Psychosocial and environmental factors : influence on work ability among social assistance workers

| Background: Poor working conditions might impair the health and work ability (WA) of public service workers. Objective: To investigate the association of work ability with environmental and psychosocial factors among workers at Social Assistance Reference Centers in Uberaba, Minas Gerais, Brazil. Methods: We administered three questionnaires: one to collect sociodemographic data, one for workplace evaluation and the Work Ability Index (WAI). The data were subjected to descriptive analysis and the Mann-Whitney U test. Results: The sample comprised 41 participants, with average age 40 years old (SD=14.83) most of whom had attended higher education (82.9%). The participants reported poor workplace ventilation (75.6%), temperature (78%) and equipment (75.6%). Most participants reported to spend long time sitting (65.9%) and absence of breaks (51.2%). Threats to personal safety (31.7%), personal belongings (53.7%) and aggression by service users (26.8%), escorts or relatives (19.5%) were the psychosocial factors most frequently reported. The average score on WAI was 37.85; 41.5% of participants were rated as with poor or moderate WA. WA was associated with workplace temperature (p=0.015), available material resources (p=0.011), threats to personal safety (p=0.005) and violence between coworkers (p=0.016) or between service users and workers and/or supervisors (p=0.035). Conclusion: The results corroborate the existence of a relationship between working conditions and WA and evidenced association between WA and workplace violence.


INTRODUCTION
Within the organization of society, work is the main activity for adults, and thus affords a path for identity affirmation and social inclusion.This central role of work is associated with individual values, skills and affinities, and is the means to enact citizenship and participate in society 1 .In Brazil, the public sector employs a substantial number of workers at three levels: federal, state and municipal.According to the Personnel Statistical Data Dashboard, there are currently 1,276,275 civil servants in the country, having had increased by 10% in the past 10 years 2 .Civil servants are a heterogenous group of workers, with different characteristics, educational level, salary, working conditions and job demands 3 .
The working conditions in public service might contribute to promote health in a favorable work environment, or be adverse 4 .Job demands are partly studied in terms of physical determinants, such as inadequate posture, absence of breaks and remaining too long in the same position 5 .The working conditions also include psychosocial factors, including control over tasks, autonomy and decision-making power, psychological demands, social support, organizational climate and workplace violence [6][7][8][9][10] .Complex in nature, psychosocial factors are defined as those which result from the interaction between organizational aspects of work and the social and environmental context 6 .As part of the context of work, physical and psychosocial demands interfere with the health of workers and have considerable influence on productivity, efficiency and work ability 5,8,10 .
The notion of work ability (WA) was developed in Finland based on a holistic view of health.WA is assessed on the basis of the self-perceived ability of workers to perform their job as a function of their available physical, mental and social resources 11 .WA is influenced by several factors, including aging, lifestyle, sociodemographic aspects, work organization, environmental, physical and mental demands, and work overload 12,13 .Therefore, WA results from the balance between work-related factors, individual resources and the macrosocial environment 14,15 .
Several environmental and psychosocial factors are related to and interfere with WA.Analysis of WA among several occupational groups detected association with high physical demands causing musculoskeletal pain and mild emotional disorders, as well as some quality of life aspects [16][17][18][19] .Also sociodemographic and occupational variables were found to have correlation with WA in different populations of workers 4,19 .On the whole, these studies evidenced factors related to the maintenance of work ability as a function of the working conditions and job demands.
Among social workers in the public sector, those at Social Assistance Reference Centers (Centros de Referência de Assistência Social -CRAS) are responsible for implementing the Social Assistance National Policy (Política Nacional de Assistência Social -PNAS), which provides basic protection and organizes the social networks for individuals under conditions of high social risk and vulnerability within definite geographical areas 20 .CRAS employees must thus deal with complex situations, however, under precarious circumstances which expose them to exhaustion and have negative impact on WA.A qualitative study performed in 2015 described the working conditions at CRAS as precarious, hindering the implementation of PNAS, with poor physical facilities, and shortage of material and human resources 21 .
We were not able to locate any study that analyzed the relationship between WA and work environment and psychosocial characteristics of work among social assistance workers.Therefore, the aim of the present study was to analyze WA and its association with environmental characteristics and psychosocial factors among CRAS employees in Uberaba, Minas Gerais, Brazil.The hypothesis underlying the present study was that inadequate workplace characteristics and psychosocial hazards have negative influence on the WA of CRAS employees.

STUDY DESIGN AND PARTICIPANTS
The present cross-sectional, observational and quantitative study was approved by the research ethics committee of Federal University of Minas Triangle Psychosocial factors and work ability (CAAE 52647216.7.0000.5154,ruling no.1,774,870).Participants were workers at the eight CRAS in Uberaba.According to the Basic Operational Standard of the Social Assistance Unified System (Sistema Único de Assistência Social-SUAS), Uberaba is considered a large municipality, with potential to receive up to 5,000 families in each CRAS 20 .CRAS facilities comprise a reception, one or more interview rooms, a meeting room, and space for therapeutic activities.The minimum staff consists of a coordinator, a social worker, a psychologist, and an administrative assistant; additional members are experts in other fields 20 .
The data on CRAS employees were obtained from the municipal Secretariat of Social Development.Potential participants were recruited by convenience sampling, and those who met the following eligibility criteria were invited to participate in the study: age over 18 years old, either sex, and having been hired to any job at CRAS more than 6 months earlier.Employees who had worked less than 6 months at one and the same CRASS, on leave, or receiving pension were excluded.

INSTRUMENTS AND PROCEDURES
The participants received information on the study aims and methods and signed an informed consent form.Data collection was performed from November 2016 through January 2017 during visits to all eight CRAS, which involved three phases: collection of sociodemographic data, workplace assessment, and WA assessment.All the participants responded the questionnaires at the workplace on one single day, according to their availability.
In phase 1, the participants responded a questionnaire for sociodemographic -age, sex, educational level, number of children, monthly income, habits (e.g.smoking) and physical activity -and occupational data: length of work in the current job, weekly and monthly working hours and shift.
In phase 2, the participants assessed their work environment by means of an adaptation of Barbosa et al.'s 22 Workplace Evaluation Protocol.This instrument comprises 25 questions distributed across domains physical demands of work, organizational climate and physical environment; the two former are responded on a 4-point Likert scale and the latter on a 3-point Likert scale.
In phase 3, the participants responded the Work Ability Index (WAI), developed at the Finnish Institute of Occupational Health, Helsinki, and translated and adapted for use in Brazil 23 .WAI is a self-report scale for WA assessment, and comprises 10 questions to measure: current WA, WA in relation to job demands, current diagnosed diseases, sick leaves, own prognosis of future WA, and mental resources 11 .The global score ranges from 7 to 49 and is categorized as follows: 7-27, low WA; 28-36 moderate WA; 37-42 good WA; and 44-49 excellent WA 23 .

DATA ANALYSIS
Data analysis was performed with software Statistical Package for the Social Sciences (SPSS), IBM®, version 20.0.Descriptive analysis included calculation of absolute and relative frequencies for categorical variables and mean and standard deviation (SD) for continuous variables.The responses on the Workplace Evaluation Protocol were dichotomized as adequate/inadequate.Association between WA and psychosocial and environmental factors was investigated by means of the Mann-Whitney U test.The significance level was set to 5%.

RESULTS
The sample comprised 41 participants, mostly female (65.9%), with average age 40.72 years old (SD=14.836).About 58.5% of the participants did not have a partner (single, divorced or widowed) and 51.2% had children.About 82.9% of the participants had attended more than 8 years of formal education; 39% had completed higher education.About 45% of the participants had a sedentary lifestyle and 7% were smokers.The average length of work in the current institution was 48.7 months (SD=59.55),i.e., 4 years (SD=4.96).Average working hours were 6.90 hours/day (DP=1.09)and 33.61 hours/week (SD=9.07).Most participants had high-level jobs (56.2%) and included psychologists (22%), social workers (22%) and managers (12.2%).About 63.2% of the participants worked one single shift (morning or afternoon); 26.8% worked the night shift (sometimes or always) and 19.5% reported to have a second job.Table 1 describes the sample sociodemographic, occupational and lifestyle characteristics.
In regard to the job physical demands, 65.9% of the sample reported to work sitting most of the time, 56.1% inadequate body posture and 34.1% to remain standing over a long period of time.A little more than half of the sample reported to work without any break (51.2%),34.1% that they need to walk a considerable part of the time, and 29.3% to often carry heavy loads.
Analysis of psychosocial factors at work revealed that there was no resting room available for 97.6% of the sample or lockers for 58.5%.About 20% of the participants reported to have witnessed aggressions by service user escorts, 7.3% by service users and 9.8% between supervisors or coworkers.Almost half of the sample (43.9%) reported threats to their personal belongings, 31.7% to their personal safety, and 9.8% had already thought of seeking a new job due to workplace violence.WA exhibited statistically significant association with workplace temperature (p=0.015); the participants who reported working in environments with inadequate temperature exhibited poorer WA.We also found significant association between poorer WA and insufficient material resources at work (p=0.011).None of the other analyzed environmental characteristics (ventilation, lighting, furniture and equipment) exhibited association with WA (Table 3).
We did not find significant association between WA and job physical demands, which indicates that the latter did not influence the former in the analyzed population (Table 4).WA reported having witnessed episodes of violence involving employees and service users (p=0.035) and between workers (supervisors and coworkers, p=0.016), as well as threats to their personal safety in the workplace (p=0.005).We did not find significant association between WA and the other analyzed psychosocial aspects.

DISCUSSION
Most participants were female, which corroborates the predominance of women in social assistance and points to increase in the proportion of women in public service jobs, as well in those requiring a high educational level 24 .Indeed, a large proportion of the participants were social workers or psychologists, which occupations are historically characterized by a predominance of women 24 .This finding might further account for the high educational level of the analyzed sample, since social workers and psychologists have the central role in PNAS and are part of the minimum staff needed to implement CRAS programs and services 20 .As a fact, the demand for better qualified workers is increasing, as evidenced by the profile of workers hired  As per the reported habits, the participants can be considered as having a healthy lifestyle.Few participants reported to smoke and most to perform physical activity on a regular basis, which might be seen as strategies to cope with high psychological job demands.The results of the National Household Sample Survey (Pesquisa Nacional por Amostra de Domicílios-PNAD) evidenced positive correlation between physical activity and educational level, i.e. the proportion of individuals who performed physical activity was higher among the adults with better educational level 26 .Physical activity is considered a means to adapt to the environment and contributes to improve or maintain the general state of health of people.In addition, it increases motivation, job satisfaction and productivity 27 .According to Cordeiro and Araújo 7 , the lifestyle of workers is a marker of WA: leisure activities, physical activity and avoiding alcohol and smoking are predictors of adequate WA.
Also the participants' daily and weekly working hours agree with those found in the 2017 SUAS Census, according to which most CRAS employees nationwide work one single shift, with an average of 6 working hours/day 25 .In the present study, the participants who worked more than 6 hours/day were managers, social agents or general service workers, whose jobs are associated with distinct demands, which might account for also distinct working hours.
According to recent studies, WA is impaired in a considerable proportion of workers, public service employees in different sectors, as e.g.health care 4,5,16,17 , the judiciary 19 and education 18 .More than 40% of the participants in the present study exhibited poor or moderate WA, which represents a considerable proportion.In Ilmarinen's view 15 , satisfactory WA prolongs the workers' permanence in the labor market, while impairment is associated with negative outcomes, such as leaves, accidents and physical and mental diseases, which do not only interfere with the workers' jobs, but also with their family, economic and social environment.
The results of the present study confirm the existence of a relationship between working conditions and WA.They also point to an association between WA and psychosocial factors related to social assistance work, particularly insecurity and workplace violence.
A high proportion of participants reported inadequate ventilation, temperature, equipment, furniture and material resources in the workplace.Poor lighting might cause headache, vision problems and tension 28 .Environmental ventilation and temperature have effects on the body, influence the thermal perception, may cause sleepiness, and reduce the blood pressure and attention 27,28 .Among the analyzed environmental aspects, inadequate temperature and lack of material resources were associated with poorer WA.Other authors who studied the relationship between WA and working conditions suggest that poor working conditions impair the workers' performance 14 , cause distress 3 , and have negative impact on their general 4 , physical 5,22 and mental 5 health.
The negative perception of the environmental temperature reported by the participants in the present study might be related to the physical aspects of CRAS.Of the eight analyzed sites, only two had been specifically built for this purpose, while the other six operated in rented facilities.Therefore, there are reasons to suspect the available space was not the ideal for the intended services, with consequent impairment of the workers' performance.
Also shortage of material resources was significantly associated with poorer WA.Reductions in the allocation of financial resources to social programs wreck institutions, which thus become compelled to operate with insufficient equipment and material and human resources, with consequent impairment of service delivery 21 .
In the present study, we did not find a direct relationship between physical job demands and WA, which disagrees with reports in the literature 5,12,22 .High physical job demands are associated with muscle exhaustion, inadequate body posture, higher odds of chronic musculoskeletal pain and other health problems 22 .As a rule, activities at CRAS pose low physical, but high mental demands, including continuous decision making and problem solving 21 .This fact might account for the lack of correlation between WA and physical demands in the present study.This hypothesis is supported by the results relative to the environmental conditions, and mainly to the psychosocial factors at work.

Psychosocial factors and work ability
Violence, as a psychosocial factor of work, has grown in recent years.It is a significant cause of concern and a source of risk to the health of workers.The association we found between WA, lack of safety and aggression in the workplace confirms the existence of a relationship between organizational climate and performance at work.Aggression and lack of safety are forms of violence to which workers are exposed.Workplace violence is defined as violent acts, including physical or verbal aggression or threats, against workers 29 .Among the outstanding factors found in the study by Fischer and Martinez 9 , verbal abuse by patients was a leading cause of fear and tension among workers, with negative impact on their well-being and WA.
Another study confirmed that workplace violence impairs the physical and mental health of healthcare workers, reduces their self-esteem, and might cause depression, apathy, irritability and memory disorders 30 . in this regard, attention should be paid to the overall context of CRAS employees' working conditions: facilities located in peripheral areas, direct care delivery to a high-risk and vulnerable population, doors open to the community, exposure to psychosocial hazards and violence, and lack of the required resources, among others.
Within such a context, besides their impact on CRAS employees' physical health and productivity, violence and unsafety also influence their performance and service delivery.Lack of safety and violence in the workplace are psychosocial factors which cause physical and mental tension and increase the level of violence between coworkers 8 .Gadegaard et al. 10 analyzed the relationship between the outcomes of workplace violence prevention policies and programs in high-risk areas, and found that support from coworkers and supervisors after a violent episode helped reducing their negative emotional impacts.Support might be received through prevention programs which provide the staff training on how to identify and manage hostile behaviors involving clients and workers to thus improve the response to violent incidents 29 .
While the cross-sectional design is useful to describe population characteristics, the resulting representation hinders causal analysis, since it consists of data collected at one single time-point.For this reason, we point to the need for longitudinal studies able to identify the factors which influence WA over time.One further limitation of the present study derives from the small sample size.However, the participants represented more than 60% of the total number of CRAS employees, and their sociodemographic and occupational characteristics are similar to that of CRAS workers in other Brazilian regions and towns 25 .

CONCLUSION
The present study represents an advance inasmuch as it analyzed social assistance workers, their working conditions and WA.Violence and unsafety were found to be potential predictors of WA.Despite its ubiquitous presence, especially in public facilities which provide services to high-risk populations, only in recent years workplace violence became a cause of concern and was categorized as an occupational health risk.The literature on workplace violence and workers' perception of their work environment is still incipient.Analysis of the work environment, its psychosocial characteristics and their relationship with WA evidenced significant results, and pointed out aggression and unsafety as risk factors.Our results indicate that several variables should be taken into account in the design and implementation of actions and policies to maintain or improve the WA of social assistance workers.In particular, we call the attention to the need for preventive programs against violence and to improve workplace safety as means to enhance the performance of workers and improve the quality of service delivery.

Table 5
describes the association between WA and psychosocial factors.The participants with the poorest

Table 1 .
Sociodemographic, lifestyle and occupational characteristics of workers at Social Assistance Reference Centers in Uberaba, Brazil, 2019 (n=41).

Table 2 .
Working conditions, job demands and psychosocial factors at work relative to workers at Social Assistance Reference Centers in Uberaba, Brazil, 2019 (n=41).

Table 4 .
2ssociation between work ability and job physical demands, Uberaba, Brazil 2019 (n=41).forhigherlevelpositionsthrough public calls and the overall elevation of the educational level of workers in the public sector2.Most workers allocated to CRAS were young adults, namely, the economically active population with the highest WA.This finding agrees with data gathered in the 2017 SUAS Census of CRAS employees all across Brazil 25 .