Background Previous research has indicated that effort reward-imbalance (ERI) is independently associated with retirement and cross-sectional multimorbidity. In addition, aging populations and pension reform across western societies has led to older adults who are at increased risk of aging related disease onset and progression, participating in the labour force for longer. The objective of this study was to examine the association between multimorbid chronic health trajectories and retirement in older workers experiencing high and low effort-reward imbalance.

Methods This study used longitudinal data from the Health and Retirement Study (2006–2016) and the English Longitudinal Study of Ageing (2004–2014) for adults aged 50–55 years. Group-based trajectory modelling was used to construct multimorbidity trajectories (0–6 of diabetes, hypertension, heart disease, stroke, lung disease and cancer) over a 10-year period separately for participants reporting low and high ERI at baseline. Logistic regression analysis fully ad-justed for relevant variables examined the association between the multimorbidity trajectories and cross-sectional retirement at the final wave.

Results Mean ERI scores were higher in UK workers (low ERI = 0.75, high ERI = 1.38), compared to U.S. workers (low ERI = 0.67, high ERI = 1.32). Four trajectory classes were identified for U.S. workers with low ERI (no conditions, no-increasing, low-increasing, and high-stable). Compared to the no conditions trajectory, the high stable trajectory was associated with retirement (HR=4.50, 95%CI=2.08–9.62). Four trajectory classes (no-increasing, low-increasing, medium-stable, and high increasing) were identified for U.S. participants with high ERI. The medium-stable (HR=3.14, 95%CI = 1.19–8.29) and high-increasing (HR=4.52, 95%CI = 1.32–15.46) trajectories were associated with retirement. UK participants with high and low ERI were each classified into 3 trajectory classes respectively (no conditions, low-increasing, high-increasing), however no significant associations with retirement were observed.

Conclusion Our findings demonstrate country differences in mean ERI scores and trajectories of multimorbidities and their association with retirement. Trajectories with high intercepts indicating multimorbidity (i.e. 2+ chronic conditions at baseline) were associated with retirement in U.S. older adults only, regardless of ERI. These results imply that baseline multimorbidity status may play a more important role than ERI on retirement in those approaching the retirement age, however further research is required. Socio-economic inequalities and social policies may provide partial explanations for these findings. Public health and workplace interventions may be warranted for workers with multimorbidities.