Effects of partial sick leave versus full-time sick leave on sickness absence and work participation: a systematic mapping review
Mapping review
|Published
The Norwegian Labor and Welfare Administration (NAV) commissioned the Norwegian Institute of Public Health to map all evidence on the effects of PTSL versus full-time sick leave (FTSL) on sickness absence and work participation.
Key message
In many countries, the high sickness absence rate in working age people is a concern. Partial sick leave (PTSL) is a return-to-work strategy that enables employees to be absent from work part of the time and remain working for a proportion of the time. The Norwegian Labor and Welfare Administration (NAV) commissioned the Norwegian Institute of Public Health to map all evidence on the effects of PTSL versus full-time sick leave (FTSL) on sickness absence and work participation.
Methods
We conducted a systematic mapping review. In January 2018, we conducted an extensive literature search, including searches in major databases, reference lists, grey literature, and we contacted labor agencies and international ministries. Two independent reviewers screened all retrieved records and appraised the included studies. We extracted data from the included studies and performed descriptive analyses. Synthesis of individual study results is not part of systematic mapping reviews.
Results
We included one small randomized controlled trial and 12 registry-based studies. The 13 studies included about 2.74 million employees on sick leave. The studies exhibited the following characteristics:
- Eleven of the studies were from Nordic countries, including four from Norway.
- All studies had either moderate or high methodological quality.
- The randomized controlled trial included Finnish employees (n=62) who were sick-listed due to musculoskeletal disorders, while the registry-based studies mostly included employees with either musculoskeletal- or mental disorders.
- There were 15 outcomes, of which the most frequently reported outcomes were work participation, sickness absence duration, disability, and social welfare benefits.
The findings indicated that PTSL may be associated with several favorable outcomes such as shorter sickness absence and higher work participation. However, firm conclusions about the effects of PTSL cannot be drawn due to the overwhelming majority of observational studies in this body of evidence.
Summary
Background
In many countries, high sickness absence rates mean that the need to increase work participation of working age people is a critical priority. Partial sick leave (PTSL), also called graded sick leave, is a return-to-work strategy that enables employees to be absent from work part of the time and remain working for a proportion of the time. It is understood as a graded leave, which allows people with reduced workability to work part time and still keep the link to the labor market. PTSL varies between 20% up to 99%, and might facilitate a progressive return to work. In many countries, especially the Nordic countries, this return-to-work strategy has been considered the first option to tackle increasing sickness absence rates and to prevent labor market exclusion.
Research on PTSL has shown largely promising results, especially in the Nordic countries. A recent evaluation of the PTSL arrangement in Norway, however, highlighted barriers to its use, such as bureaucratic red tape, lack of easy to understand and access information, and quality control. Although promising results have been published, research on the effects of PTSL has been criticized on methodological grounds, such as risk of selection bias and weaknesses in measuring work participation. To date, no systematic analysis of the evidence base on the effects of PTSL has been undertaken.
Objective
The Norwegian Labor and Welfare Administration (NAV) commissioned the Norwegian Institute of Public Health to map all evidence on the effects of partial sick leave (PLSL) versus full-time sick leave (FTSL) on sickness absence and work participation.
Method
We conducted a systematic mapping review according to international standards. A systematic mapping review (also known as systematic scoping review) is a review that maps and describes the existing literature or evidence base on a particular topic. There is no systematic synthesis of individual study results (e.g., no meta-analysis). We included studies in any language evaluating the effects of PTSL versus FTSL on sickness absence and work participation. Eligible study designs were prospective controlled studies as well as registry-based studies (RBs). The project team (reviewers) and commissioner (NAV) discussed and agreed on the research protocol.
We searched eight major databases, from inception to 2018, contacted experts, and hand searched websites of labor organizations and ministries, the bibliographies of all included studies, and literature reviews in the field. Two independent reviewers screened the retrieved references and data extraction was double-checked. We conducted independent quality appraisal of the included studies by using validated checklists. We grouped data extracted from the included studies according to their chief characteristics, performed descriptive analyses, and presented the results in text and tables.
Results
Thirteen studies, published between 2010-2017, met our inclusion criteria. All evaluated the effects of PTSL compared to FTSL in sick-listed employees. We included one Finnish randomized controlled trial (RCT) (n=62 sick-listed employees due to musculoskeletal disorders), and 12 RBs (n=2,742,497 sick-listed employees due to mainly musculoskeletal- or mental disorders). The participants in the 12 RBs were from Norway, Denmark, Finland, Sweden, and Germany.
Different grades of PTSL were observed across the studies. In the Finnish RCT, a 50% PTSL was given to 70% of all sick-listed employees, whereas 30% of employees worked shorter hours on 3–4 days per week. In the RBs, the most commonly used PTSL was 50%. The RBs analyzed nationwide registry data on sickness absences (both PTSL and FTSL) granted between 2001 and 2014, which indicated a 13-years’ timeline. Regarding the studies’ methodological quality, the included RCT had moderate quality, as did 11 of the 12 RBs while one RB had high methodological quality. It is important to note that RBs do not enable researchers to establish causal relationships between an intervention or exposure and outcomes.
Summary of main findings from the included studies
Sickness absence: The Finnish RCT and three RBs reported positive results of PTSL compared to FTSL on sickness absence at one-year follow-up.
Work participation:
Return-to-work: The Finnish RCT found that PTSL improved work participation compared to FTSL at the end of the one-year follow-up period. All of the nine RBs that measured work participation, except one Norwegian RB, showed favorable associations in employees on PTSL compared to FTSL.
Unemployment: Three RBs – from Norway, Germany and Finland – reported favorable effects of PTSL compared to FTSL on employees’ unemployment.
Degree of disability and rehabilitation benefits:
Recurrence of sick leave for any cause: The Finnish RCT found no significant differences between PTSL and FTSL in the recurrence of sick leave for any cause.
Disability: No differences between PTSL and FTSL were reported by the Finnish RCT at one-year follow-up and in one Norwegian RB, whereas two other RBs (from Norway and Finland) found that PTSL was associated with improvements on employees’ disability.
Productivity loss: Only the Finnish RCT reported on productivity loss. Data from this RCT showed there was no significant difference on productivity loss between PTSL and FTSL up to one-year follow-up.
Disability pension: One Norwegian RB reported that PTSL was associated with a higher rate of receiving disability pension compared to FTSL, whereas the two German RBs reported a decreased risk of receiving disability pension in employees on PTSL. A Finnish RB found that PTSL was associated with a lower risk of full disability pension compared to FTSL, whereas the opposite association was found for the risk of partial disability pension.
Allowance of social welfare benefits: Four RBs observed that PTSL was associated with a lower allowance of social welfare benefits compared to FTSL.
Health-related outcomes
Only two of the included studies reported on health-related outcomes. The
Finnish RCT found no differences between PTSL and FTSL on pain intensity, but showed positive results for PTSL on both self-rated general health and health-related quality of life. A German RB showed that PTSL was associated with better physical and emotional functioning in sick-listed employees. The Finnish RCT found no difference between PTSL and FTSL on sick-listed employees’ depressive symptoms, while the German RB showed that PTSL was associated with improvements on both depression and anxiety symptoms, and working ability, when compared to FTSL.
Conclusion
The evidence on the effects of PTSL compared to FTSL consists of one small RCT and 12 RBs, with a total of about 2.74 million study participants with mostly musculoskeletal- or mental disorders. The findings indicate PTSL may be associated with several favorable outcomes, such as higher work participation, but due to the overwhelming majority of observational studies in this body of evidence, firm conclusions about the effects of PTSL cannot be drawn. Both study designs suggested PTSL may be associated with shorter sickness absence duration and higher work participation. The Finnish RCT reported that employees with PTSL experienced better general health and quality of life compared to those on FTSL. However, it did not find statistical differences between PTSL and FTSL on sick leave recurrence, employees’ productivity loss, and pain. The RBs indicated a lower probability for people on PTSL of receiving both disability pension and allowance benefits, disability, as well as better scores on physical- and emotional functioning, anxiety, depression, and working ability.
Firm conclusions about the effects of PTSL are constrained due to the overwhelming majority of RBs in this body of evidence. Observational designs have considerable risk of systematic differences in the results, which make it difficult to answer questions about cause and effect. Further high quality RCTs are necessary in order to draw firm conclusions.
