Apoio social no trabalho : um estudo de coorte com servidores de uma universidade pública Social support at work : a cohort study with civil servants from a public university

REV BRAS EPIDEMIOL 2019; 22: E190068 RESUMO: Introdução: O apoio social no trabalho, relacionado à interação entre colegas e chefias na cooperação para a realização do trabalho, pode contribuir para a diminuição do desgaste do trabalhador e dos riscos à saúde. Portanto, o presente estudo objetivou analisar o apoio social no trabalho e os fatores associados dos servidores técnico-administrativos em educação de uma universidade pública. Metodologia: Estudo de coorte, com 328 servidores em exercício ativo de sua função, dos quais as informações sobre o apoio social no trabalho e as variáveis independentes sociodemográficas, relacionadas ao trabalho e à saúde, foram coletadas por meio de questionário. Os testes t de Student, do χ2 e a regressão logística foram utilizados, a fim de analisar as prevalências e os fatores associados ao apoio social no trabalho. Resultados: Os trabalhadores eram, predominantemente, homens, com idade média de 47 anos, casados, com filhos, com nível universitário ou mais, apresentaram boas condições de trabalho e de saúde e alto apoio social no trabalho (85,7%). Os fatores associados ao apoio social no trabalho foram o turno de trabalho, a depressão e a capacidade para o trabalho. Conclusão: Os fatores associados ao apoio social devem ser devidamente analisados para que seja conservada essa interação positiva no ambiente de trabalho.


INTRODUCTION
Humans, unlike most other living creatures, are inherently social beings living in a community/society, with which they interact, establishing the most diverse relationships 1 .Due to this social character of human beings, for many years, some authors have suggested that the rupture of social ties affects the body defense system, making the individual more susceptible to diseases and influencing the health maintenance, in addition to hindering adaptive behaviors in stress situations 2,3 .Other authors 4 , when investigating the influence of social ties on the risk of falling ill and dying, found that being part of a social network and receiving help from individuals who belong to it can, in fact, benefit the health and well-being.
Considering the above, social network and support are significant in maintaining the health of individuals.Social network relates to the group of people with whom the individual can keep contact or some kind of social bond, while social support concerns the resources that other people, such as friends and family, can provide in situations of need 5 .Thus, we can assume that work integrates with the social network of individuals, particularly when taking into account that people spend most of their day at work 6 .
However, when labor activities are developed under inadequate environmental, organizational, and physiological conditions, health damage can be accelerated or exacerbated as a result of these activities 7 .Therefore, it is necessary to evaluate the social relationships established in the work environment as a factor that can also affect the health of individuals, especially when analyzed together with other work-related aspects.
Palavras-chave: Saúde do trabalhador.Apoio social.Trabalho.Servidores públicos.Estudos de coortes.the 1970s and early 1980s, focused on researching the causes of work-related stress and their effects on health.With the introduction of the social support dimension, the scale was renamed to Swedish Demand-Control-Support Questionnaire (DCSQ), in which the social support at work is related to the interaction between co-workers and supervisors in cooperation toward work achievement and can contribute to reducing the strain on workers and health risks [8][9][10][11] .
Negeliskii and Lautert 12 consider that social support should be the basis of work relationships and a social organization strategy in institutions, as it can decrease or even prevent occupational stress.These authors further highlighted that valuing work relationships and environment benefits the health and ability of workers 12 .
Studies on social support at work relating to different aspects of labor activities have been carried out with workers from numerous areas and sectors; among them, we underline the workers who act on the public sector, particularly on universities [13][14][15] .Nevertheless, in Brazil, workers from the public sector are stereotyped as inefficient and costly, and, for this reason, derogatory names are associated with both the institutions and the civil servants 16 .
Integrating the group of university workers is the technical-administrative staff in education (técnico-administrativos em educação -TAEs), who stand out for encompassing a variety of roles within different training/working areas, covering positions that require schooling levels ranging from elementary to higher education 17 , representing the heterogeneity of the population, and being of great scientific importance to the field of worker's health.
In 2012, a study carried out with TAEs from a public university in Minas Gerais found an increased prevalence of civil servants with high social support at work 17 .Thus, the present study aimed to analyze the social support at work and its associated factors among TAE civil servants from a public university to give continuity to the research mentioned above.

METHODS
This prospective cohort study, with a quantitative approach, is the second stage of the base-study "Work ability and associated factors of Brazilian technical-administrative workers in education."The present study was carried out at the Universidade Federal de Juiz de Fora (UFJF), located in the city of Juiz de Fora, state of Minas Gerais, and the population of this study consisted of permanent TAE workers from this university.The workers were invited to participate in the research through direct contact (in-person at the workplace), telephone, or e-mail, when both parts agreed to meet in a place that ensured the privacy of the interviewee, where they would receive and sign the Informed Consent Form and be interviewed.
The present investigation considered as inclusion criteria: having participated in the base-study between January 2012 and April 2013.The exclusion criteria were: having filled the questionnaire of the base-study incorrectly or incompletely; being off-work due to sick leave, maternity leave, or leave of absence granted by the National Institute of Social Security; having leave to accompany a spouse; or having been transferred to another institution.
Civil servants who were not found after three attempts, who retired for the length of service, age, or disability, those dismissed, or who died were regarded as losses.We also respected the individual's right not to continue participating in the study, and these cases were considered refusals.
Data were collected through interviews conducted by trained researchers or self-administered questionnaires between August 1, 2016 and April 28, 2017, that is, 4 years after the conclusion of the base-study.
We characterized the profile of civil servants by analyzing variables related to sociodemographic characteristics (age, gender, marital status, ethnicity, schooling, number of children), work aspects (age at first job, number of jobs, weekly workload, total length of service, length of service at UFJF, night shift, psychosocial stress at work, social support at work), and health status (signs and symptoms of depression, physical activity, and work ability) 18,19 .
Work conditions were analyzed through issues concerning the occupational history and by using the DCSQ, a reduced scale that addresses stress and social support at work adapted to Portuguese 10 .The Demand-Control Model covers four specific work situations: high-strain job; passive job, active job; and low-strain job.Social support at work has two possible classifications -low or high support 9,20,21 .
Information about health status was collected using the following instruments: • Patient Health Questionnaire (PHQ-9), brief instrument that evaluates, diagnoses, and monitors depressive disorder.Kroenke et al. 22  The independent variables were sociodemographic characteristics, work aspects, and health status, and we evaluated their association with the outcome -social support at work.Chart 1 presents the variables studied and their categories.

REV BRAS EPIDEMIOL 2019; 22: E190068
We analyzed the data with the software Statistical Package for the Social Sciences, based on descriptive statistics, using measures of central tendency and dispersion for continuous variables, and frequency distributions for categorical variables.We verified the presence of statistically significant differences among the study groups using Student's t-test and the χ 2 test and selected the variables associated with the outcome with a p-value ≤ 0.20 for the logistic regression.After the multivariate analysis, variables with a p-value ≤ 0.05 were considered statistically relevant to explain the social support at work. The

RESULTS
Among all TAEs who participated in the base-study, 328 comprised the cohort of this study, as shown in Figure 1.
When comparing the participant and non-participant population, the statistical analysis showed that the two groups did not differ with respect to most sociodemographic, health, and work variables investigated in the present study, except for age, ethnicity, and schooling.
The sociodemographic profile of TAEs, presented in Table 1, indicates that they are predominantly male, with a mean age of 47 years (ranging from 25 to 67 years), a higher incidence in the age group 41 to 59 years, white, married, with children, and have higher education or beyond.
Regarding the work-and health-related characteristics of TAEs, we found that most of them had only one job, with fixed working hours, weekly workload that did not exceed 40 hours, did not work at night, had up to 15 years working at UFJF, had another job before starting to work at the university, and were older than 18 years at their first job.Most TAEs did not show any signs and symptoms of depression, were classified as active or very active concerning their level of physical activity, had passive work and high social support at work.In the follow-up period (2012/2013 to 2016/2017), some variables have changed, as many individuals got married, had children, and improved their level of schooling.Concerning work-related aspects, more civil servants started working fixed hours and up to 40 hours per week, and the percentage of workers with high social support at work increased.On the other hand, the percentage of individuals who did not work at night decreased, and two civil servants no longer have just one job.
As to health status, one TAE no longer had signs and symptoms of depression, and the number of workers with good WAI increased.In contrast, relating to the level of physical activity, the number of individuals classified as active or very active dropped.When performing bivariate and regression analyses of independent variables with the outcome -social support at work -, we identified that, in the bivariate analysis, the variables type of work schedule, night shift, signs and symptoms of depression, level of physical activity (active/very active versus sedentary), demand control, and WAI had some influence on social support at work, with a p-value ≤ 0.20, as shown in Table 2.
.Despite the lack of significant differences between participants and non-participants, we cannot refrain from considering the size of the study population as a limitation.Also, extrapolating data to the general population requires caution, given that TAEs from the university have health and working conditions that might not reflect the reality of the general Brazilian population.Nonetheless, our results can contribute to designing intervention strategies that could to promote social support, prevent occupational stress and illnesses in workers with characteristics similar to those of the sample of this study.

CONCLUSION
We concluded that it is crucial to implement actions aimed at controlling risk factors associated with low social support at work and measures that promote a balanced work environment focused on existing social relationships, both among co-workers and between employees and supervisors.These actions can help to prevent diseases among workers, which often leads to a leave of absence, a situation that can happen early in some cases.
Thus, it is essential to have an ongoing multidisciplinary support program for civil servants from the university who presented low social support at work and health deficiencies, paying special attention to psychological support and the maintenance of functional capacity.For this multidisciplinary support program for civil servants, we suggest that the university includes its student body under the supervision of faculty members, integrating the TAEs in the process so that they can maintain an active and healthy life inside and outside the work environment.
24ovided evidence of validity, and Pfizer (Copyright © 2005 Pfizer Inc., New York, NY, USA) published the translation into Portuguese.In Brazil, Osório et al.23validated the instrument.It comprises 9 items, distributed in a 4 point-scale: 0 (not at all) to 3 (nearly every day), with a score ranging from 0 to 27 to assess the frequency of signs and symptoms of depression in the prior two weeks.A score greater than or equal to 10 corresponds to a positive indicator of major depression; • International Physical Activity Questionnaire (IPAQ), which covers data related to the practice of physical activity by the individual.Validated in 12 countries, IPAQ is a questionnaire that allows the estimation of the weekly time spent on mild, moderate, and vigorous physical activities.In Brazil, Pardini et al.24validated the instrument; • Work Ability Index (WAI), based on the worker's self-perception.Researchers from universities in the state of São Paulo 25 translated and adapted WAI for Brazil, and later, a study carried out with workers from a power company in the state of São Paulo 26 validated the instrument.
Research Ethics Committees of the Escola Nacional de Saúde Pública (REC/ENSP) and UFJF approved this study under the reports no.1,574,457 and 1,673,735, respectively.1. Study variables and their categories, Juiz de Fora, 2017.

Table 1 .
Profile of the technical-administrative staff in education throughout the cohort follow-up.Juiz de Fora, 2017.

Table 2 .
Sociodemographic characteristics, health status, work conditions, and their influence on social support at work, Juiz de Fora, 2017.

Table 2 .
Continuationsocial support should be the basis of work relationships and a social organization strategy in institutions, as it can decrease or even prevent occupational stress.Valuing work relationships and environment can benefit the workers' health and ability