Pos0569 identifying the impacts of arthritis on labour market participation: an observational study in britain

Background Existing evidence of the impacts of rheumatic and musculoskeletal diseases (RMDs), including those with arthritis, compared to those without RMDs on labour market participation is limited. Furthermore, most studies are cross-sectional. Objectives To compare labour market outcomes in people with arthritis versus those without. Methods We utilise 10 waves of the UK Household Longitudinal Survey (2009-2019) which contains annual observations of a representative sample of individuals in Britain. The data contain labour market participation measures (employment, unemployment, non-participation due long-term sickness, and retirement) and data for those employed including whether the respondent has been off work for health reasons, job sector type and employing sector. These are observed over time, enabling assessments of changes in activity. Arthritis was self-reported (type of arthritis is not identifiable). Multivariable linear and non-linear probability regression models were estimated for: i) sickness absence (employed individuals), ii) transitions to unemployment (employed individuals), iii) transitions to long-term sick (employed, unemployed individuals), and iv) transitions to retirement (employed, unemployed, long-term sick individuals). The sample concerns those aged 30-64. The key covariate of interest is an indicator for whether the respondent has reported arthritis. The panel nature of the data permits accounting for unobserved heterogeneity (fixed-effects). A range of potentially confounding factors were adjusted for including: age (40-49, 50-59, 60-64, ref: 30-39), job sector (intermediate, routine, ref: managerial), employing sector (public sector, ref: private sector), year indicators, and the presence of other long-term conditions. Additional analyses interacted arthritis with age, job sector and employing sector, and other long-term conditions to test whether the impact of arthritis differs between these groups. Results The results were robust to linear or non-linear probability model specifications. Sample sizes representing individual-year observations for the sickness absence, and transitions to long-term sick, unemployment, and retirement were 166,878, 167,681, 166,994, and 169,811 respectively. 1.06% reported sickness absence and 8.31% reported arthritis. Those with arthritis had a greater probability of sickness absence than those without arthritis (1.35 percentage points (pp), p-value <0.001). The effects of arthritis did not vary by age, job sector, public sector employment, or for those with other long-term conditions. 0.04% transition to unemployment and 8.31% reported arthritis. Those with arthritis were no more likely to become unemployed (0.01 pp, p-value=0.148). The effects of arthritis was no different by age, employing sector, job sector, or for those with other long-term conditions. 0.36% transition to long-term sick and 8.38% reported arthritis. Those with arthritis were more likely to move into long-term sick (0.64 pp, p-value<0.001). The effects of arthritis did not differ by age or by whether the individual’s employing sector was public nor by job sector; but was greater for those with other long-term conditions (8.92 pp, p=0.030). 1.22% transition to retirement over the period and 8.56% reported arthritis. Those with arthritis were more likely to retire in the following year (1.43 pp, p-value <0.001). The effects of arthritis was greater with age (40-49 (0.51 pp, p<0.001), 50-59 (1.90 pp, p=0.004) and 60-64 (4.69 pp, p<0.001)); was less for those in routine job role sectors (-1.16 pp, p=0.052), but not different for those in public sector jobs or for those with other long-term conditions. Conclusion The findings provide evidence of the detrimental impacts arthritis has on labour market participation and highlights those groups of the population for which the impacts of arthritis are greater, providing insight into where targeted support may be required by employers and policymakers to support labour market participation for those with arthritis. Acknowledgements This work is undertaken as part of the Centre for Musculoskeletal Health and Work, funded by Versus Arthritis and the Medical Research Council. Disclosure of Interests None Declared.