significantly (P<0.001) higher than in the age groups of 38-43 years.
Different tendencies in the evaluation of health emerged in different age groups. Among 26-37-yearold physicians, the evaluation of one's health according to health categories differed (P<0.05), except for those who evaluated their health as "very good" or "poor" -no significant difference in the percentage was found here. No significant differences were found in the age group of 38-43 years between those who evaluated their health as "very good" or "satisfactory" (P>0.05) and between those who evaluated their health as "very poor" or "poor" (P>0.05), in the age group of 44-48 years -between those who evaluated their health as "very good" and "poor" or "very good" and "very poor" (P>0.05), in the age group of 49-55 yearsbetween physicians who evaluated their health as "very good" and "poor," "very good" and "very poor," or "good" and "satisfactory," and in the age groups of 56-70 years -between the respondents who evaluated their health as "good" or "satisfactory" (P>0.05).
The study also showed that the evaluation of one's health among physicians tended to worsen with advancing age (r=-0.250; P=0.01).
Findings of the study show that physicians working in these institutions presented differing evaluations of their health (χ²= 22.23, P<0.001). The majority of physicians working in hospitals (83.2%) and polyclinics (93.1%) evaluated their health as "satisfactory" or "good," but significantly higher percentage of such evaluations was observed among physicians working in polyclinics (P=0.004). It must be emphasized that more than one-half of physicians working in the studied healthcare units (56.3% of physicians who worked in hospitals, and 55.1% of physicians working in polyclinics) indicated that their health was good (P>0.05). However, the percentage of physicians who evaluated their health as "very good" was higher in hospitals than in polyclinics (P<0.0001).
The evaluation of physicians' health according to their specialty (Fig. 2) showed that surgeons, general practitioners (GPs), and therapists evaluated their health differently (χ²=77.93, P<0.001). A higher percentage of surgeons, compared to GPs and therapists, evaluated their health as "very good" or "satisfactory" (P<0.05). Surgeons also more frequently evaluated their health as "good," compared to therapists (P<0.05). Meanwhile, the comparison of the evaluation of one's health presented by GPs and therapists yielded no differences in any of the cases (P>0.05).

Discussion
In Lithuania, like in other countries, especially large amounts of data have been accumulated on selfrated health among schoolchildren, youth, and other groups of population, as well as on the factors influencing health status in these groups. Studies showed that Lithuanian schoolchildren presented poorer evaluations of their health, compared to their peers in other countries: as many as 21% of girls and 9.9% of boys stated that they felt themselves "not very healthy." Among 28 countries that participated in the study, Lithuania occupied penultimate position. For comparison, only 2.6% of Finnish schoolgirls and 1.3% of schoolboys of respective age stated that they felt "not very healthy" (17). According to the findings presented by other authors, 30-35% of girls and 17-21% of boys presented such evaluations of their health (18)(19)(20). Schoolchildren who experienced bullying at schools provided significantly poorer evaluation of  (7) Self-rated health among physicians % their health and frequently complained of pain of different localization, nervous tension, and insomnia (21). It is noteworthy that representatives of youth and other groups of population presented poorer evaluations of their health than schoolchildren did (22)(23)(24)(25)(26)(27)(28). Literature also provides data about subjective health evaluation among men as a prognostic indicator of the probability of death and mortality from cardiovascular diseases and ischemic heart disease (29). Detailed analysis of the findings of the study revealed the characteristics of physicians' self-rated health. More than one-half (55.2%) of the physicians evaluated their health as "good," and 33.7% as "satisfactory." Only 4.5% of the respondents stated that their health was "poor." Although it was expected that a significant part of the studied physicians would evaluate their health as "very good," only 6.4% of the respondents presented such evaluations. Even in the age groups of 26-43 years, the percentage of physicians who evaluated their health as "very good" was only 12.1-12.3%. The evaluation according to age groups revealed a similar situation. Our findings do not provide a direct explanation of this phenomenon. So far, only certain assumptions may be made. It is common knowledge that a physician's work is highly specific and responsible, and is associated with continuous psychological and physical stress. Due to low salaries, newcomers and physicians with limited experience are forced to work in several workplaces thus trying to solve financial problems in the family. Like in case of other young people, socioeconomic problems faced by young physicians are especially tender -low salaries, unsolved problems of living place, personal transport, the choice of desired kindergarten and school for one's children, and other problems affect family relationships, and enormous responsibility and continuous physical and mental stress at work create an atmosphere of dissatisfaction with work, which in the long run becomes one of the major causes of poor health. Another fact has to be mentioned. According to the number of physicians per 10 000 population, Lithuania occupies one of the leading positions in Europe (30); this would indirectly point at lower workloads and shorter waiting lists. However, in reality, this is not the case -physicians' workload is immense, and waiting lists for examinations are not getting any shorter. In our opinion, long waiting lists affect not only the patients' but also the physicians' health. The reason for this phenomenon lies in the healthcare system and healthcare policy. Physicians spend a significant amount of their working time doing social rather than their direct medical work, including endless paperwork, complicated filling of sick-leaves, etc., which should not be a part of medical services. It can be stated that trying to establish their positions in this complicated life, young and most able-bodied physicians do so at the expense of their health. We think that qualitative studies will confirm these assumptions. It must be emphasized that irrespectively of age, 63.2% of physicians (70.6% of males and 59.1% of females; P=0.058) evaluated their health as "very good" or "good." The comparison of these findings with the data on the general Lithuanian population showed that very good and good health status was observed in 41.5% (43.4% of males and 40.1% of females) of the Lithuanian population (30). Evidently, physicians evaluated their health better than other inhabitants of Lithuania of respective age. The findings of contemporaneously performed studies of the self-rated health among physicians and youth showed that young people evaluated their health similarly to physicians: 64.7% of young males and 59.2% of females evaluated their health as "very good or good" (22). These findings evidently show that socioeconomic problems of physicians are essentially similar to those faced by other Lithuanian population, but physicians -due to the specificity of their work -more frequently experience concurrent major physical and emotional stress, compared to other occupational groups. On the other hand, physicians are fully aware of possible consequences of health risk factors and know how to avoid those consequences, adhering to healthy lifestyle principles at work and in personal life and passing their experience to their patients and the general population who, unfortunately, frequently ignore those principles.

Conclusions
1. Only 6.4% of the studied physicians (8.2% of males and 5.8% of females) evaluated their health as "very good," 55.2% (62.3% of males and 53.1% of females) as "good," 33.7% (25.9% of males and 36% of females) as "satisfactory," 4.7% (2.4% of males and 5.1% of females) as "poor," and one male evaluated his health as "very poor." 2. Physicians' self-rated health was age-dependent, but the percentage of physicians who evaluated their health as "poor" was similar in all studied age groups and ranged between 2.5% and 6.7%.