Main Findings
The aim of this study was to investigate longitudinal associations between demographic, personality, disorder-related and work-related characteristics and sustainable RTW in 2 years of sick-listed workers with a lifetime diagnosis of a depressive or anxiety disorder. In 2 years, 51.6 % of the study participants returned to work sustainably. This study revealed that in the long-run not disorder-related factors, but a younger age, a higher household income level and being (self-)employed are all together associated with a higher odds of sustainable RTW in 2 years of sick-listed workers with a depressive or anxiety disorder.
Comparison with Other Studies
Most of the participants in this study had currently been diagnosed with a depressive and/or anxiety disorder at baseline. Earlier research within NESDA reported a twofold and a sevenfold higher risk of long-term sickness absence for persons with respectively an anxiety disorder or depressive disorder in the same period that the disorder was present, so cross-sectional [
26]. We selected participants of NESDA for our study, based on their long-term sickness absence. Since participants with a depression had the highest risk of long-term sickness absence, it is not surprising that many of our respondents were diagnosed with a current depression at baseline. Another study within NESDA revealed that persons with a depression, are also most likely to have recovered in 2 years [
37]. This might be an explanation for the absence of an association between the presence or severity of the disorder at the moment of sick-listing and RTW 2 years later. Moreover, our findings confirm that when one’s aim is to enhance sustainable RTW of sick-listed workers with mental health problems, it is not sufficient to solely focus on characteristics of the disorder itself, which has often been done in previous studies [
18].
The influence of a broad range of factors on RTW has been studied before in study populations consisting of sick-listed workers with physical complaints, such as low back pain. Results of these studies emphasize the importance of work-related factors in RTW, such as job satisfaction, social support, job demands and job control [
38‐
41]. In our study, univariable associations were found between sustainable RTW in 2 years and two work-related factors: a high job security and a high skill discretion. However, in the combined model, the associations between sustainable RTW and these work-related factors, did not remain significant. This may be explained by the high number of participants that was on sickness benefit at baseline. They probably had no (longer a) workplace to return to, so that characteristics of the job influenced RTW to a lesser extent.
More than half of the participants in our study reported at baseline that they were on sickness benefit. They had a more than two times lower odds of returning to work in 2 years than participants who at baseline reported that they were (self-)employed. In the Netherlands, unemployed workers, temporary agency workers and workers with an expired fixed-term contract who become sick-listed can apply for a sickness benefit from the Dutch SSA. Both unemployment and temporary employment have been related to poor (mental) health [
11,
15,
42‐
44]. Nevertheless, it seems that these workers are not sick-listed more often [
45,
46], but when they do get sick-listed, the absence of a workplace to return to will complicate their RTW importantly [
15]. This stresses the need for vocational interventions that create a RTW perspective [
47,
48], i.e. interventions that focus on a suitable job for vocational rehabilitation. As evidence for effective vocational interventions for this vulnerable group of workers is lacking, more research on this topic should be promoted.
Besides the absence of a job to return to, also other obstacles for RTW might explain the reduced odds of sustainable RTW in 2 years for sick-listed workers on sickness benefit. It is possible that these workers experience a so called ‘benefit trap’. This means that the perceived (economic) benefits of staying out of work exceed the benefits of returning to work, for example because it is not possible to find a job that pays more than the income from being unemployed or sick-listed [
49]. This could also be an explanation for the reduced odds of sustainable RTW in case of a lower household income that was found in this study. A benefit trap might be experienced by the ones with a lower income.
Apart from the sick-listed workers without a (permanent) employment contract, also the older workers seem to represent a vulnerable group. This study showed that the odds of sustainable RTW of sick-listed workers with a depressive or anxiety disorder decreases significantly per each 10 years of age increase. This finding is highly supported by earlier research [
7,
11,
18,
19]. As the workforce is ageing, work participation of older workers is of growing importance. Based on an in-depth study of older workers’ perspectives and previous research, Koolhaas and colleagues [
50] proposed a tailor-made intervention with the aim to enhance sustainable working life, with a central focus on work-related problems and obstacles, personal development opportunities and environmental factors. Knowledge about the effectiveness of these kinds of interventions for older workers is needed.
Strengths and Limitations
Systematic reviews of the literature have shown that previous prognostic cohort studies more often addressed disorder-related factors, compared to work-related and personal factors, when studying RTW of sick-listed workers with mental health problems [
18,
19]. To our knowledge this has been one of the first studies that paid equal attention to the long-term influence of demographic, personality, disorder-related and work-related characteristics. This made it possible to study the independent effects of all these different factors and this is an important strength of our study.
A second strength of this study is that longitudinal associations were studied. All independent variables were measured at baseline. At this point all participants were sick-listed. In this way, all independent variables were measured prior to the possible occurrence of the outcome. Longitudinal associations provide more information than associations that are determined in a cross-sectional study, because with only cross-sectional data it is not possible to know whether an independent variable preceded the outcome or not. Moreover, assessing longitudinal associations between RTW and multiple factors, makes it possible to determine which of these factors have a long-term influence on RTW. This provides important information for policymakers who are engaged in the development of RTW policies.
Another strength of the study is that participants with a variety in duration of sickness absence and employment status were included in the study, which made it possible to investigate the influence of sickness absence duration and employment status on sustainable RTW. A disadvantage of our selection of participants is that the study population consists of participants with a probably worse prognosis than the source population of NESDA. Therefore, generalizing these results to other groups, such as workers who are only short-term sick-listed from a paid job, may be limited.
Another limitation of the study was the interpretation of the employment status of participants. In NESDA the Tic-P was used to collect information about the employment status of participants. In this study we assumed that the participants who indicated that they were on sickness benefit had no workplace to return to. In the Netherlands being on sickness benefit usually means that someone has applied for a sickness benefit from the Dutch SSA, because of the absence of an employer. However, as employment status was self-reported by the participants, we are not sure if the participants who had indicated that they were on sickness benefit actually had no (longer an) employment contract. Nevertheless, the sick-listed workers who had indicated that they were on sickness benefit differed significantly in outcome from the sick-listed workers who had indicated that they were (self-)employed.
The outcome measure, sustainable RTW in 2 years, was also assessed with the Tic-P [
25]. This questionnaire uses a reference period of 6 months. For that reason, it was only possible to know whether the participant had returned to work for a limited period of time (6 months). This is a limitation of our study. However, the follow-up period was more than these 6 months. Our outcome measure was assessed after 2 years follow-up, with a recall period of 6 months. As we were interested in return to work on the long run, the assessment of RTW after 2 years provided us with very valuable information. The measurement of the outcome with the use of the Tic-P, did not only show whether someone was at work in 2 years, but also provided some information about the sustainability of this outcome, because information was available about days of sickness absence in the previous 6 months.
The varying number of participants in the analysis due to missing values is also a limitation. However, the hypothesis that the values were missing completely at random could not be rejected. Imputation of missing data would probably not have provided new information. For that reason, we decided not to apply any data imputation techniques.
Practical Implications and Further Research
As long-term sickness absence is more and more caused by mental health problems [
2], it is for policymakers and occupational health care professionals important to know which (modifiable) factors influence sustainable RTW of sick-listed workers with mental health problems and to anticipate on this. This study reveals that in the long run characteristics of the disorder itself, such as duration and severity, do not influence sustainable RTW. Although work participation of sick-listed workers with mental health problems has still been studied mainly in regard with the disorder itself, there is a growing awareness of the importance of a healthy and steady job. The results of this study indicate that some workers are more vulnerable than others when becoming sick-listed. Especially older workers and workers without a (permanent) employment contract had a reduced odds of sustainable RTW in the long run. This might be explained by social-political factors, such as ageing of the workforce, the availability of jobs in the labor market and the increase of flexible employment relationships [
51]. RTW programs and practices should take this larger social-political context into account. Therefore, research aiming to investigate facilitators and barriers for RTW of more vulnerable groups of sick-listed workers can be highly recommended.