OBJECTIVES: To describe the association between arthritis and early retirement (ER) in Europe and its evolution between 2004 and 2017.

METHODS: We used data from all waves of the Survey of Health, Ageing and Retirement in Europe (SHARE), with the countries which participated over the analysed period (n=11). The working sample comprised all individuals who self-report suffering from arthritis aged between 50 and 65 years old, i.e., the years before retirement age in most EU countries. We used logistic regression models to estimate the association between arthritis (including osteoarthritis and rheumatoid arthritis) and ER over time, controlling for age, gender, education level, relevant comorbidities, and using country fixed effects to proxy for context differences

RESULTS: ER was reported by 28% of arthritis patients in 2004, and 23% more recently in 2017 (-17%). This variation is lower than the one observed for the European population in general (2004: 23%; 2017: 18%. Variation: -21%), and for other relevant underlying conditions, such as stroke (-24%). In the regressions explaining ER, there was an association to arthritis in 2017 (OR: 1.21, 95%CI:1.05-1.40, p=0.011. Nr. observations=29,291), but not in 2004 (OR:1.10; 95%CI: 0.98-1.25, p=0.112. Nr. observations=15,724). The opposite was observed in some other diseases, such as stroke (2004: 1.38, 95%CI: 1.05-1.83, p=0.023; 2017: 1.15, 95%CI: 0.79-1.67, p=0.46) and diabetes (2004: 1.20, 95%CI: 1.03-1.41, p=0.021; 2017: 1.19, 95%CI: 0.96-1.47, p=0.11).

CONCLUSIONS: There was an overall decrease in ER in the European population, less intensive in arthritis patients. In fact, more recently, the association between this disabling condition and ER became statistically significant. Our evidence suggests that stronger integration policies should aim at reducing early retirement of more vulnerable workers and further advances in treatment options are needed, namely for those suffering from the most prevalent forms of arthritis, such as osteoarthritis.