SciELO - Scientific Electronic Library Online

 
vol.19 issue2Evaluation of ophthalmologic assistance from the perspective of usersDesign of a digital and self-reported food frequency questionnaire to estimate food consumption in adolescents and young adults: birth cohorts at Pelotas, Rio Grande do Sul, Brazil author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

Share


Revista Brasileira de Epidemiologia

Print version ISSN 1415-790XOn-line version ISSN 1980-5497

Rev. bras. epidemiol. vol.19 no.2 São Paulo Apr./June 2016

http://dx.doi.org/10.1590/1980-5497201600020016 

ARTIGO ORIGINAL

Testing the "Work Ability House" Model in hospital workers

Maria Carmen Martinez1 

Maria do Rosário Dias de Oliveira Latorre2 

Frida Marina Fischer1 

1Department of Environmental Health, School of Public Health, Universidade de São Paulo - São Paulo (SP), Brazil

2Department of Epidemiology, School of Public Health, Universidade de São Paulo - São Paulo (SP), Brazil

ABSTRACT:

Objective:

To test the Work Ability House model, verifying the hierarchy of proposed dimensions, among a group of hospital workers.

Methods:

A cohort study (2009-2011) was conducted with a sample of 599 workers from a hospital in the city of São Paulo. A questionnaire including sociodemographics, lifestyle and working conditions was used. The Brazilian versions of Job Stress Scale, Effort-Reward Imbalance, Work-Related Activities That May Contribute To Job-Related Pain and/or Injury, and the Work Ability Index (WAI) were also used. A hierarchical logistic regression analysis was performed: the independent variables were allocated into levels according to the dimensions of the theoretical model in order to evaluate the factors associated with work ability.

Results:

Variables associated with impairment of work ability in each dimension were as follows: (a) sociodemographics: age < 30 years (p = 0.20), (b) health: without report of occurrence of work injuries (p = 0.029), (c) professional competence: low educational level (p = 0.008), (d) values : intensified in overcommitment (p < 0.001), and (e) work: intensification of effort-reward imbalance (p = 0.009) and high demands (p = 0.040).

Conclusion:

The results confirmed the dimensions proposed for the Work Ability House model, indicating that it is valid as a representation of a multidimensional construct of multifactorial determination and can be used in the management of work ability.

Keywords: Work capacity evaluation; Occupational health; Workers; Workload; Work environment; Health personnel

INTRODUCTION

The concept of work ability (WA) concerns the ability of the worker to perform his/her tasks at work. It is conditioned by the work demands, health status, and physical and mental abilities1,2,3. WA is considered to be a measurement of functional aging1,2,3, and it is seen as an index for the health of the worker2,4. This concept has been expanding based on the centrality of health for models that integrate aspects related to health, well-being, and macrosocial environment2,4,5.

Theoretical models have been proposed to explain the process to determine WA and/or the dimensions associated with this construct, such as the one based on the Stress-Strain Model or the Tetraedric Model4.

Among these models, the multidimensional Work Ability House stands out. It considers that the WA depends on the balance between individual resources, work-related factors, and the macrosocial environment2,4,5. The model is expressed by a four-floor House and a roof inserted in an encircling environment. The individual resources comprise the dimensions represented in the three lower floors. The first floor represents the base that supports the building and concerns health status and functional capacity, including physical, mental, and social aspects. The second floor relates to professional competence (knowledge and skills, training and learning at work) and its continuous development used to meet the demands of working life. The third floor represents the internal aspects of the individual, manifested as values, attitudes, and motivation. These aspects can be affected by the external environment, that is, by the relationship between work, society, and personal life. The last floor represents factors related to work such as work conditions, demands and content, organization and communitarian environment, management and supervision: this is the "heaviest" floor in the building and can affect the other dimensions, which also support it. The WA is also influenced by the surroundings of the macrosocial environment, which includes matters related to public and social policies, health care and occupational safety, and in special, the structure and support of family and community. The roof of this building is the WA, resulting from the interaction and balance between the previous dimensions2,4,5.

This model has been assessed by a number of international studies6,7. In Brazil, analyses about WA have been conducted since the 1990s, using the Work Ability Index (WAI) as the research instrument3; however, there are only a few national studies testing the theoretical model under discussion. Considering these matters, this study aimed at testing the theoretical Work Ability House model, verifying the hierarchy of the proposed dimensions for a group of workers in the hospital sector of the city of São Paulo.

METHODS

This is a two-year follow-up longitudinal study (2009 to 2011) carried out in a private high complexity hospital in the city of São Paulo, Brazil. In 2009, all active workers were invited to participate in the study. This occupational group was chosen because hospital work is characterized by relevant physical and mental demands, which are associated with negative outcomes for the worker, such as WA impairment8,9,10.

The adherence rate was 87.9% (1,226 people). Among them, 599 workers (48.5%) participated in 2011, and most of them were from the Nursing (51.8%) and Hospitality sectors - hygiene, gastronomy, and patients' admission (18.5%). The main causes of losses were dismissals (54.7%) and not answering the questionnaire (39.7%).

Participants differed from nonparticipants in terms of the following factors: gender (57.2% of women versus 40.9% of men, p < 0.001), work sector (greater losses in the administrative sectors of Planning and Commercial, respectively, with 87.1 and 81.6% of losses, respectively, p < 0.001), age (participants mean age 35,7 years, SD = 8.3 years versus losses 34.6 years, SD = 8.9 years, p = 0.022), and working time (years) in the studied hospital (participants 6.1 years, SD = 6.5 years versus losses with 4.8 years, SD = 5.8 years, p < 0.001).

Data collection was performed by a self-report comprehensive questionnaire. The first part included items on sociodemographics, lifestyle, and functional aspects. The second part was the short version of the Job Stress Scale (JSS)11, based on the demand-control model. JSS assesses the strain resulting from stressors (demand, control and social support) of the psychosocial work environment11. The third part included the Effort-Reward Imbalance (ERI) questionnaire12, whose variables (effort, reward, and overcommitment) also evaluate other psychosocial work stressors. The fourth part was the questionnaire of Work-Related Activities that May Contribute to Job-Related Pain and/or Injury (WRAPI)13. The last part was the Work Ability Index (WAI)1,3, used to measure the variable of interest in this study - WA. All questionnaires were validated to Brazilian Portuguese and are being used in current use.1,3,11,12,13.

The results of Cronbach's alpha to assess the reliability of questionnaires in the beginning of the follow-up were: WAI = 0.69; demand = 0.69; control = 0.57; social support = 0.82; effort = 0.74; reward = 0.83; overcommitment = 0.75; and WRAPI = 0.92. Considering the complexity of the phenomena to be assessed and their importance to understand the analyzed construct8, we chose to maintain the dimensions that presented alpha < 0.70.

The study variables were selected and placed in groups according to the dimensions (or floors) of the Work Ability House model:

  • • sociodemographics features: sex, age, marital status, family income and responsibility for underage children;

  • • health and functional capacity: alcohol consumption, smoking, nutritional status (based on body mass index), practice of regular physical activity, and recent work injury;

  • • professional competence: age at the time of joining the workforce, working time at the studied hospital, years in the profession, and position;

  • • values: overcommitment (6 to 24 points);

  • • work-related characteristics - work sector, work shift, working hours (adding those in the hospital, a second job, and domestic chores), work violence (7 to 21 points), demands at work (5 to 20 points), control at work (6 to 24 points), social support at work (6 to 24 points), ERI (0.17 to 5.00 points), and WRAPI (0 to 150 points);

  • • work ability - WAI, with a score of 7 to 49 points.

The independent variables were measured in the beginning of follow-up (2009). The exception included the variables regarding work stressors and WAI, assessed in the beginning and in the end of the follow-up. For each one of these variables, the difference between the initial and final scores was calculated, and a new variable was provided, categorized into "no changes," "aggravation," or "improvement". These variables were then dichotomized for the logistic modeling. Cutoff points were analyzed according to the distribution of frequencies, as to the best of our knowledge we did not find references in the literature.

A descriptive analysis was conducted by means, medians, standard deviations, and minimum and maximum values for the quantitative variables and proportions for categorical variables. The associations between independent variables and WA were assessed by the χ2-test. The theoretical Work Ability House model was tested by a hierarchical multiple logistic regression with predicted levels of hierarchy. In each level, modeling was conducted step by step. Gender was maintained as a control variable. The risk measurement was the odds ratio (OR), and in all analyses, the associations were considered to be significant when p < 0.05.

The research project was approved by the Research Ethics Committee of the School of Public Health, Universidade de São Paulo, protocol n. 257.518. The project was in agreement with the principles of the Declaration of Helsinki, established by the World Medical Association (WMA). The participation in in this study was voluntary. Workers signed an informed consent form. Individual results were kept confidential.

RESULTS

In 2009, the mean score of the WAI was 43.0 points (SD = 4.0); in 2011, it was 42.5 points (SD = 4.7). The change in WAI score since the beginning to the end of the follow-up was, in average, of -0.5 points (SD = 4.6), representing a slight impairment.

Table 1 shows the descriptive analysis of the variables representing the sociodemographic characteristics, and dimensions of health and professional skills. The highest proportions of participants were women (72.6%), married people/or living with a partner (50.1%), and monthly family income higher than 5 minimum wages (51.6%). The mean age was 36.7 years (SD = 8.3), and 73.1% of them were older than 30 years (73.1%). Regarding health, 91.3% of them reported sporadic alcohol consumption, 90.8% were non-smokers, 54.9% were eutrophic, 36.9% reported the regular practice of physical activities, and most denied recent occurrence of workplace injury (88.1%). For the variables representing professional competence, 94.2% had at least incomplete high school, and 69.8% were in the current profession in the past 6 years. The highest proportions were of Nursing Technicians (29.2%), Specialized Administrative staff (17.0%), Registered Nurses (16.9%), and General Assistants, all working in different sectors (15.0%).

Table 1: Descriptive statistics of sociodemographics, health and professional competence, according to changes in work ability, private hospital, São Paulo, 2009 - 2011. 

Variable Maintaining Aggravation Total p-value*
n % n % n %
Sociodemographics
Gender
Female 281 72.6 154 72.6 435 72.6 0.993
Male 106 27.4 58 27.4 164 27.4
Age group
< 30 90 23.3 70 33.0 160 26.7 0.010
≥ 30 296 76.5 142 67.0 438 73.1
Not informed 1 0.3 0 0.0 1 0.2
Marital status
Single 144 37.2 90 42.5 234 39.1 0.420
Married/partner 201 51.9 99 46.7 300 50.1
Separated/divorced/widow(er) 38 9.8 20 9.4 58 9.7
Not informed 4 1.0 3 1.4 7 1.2
Monthly Family income
≥ 5.1 minimum wages 209 54.0 100 47.2 309 51.6 0.074
< 5.0 minimum wages 164 42.4 107 50.5 271 45.2
Not informed 14 3.6 5 2.4 19 3.2
Responsibility for underage children/
No 188 48.6 108 50.9 296 49.4 0.656
Sporadically/yes 188 48.6 100 47.2 288 48.1
Not informed 11 2.8 4 1.9 15 2.5
Health and functional capacity
Alcohol consumption
Sporadic consumption (0 - 1 day/week) 355 91.7 192 90.6 547 91.3 0.736
Regular consumption (≥ 2 days/week) 23 5.9 14 6.6 37 6.2
Not informed 9 2.3 6 2.8 15 2.5
Smoking
Never smoked/former smoker 345 89.1 199 93.9 544 90.8 0.135
Yes, I smoke 37 9.6 13 6.1 50 8.3
Not informed 5 1.3 0 0.0 5 0.8
Nutritional status
Eutrophic 203 52.5 126 59.4 329 54.9 0.249
Overweight 132 34.1 60 28.3 192 32.1
Obesity 46 11.9 23 10.8 69 11.5
Not informed 6 1.6 3 1.4 9 1.5
Practice of physical activities
Yes 143 37.0 78 36.8 221 36.9 0.994
No 235 60.7 128 60.4 363 60.6
Not informed 9 2.3 6 2.8 15 2.5
Work injury
No 335 86.6 193 91.0 528 88.1 0.032
Yes 35 9.0 9 4.2 44 7.3
Not informed 17 4.4 10 4.7 27 4.5
Professional competence
Educational Level
Elementary school 8 2.1 12 5.7 20 3.3 0.023
incomplete/finished high school 366 94.6 198 93.4 564 94.2
Not informed 13 3.4 2 0.9 15 2.5
Years in the profession
Less than 6 96 24.8 72 34.0 168 28.0 0.011
6 and more 285 73.6 133 62.7 418 69.8
Not informed 6 1.6 7 3.3 13 2.2
Position
Others 367 94.8 190 89.6 557 93.0 0.017
Technicians 20 5.2 22 10.4 42 7.0
Not informed 0 0.0 0 0.0 0 0.0
Total 387 100.0 212 100.0 599 100.0

2 test. Obs.: Results for the first year of data collection (2009).

Table 2 presents the variables representing the dimensions of values and work. During follow-up, 31.4% of the workers reported aggravation in overcommitment. In the beginning of the follow-up (2009), the mean of overcommitment was 12.3 points (SD = 3.1), in a score ranging from 6.0 to 24.0 points. Table 2 shows that participants were working mainly in the Nursing Service (51.8%) and Hospitality Sectors (18.5%). The distribution regarding work shift was relatively homogeneous; 64.8% of the workers did not consider being exposed to circumstances of workplace violence, and 41.7% denied changes in the weekly work load throughout the studied period.

In 2009, the mean score of work demands was 14.1 points (SD = 2.3), and 32.7% reported aggravation during follow-up. The mean score of work control was of 17.8 points (SD = 2.4), and 22.5% reported aggravation. The mean score of social support was 20.7 points (SD = 2.8), and 35.2% reported aggravation. The mean ERI score was 0.42 points (SD = 0.18), and 36.2% reported aggravation. The mean WRAPI was 57.5 points (SD = 34.6), and 34.2% reported aggravation.

Table 2: Descriptive statistics of personal values and work features , according to changes in work ability, private hospital, São Paulo, 2009 - 2011. 

Variable Maintaining Aggravation Total p-value*
n % n % n %
Values
Overcommitment
No changes 161 41.6 74 34.9 235 39.2 < 0,001
Aggravation 98 25.3 90 42.5 188 31.4
Improvement 120 31.0 43 20.3 163 27.2
Not informed 8 2.1 5 2.4 13 2.2
Work
Work violence (points) 0.602
7 (no violence) 254 65.6 134 63.2 388 64.8
8 67 17.3 29 13.7 96 16.0
9 34 8.8 23 10.8 57 9.5
10 or more 14 3.6 9 4.2 23 3.8
Not informed 18 4.7 17 8.0 35 5.8
Work sector
Corporate areas 34 8.8 12 5.7 46 7.7 0.320
Hotel 62 16.0 49 23.1 111 18.5
Other operations/services 13 3.4 6 2.8 19 3.2
Nursing services 207 53.5 103 48.6 310 51.8
Medical superintendence 32 8.3 16 7.5 48 8.0
Services of diagnosis and therapy 21 5.4 13 6.1 34 5.7
Supplies 18 4.7 13 6.1 31 5.2
Not informed 0 0.0 0 0.0 0 0.0
Work shift
Administration 102 26.4 43 20.3 145 24.2 0.246
Morning 101 26.1 55 25.9 156 26.0
Afternoon 75 19.4 53 25.0 128 21.4
Night 106 27.4 56 26.4 162 27.0
Not informed 3 0.8 5 2.4 8 1.3
Work load
No changes 153 39.5 97 45.8 250 41.7 0.531
Aggravation 76 19.6 40 18.9 116 19.4
Improvement 89 23.0 45 21.2 134 22.4
Not informed 69 17.8 30 14.2 99 16.5
Work demands
No changes 180 46.5 110 51.9 290 48.4 0.010
Aggravation 123 31.8 73 34.4 196 32.7
Improvement 81 20.9 23 10.8 104 17.4
Not informed 3 0.8 6 2.8 9 1.5
Effort-reward imbalance
No changes 75 19.4 41 19.3 116 19.4 < 0.001
Aggravation 117 30.2 100 47.2 217 36.2
Improvement 181 46.8 59 27.8 240 40.1
Not informed 14 3.6 12 5.7 26 4.3
Social support
No changes 184 47.5 91 42.9 275 45.9 0.002
Aggravation 119 30.7 92 43.4 211 35.2
Improvement 82 21.2 26 12.3 108 18.0
Not informed 2 0.5 3 1.4 5 0.8
Control over work
No changes 208 53.7 115 54.2 323 53.9 0.524
Aggravation 84 21.7 51 24.1 135 22.5
Improvement 88 22.7 40 18.9 128 21.4
Not informed 7 1.8 6 2.8 13 2.2
Situations that favor pain/injury
No aggravation (delta < 15,0) 254 65.6 113 53.3 367 61.3 0.004
Aggravation (delta ≥ 15,0) 117 30.2 88 41.5 205 34.2
Not informed 16 4.1 11 5.2 27 4.5
Total 387 100.0 212 100.0 599 100.0

2 test. Obs.: Figures for the first year (2009). For the variables categorized according to type of change, the results presented difference in the scores between 2009 and 2011.

Tables 1 and 2show the variables that were significantly associated with WA impairment in univariate analyses such as age group (p = 0.10), report of workplace injury (p = 0.032), all variables related to professional competence, overcommitment (p< 0.001), work demands (p = 0.010), ERI (p< 0.001), social support (0.002), and WRAPI (p = 0.004).

Table 3 presents the results of multiple hierarchical analyses. Among the demographic variables, the age group was kept in the modeling (OR = 0.64; p = 0.020). Regarding the dimension of health/functional capacity, previous occurrence of workplace injury was associated with WA impairment (OR = 0.43; p = 0.029). Regarding professional competence, educational level remained in the modeling (OR = 0.27; p = 0.008). Overcommitment, representing the values, remained associated with WA impairment (OR = 2.11; p < 0.001). Regarding work, the variables associated with changes in WA were ERI (OR = 1.72; p = 0.009) and work demands (OR = 1.77; p = 0.040). Some of the dimensions related with "values" and "work" showed higher chances of WA impairment, even after the adjustment by other variables. Gender was kept in the model as a control variable.

Table 3: Determinants of work ability identified by the hierarchized logistic regression, private hospital, São Paulo, 2009 - 2011. 

Dimension/variable Univariate Multiple
ORcrude p-value ORadjusted 95%CI of ORadj p-value
Sociodemographics
Gender
Female 1.00 0.993 1.00 0.63 - 1.35 0.672
Male 1.00 0.92
Age group
< 30 1.00 0.011 1.00 0.43 - 0.93 0.020
≥ 30 0.62 0.64
Family income
≥ 5.1 minimum wages 1.00 0.074 1.00 0.53 - 1.05 0.093
< 5.0 minimum wages 1.36 0.74
Health/functional capacity*
Previous occurrence of work injuries
No 1.00 0.032 1.00 0.20 - 0.92 0.029
Yes 0.45 0.43
Smoking
Never smoked/quit 1.00 0.138 1.00 0.30 - 1.18 0.141
Current smoking 1.64 0.60
Professional competence (knowledge/skills)**
Years in the profession
< 6 1.00 0.012 1.00 0.47 - 1.12 0.149
≥ 6 0.62 0.73
Job title
Others 1.00 0.019 1.00 0.95 - 3.46 0.073
Technicians 2.13 1.81
Educational level
Concluded elementary school 1.00 0.028 1.00 0.11 - 0.71 0.008
Incomplete high school and more 0.36 0.27
Values (attitude/motivation)***
Overcommitment
No aggravation 1.00 <0.001 1.00 1.46 - 3.06 < 0.001
Aggravation 2.21 2.11
Work****
Work shift
Others 1.00 0.088 1.00 0.82 - 1.95 0.284
Afternoon 1.42 1.27
Work sector
Others 1.00 0.033 1.00 0.99 - 2.47 0.055
Hotel 1.58 1.56
Effort-reward imbalance
Improvement 1.00 < 0.001 1.00 1.15 - 2.58 0.009
No improvement 2.25 1.72
Social support
No aggravation 1.00 0.002 1.00 0.88 - 1.93 0.187
Aggravation 1.76 1.30
Work demands
Improvement 1.00 0.003 1.00 1.03 - 3.04 0.040
No improvement 2.13 1.77
Situations that favor pain/injury
No aggravation 1.00 0.004 1.00 0.80 - 1.80 0.371
Aggravation 1.69 1.20

*Multiple analysis adjusted by the variables sex and age; **Multiple analysis adjusted by the variables sex, age, and history of work accident; ***Multiple analysis adjusted by the variables sex, age, history of work accident, and educational level; ****Multiple analysis adjusted by the variables sex, age, history of work accident, educational level, and overcommitment. Obs.: Figures for the first year (2009). For the variables categorized according to type of change, the results presented difference in the scores between 2009 and 2011.

DISCUSSION

The results of this study confirmed the Work Ability House as a multidimensional model, in which characteristics of the individual, work and encircling environment are associated to WA. In this study, the factors that have been associated with changes in WA were age group, previous occurrence of workplace injury, years in the current profession, educational level, overcommitment, ERI, and work demands. Analyses were adjusted by the variables of each evaluated dimension (demographics, health, professional competence, values and work) in the hierarchical modeling. Moreover, in agreement with the theoretical model, the work dimension was the one with higher risk for WA impairment, with the variable regarding values (overcommitment).

The first set of variables included those related to sociodemographics features. These variables do not compose the four floors from the central structure of the House, but they are part of the surrounding environment5. The theoretical model emphasizes that factors surrounding the House influence WA, even if less directly than the floors composing its core structure5. Only the age group remained associated with WA impairment. These results do not mean that the social surrounding are not important for WA. They only show that, in the present study, such factors were of minor relevance. Partly this is explained as it was included only demographics and family features without the inclusion of broader aspects of the macro environment. Older age (≥30 years) proved to be a protective factor for WA. Even though the effect of chronological aging in relation to functional aging is consistently demonstrated2,5, this effect is not always linear or present. It can be mediated by the level of knowledge, experience, skills, and job ties, which older workers tend to show more than younger ones9,14. Another aspect is the possibility of the healthy worker effect, as those who remain active are the ones with better health.

The core structure of the House has the individual resources, including health and functional capacity, professional competence, and values4,5. The second set included the dimension of health/functional capacity. This dimension composes the first floor of the House, the base that supports the building, because this is the resource that is more clearly related with WA4,5,6. In this dimension, the previous occurrence of workplace injury appeared as a protective factor against WA impairment. This result requires a careful interpretation. Workplace injuries generate temporary or permanent disabilities, so they can compromise the functional capacity of the workers15. Most work injuries involving health professionals are related to musculoskeletal disorders and hands needlestick/sharp objects injuries16,17. Musculoskeletal injuries may generate a prolonged or definitive disability, and in the latter there is the risk for transmission of infectious diseases, leading to emotional and behavioral changes16,17. The healthy worker effect may have excluded those who presented more severe lesions, returning to work those with better health conditions.

The third set included variables representing the second floor of the House, concerning professional competence. In this dimension, workers with a higher educational level presented lower WA impairment. A study conducted with Finnish workers showed that one out of three workers with lower schooling had WA compromise, while this relationship was lower than one for those with higher schooling18,19. These differences must be interpreted from the point of view of the socioeconomic conditions reflected by education, translated into economic, occupational, and social terms, along with health conditions and professional specialization19.

The variables representing the third floor of the House concerns the internal aspects of the individual, manifested in values. Values were demonstrated by overcommitment. Overcommitment is defined as an individual motivational pattern of excessive search for accomplishment and high performance at work, which can become more intense owing to the pressure in the work environment, thereby making these professionals more prone to exhaustion and stress20. In this study, individuals with higher overcommitment presented higher WA impairment than the other workers, regardless of the other variables. This association is identified in other studies9,21.

The last set included the variables representing work. Work, with individual characteristics and resources, composes the structure of the House4,5. It is considered to be the wider and heavier floor; so, it can affect the others. If the workloads are disproportional to the individual resources, WA will be impaired4,5. In the work dimension, the variable associated with WA impairment was the greater imbalance between efforts and rewards and the exposure to work demands. The social and organizational context of work represented by ERI is pointed out as a predictor of WA, even more than other evaluated stressors9,22. It even presents a predictive value for the early exit of the nursing profession8. The ERI model is structured based on the conception of social reciprocity, in which the imbalance between the efforts made and the rewards obtained can generate negative emotions, which is prone to neuroendocrine and autonomic activation. If these situations are maintained, they can trigger adverse effects on health20. Interventions in these aspects help to reduce the stress load, with favorable effects on health and WA10. The associations between the perceptions of aggravation in the exposure to psychosocial work demands and WA impairment reflect the fact that the more intense and frequent the work demands, the higher the risks to health and WA20,23 among healthcare workers8,10.

Studies with different methodologies confirm the theoretical considerations of the Work Ability House model6,7,23, which results agree with the ones found here. A study conducted in a population sample of Finnish workers showed the dimensions of work and health presented higher power of explanation for the WA results6. Values, competence, and community surroundings also were associated with WA6, thus confirming the complex structure of the model. A study evaluating Finnish teachers identified that the variables in the different dimensions of the House were associated with WA - use of medications, body mass index, percentage of body fat, aerobic capacity, muscle strength, stress, burnout, motivation, work organization, and work community24. In a systematic review, the authors emphasized the multifactorial nature of the construct, after identifying a variety of factors associated with WA impairment such as, lack of free time for physical activities, impaired musculoskeletal capacity, chronological aging, obesity, high mental and physical demands, lack of autonomy, and precarious physical work place23.

WA impairment has a predictive value for negative outcomes for workers, institutions and society, resulting in absenteeism, lack of productivity, illnesses, early exit of the profession and higher mortality, including the health sector2,10,18,23. The knowledge of WA determinants allows subsidizing institutional and public policies in order to promote health and well-being for the workers, to protect and recover WA and favor employability2,5,10,18,23. A valid theoretical model to understand WA determinants represents a useful resource in the management of the worker's health, applicable in planning, development, and evaluation of intervention actions addressed to the individual and collective aspects of work4,5,24.

The longitudinal design of this study allows establishing causality in the observed relations and confirming the tested theoretical model. However, some limitations must be mentioned. The first one is the rate of response (48.5%). In the period of the study (3 years), there was a significant turnover; so, the sample losses were mainly caused by dismissals (54.7%). High rates of turnover are commonly observed in the hospital sector, especially among nursing staff25. This occurs as hospital work is characterized by relevant physical and mental demands resulting from the work object (involving human health and life), the physical environment, the processes, and organization of work, which are usually unfavorable, conflicting interpersonal and work relationships and restricted forms of recognition8,25. Another limitation was the restricted number of variables analyzed in each dimension of the House, because of the structure of the cohort; so, some measurements could not be assessed such as objective aspects of functional capacity. Finally, the study was conducted in a specific work group. Despite the limitations, the external validity can be extended for institutions with similar work characteristics and organization.

CONCLUSIONS

The results of this study among hospital workers confirmed the dimensions proposed for the Work Ability House model. It showed it is a valid model representing WA as a multidimensional construct, which is determined by different causes. These results have implications for institutional and public policies, because the tested model represents a useful tool in planning, development, and evaluation of actions addressed to the promotion and recovery of WA. More studies approaching other occupational groups are welcome

REFERENCES

1. Tuomi K, Ilmarinen J, Jahkola A, Katajarinne L, Tulkki A. Índice de capacidade para o trabalho. São Carlos: EduFSCar; 2005. [ Links ]

2. Ilmarinen J. Maintaining work ability. In: Towards a longer worklife! Ageing and the quality of worklife in the European Union. Helsinki: Finnish Institute of Occupational Health; 2006. p. 132-48. [ Links ]

3. Martinez MC, Latorre MRDO, Fischer FM. Validity and reliability of the Brazilian version of the Work Ability Index questionnaire. Rev Saúde Pública 2009; 43: 55-61. [ Links ]

4. Ilmarinen J, Gould R, Jäevikoski A, Järvisalo J. Diversity of work ability. In: Gould R, Ilmarinen J, Järvisalo J, Koskinen S, editors. Dimensions of work ability: Results of the Health 2000. Helsinki: Finnish Centre of Pensions, The Social Insurance Institution, National Public, Health Institute, Finnish Institute of Occupational Health; 2008. p. 13-24. [ Links ]

5. Ilmarinen J. 30 years' work ability and 20 years' age management. In: Nygård CH, Savinainen M, Kirsi T, Lumme-Sandt K, editors. Age Management During the Life Course Proceedings of the 4th Symposium on Work Ability. Tampere: Tampere University Press; 2011. p. 12-22. [ Links ]

6. Ilmarinen J, Tuomi K, Seitsamo J. New dimensions of work ability. Int Congr Ser 2005; 1280: 3-7. [ Links ]

7. Järvelin S, Louhevaara V. Predictors of perceived work ability in mentally demanding work. In: Nordic Ergonomics Society's annual conference: ergonomics for the future; 2007 oct 1-3; Lysekil/Sweden: Nordic Ergonomics Society, 2007. [ Links ]

8. Fischer FM, Martinez MC. Individual features, working conditions and work injuries are associated with work ability among nursing professionals. Work 2013; 45: 509-17. [ Links ]

9. Fischer FM, Martinez MC. Work ability among hospital food service professionals: multiple associated variables require comprehensive intervention. Work2012; 41: 3746-52. [ Links ]

10. Hasselhorn H-M, Müller BH, Tackenberg P, NEXT-Study Group. NEXT Scientific Report - July 2005. Wuppertal: University of Wuppertal; 2005. [ Links ]

11. Alves MGM, Chor D, Faerstein E, Lopes CS, Wenerck GL. Versão resumida da "job stress scale": adaptação para o português. Rev Saúde Pública 2004; 38: 164-71. [ Links ]

12. Chor D, Werneck GL, Faerstein E, Alves MGM, Rotenberg L. The Brazilian version of the effort-reward imbalance questionnaire to assess job stress. Cad Saúde Pública2008; 24: 219-24. [ Links ]

13. Coluci MZO, Alexandre NMC. Adaptação cultural de instrumento que avalia atividades do trabalho e sua relação com sintomas osteomusculares. Acta Paul Enferm 2009; 22: 149-54. [ Links ]

14. Von Bonsdorff ME, Kokko K, Seitsamo J, von Bonsdorff MB, Nygård C-H, Ilmarinen J, Rantanen T. Work strain in midlife and 28-year work ability trajectories. Scand J Work Environ Health 2011; 37: 455-63. [ Links ]

15. Santana VS, Xavier C, Moura MCP, Oliveira R, Espírito-Santo JSE, Araújo G. Gravidade dos acidentes de trabalho atendidos em serviços de emergência. Rev Saúde Pública 2009; 43: 750-60. [ Links ]

16. Oliveira QB, Santos RS, Santos CMF. Acidentes de trabalho na equipe de enfermagem: uma revisão de literatura. Rev Enfermagem Contemp 2013; 2: 32-52. [ Links ]

17. Silva AID, Machado JMH, Santos EGOB, Marziale MHP. Acidentes com material biológico relacionados ao trabalho: análise de uma abordagem institucional. Rev Bras Saúde Ocup 2011; 36: 265-73. [ Links ]

18. Gould R, Ilmarinen J, Järvisalo J, Koskinen S. Dimensions of work ability - summary and conclusions. In: Gould R, Ilmarinen J, Järvisalo J, Koskinen S, editors. Dimensions of work ability: Results of the Health 2000. Helsinki: Finnish Centre of Pensions, Social Insurance Institution, National Public, Health Institute, Finnish Institute of Occupational Health; 2008. p. 165-75. [ Links ]

19. Martelin T, Sainio P, Koskinen S, Gould R. Education. In: Gould R, Ilmarinen J, Järvisalo J, Koskinen S, editors. Dimensions of work ability: Results of the Health 2000. Helsinki: Finnish Centre of Pensions, Social Insurance Institution, National Public, Health Institute, Finnish Institute of Occupational Health ; 2008. p. 42-4. [ Links ]

20. Siegriest J. Effort-reward imbalance and health in a globalized economy. Scand J Work Environ Health 2008; 0(Suppl 6): 163-8. [ Links ]

21. Conway PM, Campanini P, Sartoria S, Dotti R, Costa G. Main and interactive effects of shiftwork, age and work stress on health in an Italian sample of healthcare workers. Appl Ergon 2008; 39: 630-9. [ Links ]

22. Bethge M, Radoschewski FM, Gutenbrunner C. Effort-reward imbalance and work ability: cross-sectional and longitudinal findings from the Second German Sociomedical Panel of Employees. BMC Public Health 2012; 12: 875. [ Links ]

23. Van den Berg TIJ, Elders LAM, Zwart BCH, Burdorf A. The effects of work-related and individual factors on the Work Ability Index: a systematic review. Occup Envron Med 2009; 66: 211-20. [ Links ]

24. Maltby T. Extending working lives? Employability, work ability and better quality working lives. Social Policy & Society 2011; 10: 299-308. [ Links ]

25. Oliveira SAO, Paiva RFR. Possibilidade de diminuir o turnover da equipe de enfermagem nos serviços hospitalares. Rev Gestão & Saúde 2011; 2: 60-73. [ Links ]

Financial support: Hospital Samaritano de São Paulo supported the development of the study by using equipment and materials (computers, publications and meeting rooms), and providing human resources to help data collection and typing. The institution also provided resources for the participation in congresses on the study subject

Received: July 14, 2014; Accepted: May 05, 2015

Corresponding author: Maria Carmen Martinez. Department of Environmental Health, School of Public Health, Universidade de São Paulo. Avenida Doutor Arnaldo, 715, CEP: 01246-904, São Paulo, SP, Brasil. E-mail: mcmarti@uol.com.br

Conflict of interests: nothing to declare

Creative Commons License Este é um artigo publicado em acesso aberto sob uma licença Creative Commons