Health and motivation as mediators of the effects of job demands, job control, job support, and role conflicts at work and home on sickness presenteeism and absenteeism

To be included in the analyses the following criteria had to be met: (1) having been at the current workplace for at least 1 year, (2) between 18 and 65 years of age, and (3) have a complete set of data on presenteeism and sickness absenteeism available. The database originally consisted of 193,640 employees but after database clearing for research purposes and applying the mentioned criteria, 46,069 subjects were excluded (72% women, 28% men), leaving a population of 147,571 individuals. There was no information on the actual number of employees that were invited to respond to the survey, the only information available was the response rates given above. Because the main analyses used listwise deletion, 17,410 subjects with missing data for any of the variables included in the analyses were also excluded, giving a final study population of 130,161 employees, of which 73% were women and 27% men. The study population is described in Table 1. The majority was employed for municipalities, followed by county councils. The most commonly reported workplaces were schools (municipalities), hospitals (county councils), and post offices (state-owned). This information should be treated tentatively, however, since information about subject’s workplaces was missing for 45% of the population. As can be seen in the table, 12% of the study group reported being sickness present on more than 5 occasions over the previous year, and 8% reported more than 25 days on sick leave during this period. Among the 17,410 excluded employees, 63% were woman and 37% men; the mean age of this group was 47.0 years (sd 10.8) and the extent of presenteeism and absenteeism was almost identical to the study group.

The psychosocial workplace factors were assessed using the General Nordic Questionnaire for Psychological and Social Factors at Work (QPSNordic) if no other means of assessment is given (Dallner et al. 2000). The variables included in the questionnaire were primarily chosen based on the theoretical assumptions described in the introduction and on the results from the meta-analysis of Miraglia and Johns (2016). Furthermore, in line with this meta-analysis, global variables were preferred when possible, that is, different aspects of a construct were merged into one single index (see below). All QPS-scales have five scale steps from 1 to 5.

For the background factors of gender, age, education, and employer, we used single items (Bergstrom et al. 2008).

Work motivation was measured with a modified work motivation scale developed by Björklund (Björklund et al. 2013). The questions included were the following: “Do you feel stimulated by your work tasks?”, “Are you motivated to work?”, “How often do you feel a strong will to work?” and “Would you spend less time at work if possible?” The response format ranged from “1 = never” to “5 = always”. Cronbach’s alpha was 0.78.

Sickness presenteeism was measured using the following item: “Has it happened over the previous 12 months that you have gone to work despite feeling that you should have taken sick leave due to your state of health?” The response options were “1 = no, never,” “2 = yes, once,” “3 = yes, 2–5 times,” and “4 = yes, more than five times” (Aronsson et al. 2000).

Sickness absenteeism was measured using a question from the Work Ability Index (Ilmarinen 2007). The question used was “How many days in total have you been away from work due to your own illness (sick leave, health care, treatment or examination)?” The response options were “no days,” “1–7 days,” “8–24 days,” “25–99 days,” and “100–365 days.” The categories “25–99 days” and “100–365 days” were merged for our analyses.

General health. Health-related quality of life (HRQoL) was assessed using the Swedish version of the EuroQol EQ-5D-3L (Bjork and Norinder 1999; Brooks 1996). The EQ-5D is a generic HRQoL questionnaire based on five dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. These dimensions can be used to generate individual health profiles that represent health states. In this study, we applied utility weights of health states from a general population using the Danish time-trade-off values to convert the health profiles to values between 0 and 1 (Wittrup-Jensen et al. 2009). A value of 0.00 indicates the worst possible health state while a value of 1.00 indicates the best.

The variables were first checked for their statistical distribution and then correlational analyses were conducted using Pearson’s r. Secondly, we estimated the different parameters of the path model with manifest indicators, and their direct and indirect effects on sickness absence and sickness presence. Structural equation models (SEM) and the data program AMOS 25 was used. To assess model fit, we used different indicators. These were the root mean square residual (RMR), which is the square root of the discrepancy between the sample covariance matrix and the implied covariance matrix. Values close to 0 represent a good fit and, as a rule of thumb, values < 0.08 are deemed to be acceptable (Hooper et al. 2008). We also used the goodness of fit index (GFI). The GFI can be compared to R-squared, and ranges from 0 to 1 and is considered satisfactory when > 0.90. Furthermore, we used the adjusted goodness of fit (AGFI) acceptable value > 0.90, the comparative fit index (CFI) acceptable value > 0.95 and the Root mean square error of approximation (RMSEA), acceptable value < 0.10. We also calculated a Bayesian posterior predictive p-value, which should be near 0.5 for a correct model, with values toward the extremes of 0 or 1 indicating that a model is not plausible.

Since some of our variables were categorical and/or skewed, we decided to use the asymptotic distribution-free (ADF) method (Jones and Waller 2015). Moreover, to handle skewness resulting from the indirect effects (mediating effects) produced by the direct effects and their distributions, we used 5000 bootstrap replicates to estimate bias-corrected confidence intervals and p-values (Lockwood CM and DP. 1998; Valente et al. 2016). Finally, to check our estimates, we ran Bayesian estimation via the Markov chain Monte Carlo (MCMC) algorithm with a uniform prior (Byrne 2010; Kaplan and Depaoli 2012). The estimates of the two methods were close and in general identical when rounded to the second or third decimal place. In the sequel, we report the estimates from the ADF bootstrap procedure if not stated differently. Listwise deletion of data was used, since Amos cannot accept missing data when using any estimation criterion, such as ADF, except for when using maximum likelihood. In the results section, we will use the concept of “effect” in a statistical sense; that is, we do not imply causal effects between variables.

Table 2 shows the correlations (Pearson) between variables in the model. As could be expected, sickness absence and sickness presence were found to be positively correlated (0.31). The mediating variables—health and motivation—negatively correlated with sickness presenteeism and with sickness absenteeism in both the correlation (Table 2) and mediation analyses (Table 4); that is, the better the health and the higher the motivation, the less presenteeism and the less absenteeism. All work factors and the two mediating variables showed relatively higher correlations with sickness presence than with sickness absence. The associations between age and sickness presence and age and sickness absence are negative, which indicates that sickness presenteeism as well as sickness absence decrease with increasing age.

Fit indices for the SEM model are presented in Table 3. The AGFI, CFI, and RMSEA indicated a poor fit, whereas the RMR and GFI were acceptable according to the indices. The generally poor fit may partially be explained by the relatively weak correlations given in Table 2 (Hooper et al. 2008). However, to investigate the model fit further we dissembled the model to separate smaller submodels to identify possible problem areas. From this we discovered that the variable job demands, and its low correlation with several of the other variables, was causing convergence problems in the sub models. When it was removed the models fitted well and all coefficients were significant at the 0.05 level. However, as the overall model converged and to be true to the theoretical model described, we decided to keep all variables in the final model.

Figure 1 shows a simplified description of the analytic model depicting its indirect paths. Table 4 shows all direct, indirect and total standardised effects on presenteeism and absenteeism. Not unexpectedly, health has by far the strongest effect on sickness presence and sickness absence. In the following we use the word reduce instead of “negatively associated” etc. without indicating causality.

For two of the factors—job support and job demands—the mediation analysis contributed to nuanced images and increased understanding of the paradoxes in the research area.

The direct effect of job support was increased sickness absenteeism and the indirect effects were decreased sickness absence. Job support is also directly and indirectly associated with decreased presenteeism, which means that the total effects of job support is health promoting. The result may reflect that good job support encourages the kind of social climate where people do not suppress their health problems when ill. A trusting relationship with one’s supervisor and colleagues may also reduce feelings that sickness absence is unjustified. The need for sickness presence and sickness absence decrease via the health and motivation paths. In the meta-analytical study (Miraglia and Johns 2016), job support increased sickness presence via the motivation path, which probably reflected that high motivation stimulates working when ill. Why the Swedish workers act in the opposite manner is a question for further research.

The relationship for job demands is also somewhat complicated and unexpected. The direct effect of increased demands is increased presenteeism (0.076) and decreased absenteeism (− 0.049). Job demand is also connected to decreased absenteeism via the health and motivation path. The indirect effect on sickness presenteeism is a reduction in presenteeism but the effects are not so strong that they outweigh the strong direct effect in opposite direction. Paradoxically, high work demands appear to be important factor for low absenteeism. However, some of the “good” effects of high demands on sickness absence seem to be the result of substitution (total effect 0.040). Our study was cross-sectional so we cannot draw any conclusions that there are good long-term effects. Actually, it may also be that motivated employees with good health may be given, or take on the responsibility for, challenging work tasks. Previous research has shown that high sickness presence is a long term risk factor for health (Bergstrom et al. 2009) and for sickness absence (Bergström et al. 2009).

The result in our study was the reversed to the association in the meta-analytic study, where high demand was associated with high absenteeism. We can only speculate about the reason for this. One reason could be different samples. To a large extent the Swedish sample consisted of white-collar workers (60 percent had university or high school education). Increased demand in these groups may mean more of positive stimulation than of harmful load.

There may also be methodological reasons for the contrasting results. A general challenge with interpreting results on job demands concerns its potential non-linearity. A moderate increase in job demands from a low or very low level may be considered to be a health-promoting and motivating characteristic of work, provided that job resources are present, whereas an increase in job demands among employees who already work under higher demands tends to be health eroding. If this Swedish population differs from the population investigated by Miraglia and Johns (2016) concerning the distribution of the scores of the job demand variable, there could hypothetically be such effects. However, to what extent non-linearity may operate in this way and influence the results with respect to our rather heterogeneous study data is impossible to know. The effect of non-linearity is a matter for further research.

For the three remaining psychosocial factors the results were mainly in accordance with what could be expected considering previous research. Job control were directly and indirectly associated with lower absenteeism and lower presenteeism. However, in the meta-analytical study, job control increased sickness presence via the motivation path, i.e., the same difference was observed as for job support. Role conflict and work to family conflict were associated with higher presenteeism and higher absenteeism. It is of interest that the total effect of work to family conflict on presenteeism was as high as 0.21, which is far higher than any other effect.

The study has had the ambition to contribute to a more holistic research model by analyzing mediation and combining sickness presence and sickness absence in the analyses. One conclusion is that the individuals’ decisions between engaging in sickness absence or sickness presence, i.e., substitution should have higher priority in future research, both for theoretical and practical reasons (Cooper and Lu 2018). There is also a need to develop valid measures that combine sickness presence and sickness absence (Aronsson et al. 2011; Caverley et al. 2007).

In the current study the effects of job demands were unexpected and differed from the cited meta-analysis (Miraglia and Johns 2016). A possible reason for may be the aforementioned non-linearity. The differing results may also reflect that there are different types of job demands. In occupational research, demands may be divided into quantitative and qualitative demands as well as into physical, emotional, and cognitive demands. In our study, the demand scale measured quantitative demands and demands for attentiveness and for quick or difficult decision-making. Interesting in this context is a recent meta-analytical study of the reasons for going on disability pension, which can be seen as an endpoint of sickness absence (Knardahl et al. 2017). That study found very limited evidence for an effect of job demands on disability pension. This suggests that there is a need for more specific knowledge on the relations between different types of job demands and health, motivation, sickness absence, and sickness presence. A more differentiated job demand measure is a recommendation for further research and stratified analyses related to skill demands.

Motivation as mediator seems to play somewhat different roles in our study and in the meta-analytic study (Miraglia and Johns 2016). It is difficult to understand and interpret but one possible reason may be that the corresponding indicator in the meta-analytic study was job attitudes, a factor that was composed of both motivation and job satisfaction, the latter a factor that was missing in our data.

The amount of data utilized in this study was large. It was mainly gathered from public-sector employees in Sweden, and care personnel and schoolteachers were the largest occupational groups—characteristics that deserve consideration when generalizing the results. As always, the use of cross-sectional self-reports may increase the risk for common method variance. However, for sickness presenteeism, it is difficult to find data sources other than self-reports.

A weakness is that the questionnaire was not primarily created for the purposes of conducting mediation analyses of work factors and presenteeism and some variables were questionable as indicators of comparisons with the meta-analytic study. This is something that should be considered in further research.