Article Text
PDF
Abstract
Objectives The use of part-time sickness absence (pSA) enables return to part-time work from full sickness absence. However, subsequent labour market outcomes of pSA users depend on various individual and work-related characteristics. We investigated labour market paths of private and public sector employees after having a pSA spell. Moreover, we examined individual and work-related factors associated with following them.
Design Longitudinal register-based cohort study.
Setting Finnish employed population.
Participants 9896 receivers of partial sickness allowance aged 45–56 in the years 2010–2014.
Outcome We constructed labour market trajectories based on the proportion of time spent in various labour market statuses measured over 3 years after the end of the pSA spell using multiresponse trajectory analysis. We then examined how different individual and work-related factors were associated with assignment to the different trajectory groups using logistic regression analyses.
Results The majority of the pSA users followed paths where work participation was consistently elevated (Sustained Work group, 40.4%), or only slightly reduced (Slightly Reduced Work group, 31.6%). Moreover, more than 1/10th of the users followed a path where receiving partial work disability benefits became predominant (Partial Work Disability group, 12.5%). The rest followed paths where other non-employment (Other Non-Employed group, 7.8%) or full work disability (Full Work Disability group, 7.7%) became the prevailing status. Lower educational level and income predicted assignment to all other groups than the Sustained Work group. Additional predictors were identified, yet these differed between the trajectory groups.
Conclusions The majority of the pSA users maintained a connection to working life, yet weaker working life paths were also identified. The paths were determined by various individual and work-related factors that can help health professionals and employers to better target support measures particularly towards individuals whose connection to working life is at risk to weaken after the use of pSA.
- REGISTRIES
- Risk Factors
- Health policy
- Retrospective Studies
Data availability statement
Data may be obtained from a third party and are not publicly available. We used register data linked together from Statistics Finland, the Finnish Centre for Pensions and the Social Insurance Institution of Finland. The authors do not have the permission to share this third-party data. Due to data protection regulations, the data can only be accessed by individual researchers who have obtained permission to use the data through an application process (https://www.findata.fi/en/services/data-permits/, https://www.stat.fi/tup/mikroaineistot/hakumenettely_en.html).
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
The large register-based dataset was representative of Finnish part-time sickness absence users and did not suffer from missing information due to non-response or attrition.
The 3-year follow-up time enabled us to capture long-term patterns of labour market participation after the use of part-time sickness absence.
The data included comprehensive information on employment and the use of social security benefits for the basis of determining various labour market statuses.
By using job exposure matrix, we did not capture variation in the working conditions between individuals holding the same occupational title.
We did not use specific diagnoses but main diagnostic groups, within which there may be heterogeneity among the cases.
Introduction
Through the last decades, Finland and other Nordic countries as well as some countries in continental Europe have strongly promoted part-time work during sickness absence (SA) in order to fasten return to work and to increase work participation.1 2 Consequently, the proportion of part-time SA (pSA) of all compensated SA, including also full SA (fSA), has been increasing, being around 8% in Finland in 2021.3
Most previous studies suggest that pSA or graded return to work instead of fSA reduces the duration of SA, enhances return to work and increases overall work participation,4–10 yet some studies have not found such effects.11 12 Moreover, labour market outcomes after pSA may depend on various individual and work-related characteristics. More favourable labour market outcomes after pSA or graded return to work have been reported, for example, among persons with mental disorders than musculoskeletal diseases5 7 13 and among private than public sector employees,7 while the influence of some other factors such as physical and psychosocial occupational exposures and other disease groups remains unclear.
Moreover, following pSA, there is a large number of possible transitions, that is, transitional paths between different labour market statuses such as work, partial and full work disability, and unemployment. Return to work or exit from the labour market may, therefore, often be multinatured and multiphased processes rather than consisting of single events. Even though previous studies have examined average participation in different labour market statuses over time after the use of pSA (eg,7), only little is known of the potentially heterogeneous paths and factors associated with following them.
A more comprehensive investigation of labour market trajectories after the use of pSA would provide important insight into the factors associated with labour market participation after returning to part-time work from SA. Moreover, information on factors associated with following certain paths can help to better target work-promoting interventions and utilising the remaining work capacity.
Using Finnish register data and multiple response trajectory analyses, we investigated typical labour market paths of 45–56 years old employees over a 3-year period after having a pSA spell. In addition, we examined individual and work-related factors associated with being assigned to particular trajectory groups after the use of pSA.
Material and methods
Data sources
As the base data, we used a nationally representative 70% random sample of the working age population living in Finland on the last day of year 2007. Register-based longitudinal information was available for this sample until the end of October 2017. The data included information on episodes of employment, unemployment, earnings-related pensions and vocational rehabilitation within the earnings-related pension scheme obtained from the Finnish Centre for Pensions, on episodes of compensated SA, national pensions, and vocational and other rehabilitation obtained from the Finnish Social Insurance Institution, and on sociodemographic and work-related factors obtained from the FOLK data of Statistics Finland. Data from these three register holders were linked on the basis of social security numbers of the participants, pseudonymised for analyses. Due to the inclusion of sensitive data, such as SAs with medical diagnoses, our sample size was limited to 70% of the total population.
Study design
In Finland, compensation for SA by the Social Insurance Institution has a waiting period of 10 weekdays (including Saturday) that is typically paid by the employer. At the time of our study, the waiting period was always spent on fSA, while part-time work during the waiting period became an available option from the beginning of 2023. After the waiting period, the Social Insurance Institution of Finland starts to pay sickness allowance, which we refer to as compensated SA. This study only covers information on compensated SA periods. During compensated SA, pSA is a voluntary option for persons who are eligible for continuing fSA, if based on medical assessment they can work without harm to their health and part-time work can be arranged by their employer. Partial sickness allowance is 50% of full allowance, and the employee returns to working 40%–60% of the time while receiving it. In Finland, pSA is typically preceded by a compensated fSA period.7 14 The maximum length of pSA was 72 weekdays until 2013 and 120 weekdays since 2014 (after the time of our study this was changed to 150 weekdays from the beginning of 2023).
We carried out a register-based study of private and public sector employees who had a pSA spell due to any medical reason starting after the beginning of January 2010 and ending by October 2014 (hereafter called index spells) and who turned 45–56 years old during the year the index spell ended. This recruitment period was set due to a legislative change on 1 January 2010 that enabled the use of pSA immediately after the waiting period of 10 weekdays and our data being available until the end of October 2017, thereby allowing the follow-up of labour market participation for each study person over a full 3-year period after the last day of the index spell. We chose the first pSA spell of an individual occurring during the recruitment period. The spells were interpreted as one if there were no more than 30 days between them. A large proportion (46%) of these pSA spells lasted 60–90 days. The final study population consisted of 9896 individuals.
Labour market statuses
We used information on episodes of employment and of receiving social security benefits to calculate the proportion of time spent in different daily measured labour market statuses within every 3 months over the 3-year follow-up (36 months) period after the index pSA spell, resulting in 12 follow-up periods. The statuses were (1) work (without receiving work disability, unemployment or pension benefits), (2) partial work disability (pSA, temporary or permanent partial disability retirement), (3) full work disability (fSA, temporary or permanent full disability retirement, vocational and other rehabilitation) and (4) other (mainly unemployment, but including also other type of being outside the work force or being deceased). The proportions of time spent in more specific statuses are presented in online supplemental table 1.
Supplemental material
Covariates
We included individual and work-related factors as covariates, that is, background characteristics that are known to be associated with work (dis)ability or unemployment, including age,15–19 gender,15–20 socioeconomic status,16 18 19 that is, education (tertiary, secondary, primary) and income (taxable income containing annual earnings and social security benefits), employment sector15 (private, public), physical heaviness of work16 17 21 (proportion exposed), job control19 22 23 (mean score), as well as diagnostic group16 24 and calendar year of the index pSA spell.
Age at the year the index spell ended was examined in groups 45–50 and 51–56 years. The employment sector was based on the sector of pension-insured employment at the beginning of the index spell. Education and occupation were measured at the end and income during the year preceding the start year of the index spell.
Information on occupation was used for defining the occupational exposures. If occupation could not be defined for the year preceding the index pSA spell, the information was derived from the 4 years prior to that. Occupational exposures, that is, physical heaviness of work and job control, were then estimated using gender-specific job exposure matrices (JEM), which were developed earlier based on a large population survey and are described in more detail in previous studies.25 26 We chose to examine these two exposures as previous studies have provided consistent evidence on the associations of physical workload and job control with work participation.27–29 The JEM for physical heaviness of work and job control have also shown good validity.25 26 The JEM information was linked to occupational codes in the register data using a classification by Statistics Finland (Classification of Occupations 2010), based on the International Standard Classification of Occupations (ISCO-88).
The medical reason for the index pSA spell was classified according to the International Classification of Diseases, 10th Revision (ICD-10). We examined the following disease groups: (1) diseases of the musculoskeletal system and connective tissue (M00–M99), (2) mental, behavioural and neurodevelopmental disorders (F00–F99), (3) diseases of the nervous system (G00–G99), (4) diseases of the circulatory system (I00–I99), (5) neoplasms (C00–D49), (6) injury, poisoning and certain other consequences of external causes (S00–T88) and (7) other diseases.
Statistical analyses
We constructed labour market trajectories based on the proportion of time spent in the four different labour market statuses measured within every 3 months over the 3 years after the end of the index pSA spell using multiresponse trajectory analysis . 30
The best fitting number of latent subgroups and their shape were determined in terms of model fitting based on the Bayesian information criterion. The normal distribution was used as the underlying statistical model.
We then examined how the different covariates were associated with assignment to the different trajectory groups including group membership as dummy outcomes in logistic regression analyses. We assessed age-adjusted and gender-adjusted models as well as models mutually adjusting for all covariates. The results are presented as OR and their 95% CIs.
Stata V.15.1 (StataCorp) was used in all analyses.
Results
Almost 55% of the study population of the pSA users belonged to the older age group 51–56 years and over 76% were women. Approximately 45% had a musculoskeletal disease and almost one-third had a mental disorder as the reason for their pSA spell. Further characteristics of the study population are presented in table 1.
Distribution of background characteristics among the study population of part-time sickness absence (pSA) users
Trajectory groups
Five labour market trajectory groups were identified (figure 1). In the largest group (40.4%), the proportion of time spent working was at a high level throughout the 3 year follow-up (Sustained Work group). In the second largest group (31.6%), the proportion of time spent working was at a high level but slightly declined during the follow-up and was replaced mostly by an increase in the time spent with partial work disability (Slightly Reduced Work group). In the third largest (12.5%) group, the proportion of time spent working was at a medium level, then strongly declined and was replaced particularly by an increase in the time spent with partial work disability (Partial Work Disability group). In a smaller group (7.7%), the proportion of time spent working was at a medium level, then strongly declined and was replaced particularly by an increase in the time spent with full work disability (Full Work Disability group). Finally, in another smaller group (7.8%), the proportion of time spent working was at a relatively high level, then steadily declined and was largely replaced by an increase in the time spent in other labour market statuses (Other Non-Employed group).
Empirically derived groups of labour market trajectories during the period of 3 years after the use of part-time sickness absence.
Predictors of assignment to the trajectory groups
Assignment to the largest group with the best attachment to work, that is, to the Sustained Work group, was predicted by younger age (45–50 years), male gender, having tertiary education, annual income more than €60 000, an occupation where the exposure to physically heavy work or low job control was less likely and the pSA was due to a neoplasm or injury (table 2). After mutual adjustment, assignment to this group was predicted also by having pSA due to a circulatory disease, whereas the associations of gender and job control disappeared (table 3).
Age-adjusted and gender-adjusted predictors of being assigned to different groups of labour market trajectories after the use of part-time sickness absence
Mutually adjusted predictors of being assigned to different groups of labour market trajectories after the use of part-time sickness absence
The remaining four trajectory groups had two common predictors: having less than tertiary education and annual income less than €30 000, although the association with educational level generally weakened after mutual adjustment.
Moreover, after mutual adjustment, assignment to the Slightly Reduced Work group was predicted by female gender, public sector employment and having an occupation where exposure to physically heavy work was more likely. Assignment to the Partial Work Disability work group was predicted by older age and public sector employment, while on the contrary, assignment to the Other Non-Employed group was predicted by younger age and private sector employment. Finally, assignment to the Full Work Disability group was predicted by older age, male gender and having pSA due to a nervous disease.
Discussion
Using nationally representative data on pSA users aged 45–56 years who had returned to part-time work from fSA, we identified five typical labour market participation trajectories over a subsequent 3-year follow-up period over the years 2010–2017. To our knowledge, this is the first study to provide information on latent trajectories of labour market participation after the use of pSA.
In this study, the majority of the pSA users followed paths where work participation was constantly at a high level or only slightly reduced after the pSA spell. These favourable outcomes in terms of work attachment are in accordance with previous studies showing that the use of pSA instead of fSA has overall positive effects on later work participation.4–10 Returning to part-time work with the use of pSA may facilitate later return to full work duties, improve labour market attachment, and have positive health effects. Even though generally the evidence of the effect of work on health is somewhat limited, work participation has been reported to have several beneficial effects on mental31 and physical health32 as well as on general well-being.33
Moreover, over 1/10th of the pSA users followed a path where receiving partial disability benefits became common. Since those having partial work disability typically work at least part of the time, this outcome can be considered as successful with regard to work attachment, at least when return to full work duties is not possible. For example, in year 2021, approximately 80% of partial disability pensioners in Finland were partially at work.34 Also, previous studies7 10 35 36 have shown that the use of pSA often initiates a more long-term partial work disability path, during which the individuals continue to participate partially in the labour market.
The remaining persons, slightly over 15% of the pSA users, followed paths where work attachment substantially weakened, which was attributable to the increased time spent in other non-employment or full work disability. For these people, initial part-time return to work while being sick does not appear to be sufficient to ensure sustained work participation in the long term. Additional interventions improving their employability, health and work ability may therefore be needed.
Predictors of labour market paths after pSA
In this study, we found several individual and work-related characteristics that may help health professionals, employers and human resource personnel to anticipate and influence labour market outcomes after pSA. Those who were younger, had higher education and higher income more often followed the labour market path most attached to work after pSA. These characteristics are likely to be associated with better work ability or employment opportunities more generally.
Some gender differences were also observed. We found that men were more likely to follow a labour market path where working was replaced by full work disability, whereas women were more likely to follow a path where work participation only slightly reduced. Men may be employed in types of work where modifying tasks is more difficult, and therefore, returning to work is more challenging. Moreover, previous studies have shown that men have fewer contacts with healthcare services37 38 and lower SA rates20 39 than women. This may indicate that men have poorer or delayed access to treatment or that men are less eager to report their disabilities to healthcare professionals. Hence, when men take SA, it may be at a later and more severe stage of work disability compared with women, resulting in less favourable labour market outcomes also after the use of pSA.
Working in the private sector compared with the public sector was particularly strongly associated with following the path where working was replaced by other non-employment, which was typically unemployment. This may be related to the fact that employment careers are less secure in the private than the public sector. On the contrary, working in the public sector was strongly associated with following the path where work was replaced by partial work disability. This can be due to system-based reasons. In the private sector, the definition of incapacity for work due to illness is based on the possibility of doing any type of work that a person can be expected to perform, considering his or her background such as education, age and work history. In the public sector, the incapacity for work is more narrowly defined on the based on professional criteria, only considering the person’s performance in their own work. It appears that in the public sector all attempts are made to keep the person in his/her occupation at least part time, thereby encouraging the partial work disability path. There are also differences between the sectors in determining the disability pension payments of large employers. Contrary to the private sector, in the public sector partial disability pensions do not directly affect the determination of the disability pension contributions, which can encourage their use. Consequently, the risk of partial disability retirement has been shown to be higher in the public sector than in the private sector.40 Moreover, previous studies have found that partial disability retirement is more common in larger than in smaller workplaces,41 42 which may indicate that it is easier to provide part-time work in larger working units. This may partly explain the differences in the use of partial disability retirement across sectors, as work units in the public sector are usually larger compared with those in the private sector.
Musculoskeletal and mental disorders were the most common reasons for the use of pSA. According to our findings, those suffering from these diseases, however, did not follow any distinct labour market path. Instead, several smaller disease groups were associated with a specific labour market path. Those with an injury, circulatory disease or neoplasm were more likely to follow a path that was best attached to work after the use of pSA than those with other diagnoses. For these diseases, effective treatment is often provided and work ability can be fully restored. Instead, those with nervous diseases were much more likely to follow a full work disability path than those with the other diagnoses. It may be that the chronic and progressive nature of several nervous diseases leads to a long-term work disability. However, the treatments for diseases constantly develop, and labour market outcomes may thereby improve in the future.
Occupational exposures appeared to influence the outcomes to a limited extent especially once all background factors were controlled for. Those having an occupation where physically heavy work is less common were more likely to follow the labour market path that was best attached to work. Returning to full work duties after pSA and staying at work might be easier for those working in physically lighter occupations. Several studies have documented the negative impact of physically heavy work on health, work ability and the risk of SA and early retirement.43–47 However, in this study, having a physically heavy occupation was associated with following a path where working only slightly reduced, but not for example, with a path where the connection to working life was lost due to full work disability.
High job control had only a weak association with the labour market path best attached to work, which disappeared after all other background factors were accounted for. Although previous studies have found associations of high job control with retention in paid employment and having less SA,48–50 it appears based on our findings to have less influence on further labour market outcomes among pSA users who have already developed a disease causing the work disability.
These limited associations between occupational exposures and labour market outcomes after the use of pSA may partly be attributable to the circumstance that reduced work time during pSA is by itself a measure to reduce exposure to harmful physical and psychosocial conditions at work. Users of pSA are a selected group of sick-listed employees who were willing to return to part-time work and for whom this could be arranged by their employer.
Strengths and limitations
The strengths of our study include rich nationally representative register data that do not suffer from non-response or attrition during the follow-up. The 3-year follow-up time enabled us to capture long-term patterns of labour market participation after the use of pSA. Moreover, the data included detailed information on employment and the use of social security benefits for the basis of determining various labour market statuses.
Some limitations should be considered when interpreting the results. The use of JEM can be seen as both a strength and a limitation. JEM-based exposure estimates are not prone to recall bias or other types of information bias. Furthermore, JEM is a useful tool for exposure assessment in register-based studies with no individual-level information on occupational exposures. However, the limitations of the JEM should not be neglected. Earlier studies have reported that job titles could explain on average about 30% (range 5%–55%) of the variance in individual self-reported exposures,51–53 yet the value varies across different exposures. By definition, a JEM assigns the same exposure estimates to all workers within the same occupation. However, within the same occupation, exposures might vary within worker (over time variation) and between workers (variation in tasks, activities and work processes). By neglecting within occupation variation, exposure assessment by JEM induces a non-differential misclassification bias,54 which will attenuate the observed associations between an exposure and outcome towards null.55 The performance of the JEM will also be reduced when the variation of an exposure within the occupation is larger than between the different occupations. The JEMs that were used in our analyses have shown a good accuracy, especially for job control and physical exposures25 26 and a good predictive validity for various health-related outcomes.56–59
However, by using the JEM, we did not capture variation in the working conditions between individuals holding the same occupational title and likely underestimated the associations between exposures and our outcome of interest. This may partly explain the limited contribution of occupational exposures to our findings. Another study limitation is that we did not use specific diagnoses but main diagnostic groups, within which there may be heterogeneity among the cases. Importantly, the distribution of diseases has changed after the time of our study, with mental disorders being now the largest disease group among pSA users in Finland.60
Conclusions
The majority of the pSA users maintain a connection to working life after their pSA spell at least via partial work participation. The labour market paths after pSA are diverse and associated with various individual and work-related factors that can help, for example, health professionals and employers to anticipate and influence future labour market outcomes. Further studies are needed to investigate, for example, how labour market paths after the use of pSA differ between genders, how part-time working solutions could be better used in the private sector and how to support work participation among those with reduced work ability due to neurological diseases. Moreover, future studies should focus on investigating how individual-level differences and changes in working conditions contribute to work participation after pSA.
Data availability statement
Data may be obtained from a third party and are not publicly available. We used register data linked together from Statistics Finland, the Finnish Centre for Pensions and the Social Insurance Institution of Finland. The authors do not have the permission to share this third-party data. Due to data protection regulations, the data can only be accessed by individual researchers who have obtained permission to use the data through an application process (https://www.findata.fi/en/services/data-permits/, https://www.stat.fi/tup/mikroaineistot/hakumenettely_en.html).
Ethics statements
Patient consent for publication
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @ElliHartikainen
Contributors EH, LS, SS, JB, PB, TL and EV-J participated in designing the study, contributed to interpretation of the results, discussed and revised the drafts, and prepared the final manuscript. SS, EH, TL and LS prepared the data for analyses. EH conducted the statistical analyses, wrote the first drafts of the article and acts as the guarantor of this article.
Funding This study was supported by the Finnish Work Environment Fund (project number 190400) and Keva.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.