Retrospective Analysis of Individuals with at least Two Years Outside Working Life - Finnish Population-Based Study

Background There is limited knowledge on service and benefits use behind long-term unemployment. In this study, we identify different pathways behind long-term unemployment utilising information on previous labour market position. Further, we analyse differences in health care and social benefits use based on these pathways. This study aims to produce knowledge useful for early identification and supportive actions for those outside of labour market or in risk of such. Methods Among Finnish working age population, individuals who had been mainly unemployed in years 2020 and 2021 and had information on their labour market position for years 2013-2021 were identified (n = 72,485). For the analysis, register information from Statistics Finland, the Social Insurance Institution of Finland and Finnish Institute for Welfare and Health was used. To identify diverse pathways based on previous labour market position, sequence analysis was conducted. Linear regression models were used to analyse cluster differences in health care and social benefits use while controlling for potential confounders age, sex, education, living situation, migrant background and low income. Results Five clusters were identified based on previous labour market position: unemployed (54%), students (3%), employed (23%), inactive (10%) and individuals with unstable working career (10%). Health care attendance was most frequent among those inactive and unemployed. These clusters had also highest share of those with mental and musculoskeletal disorders and were most with sickness absence and with basic social assistance. In these associations between health care and benefits use and clusters, confounding due to sociodemographic factors seems to play a minor role apart from students, with whom we saw decreased risk of mental health diagnoses compared to unemployment when confounding due to age and sex was taken into account. Income differences between clusters were large in beginning of the observation years. In the end of the observation time, unemployed, inactive and students were at similar level of income while those in employment and those returned to unemployment had higher earnings. Conclusions Previous labour market position should be considered while planning preventive efforts and developing individual multisectoral services and social security for those in risk of labour market marginalisation.