Sickness absence among young and middle-aged workers in Norway: The impact of a population-level intervention

Background The study objective was to evaluate the impact of a population-level intervention (the IA Agreement) on the reduction of sickness absence (SA) among young and middle aged workers in Norway. Methods Using an observational design we conducted a quasi-experimental study to analyse registry data on individual SA for all employed individuals in 2000 (n=298 690) and 2005 (n=352 618), born in Norway between 1976 and 1967. The intervention of interest was the tripartite agreement for a more inclusive working life (the IA Agreement). We estimated difference in pre-post differences (DID) in SA between individuals working in IA companies with the intervention and companies without, in 2000 and 2005. We used logistic regression models and present odds ratios (DID OR) with accompanying 95% CI. We stratified analyses by sex, industry and company size. Results. We found no significant impact on the overall risk of sickness absence after implementing the intervention among young and middle aged workers. Stratified by sex, the intervention resulted in a slight decrease in SA among female workers (DID OR 0.93 (0.91-0.96)) while the intervention showed no impact among male workers (DID OR 1.01 (0.97-1.06)). We found that companies signing the IA Agreement were large companies (≥50­ employees) often within the manufacturing and health and social sectors. In large manufacturing companies, we found a reduction in SA among workers both in companies with and without the intervention, resulting in no statistically significant impact of the actual IA intervention on SA. In large health and social companies, we found an increase in SA among workers both in companies with and without the intervention. The increase was smaller among the workers in companies offering the IA intervention compared with workers in companies without, resulting in a positive impact of the actual IA intervention in the health and social industry. This impact was statistically significant only among female workers. Conclusions The results indicate that the impact of the IA Agreement on SA varies considerably depending on sex and industry. These findings suggest that reducing SA may warrant industry-specific interventions.