Educational disparities in dementia incidence and healthcare utilization: evidence from a cohort study in Italy

While educational disparities in dementia incidence are well-known, whether and to what extent they persist beyond dementia onset is less explored. In this study, we investigated educational disparities in the risk of dementia diagnosis in administrative health records (dementia incidence) and subsequent healthcare utilization among dementia patients. We analysed the Lazio Region Longitudinal Study (Italy) from 2012 to 2022. We applied Cox regression to investigate disparities in dementia incidence and three subsequent healthcare utilization outcomes (all-cause hospitalizations, potentially preventable hospitalizations, and emergency visits). In a cohort of dementia-free 50-90-year-olds (907 453 men and 1 083 538 women), we found strong and age-patterned disparities in dementia incidence. Compared to highly-educated, the incidence in low-educated men and women was higher, especially at ages 50-64 (HR = 2.09, 95 % CI: 1.69-2.58 and HR = 2.17; 95 % CI: 1.71-2.74). In the follow-up of 27 158 men and 40 797 women incident dementia cases, low-educated had higher risk of all-cause hospitalizations (HR = 1.24; 95 % CI: 1.16-1.32 and HR = 1.18; 95 % CI: 1.09-1.27), potentially-preventable hospitalizations (HR = 1.27; 95 % CI: 1.17-1.37 and HR = 1.19; 95 % CI: 1.08-1.31) and emergency visits (HR = 1.33; 95 % CI: 1.26-1.41 and HR = 1.27; 95 % CI: 1.18-1.35). Disparities in hospitalization are reduced after adjusting for health conditions pre-existing dementia identification, less so those in emergency visits. Overall, disparities in dementia incidence persisted to a lesser extent in subsequent healthcare utilization and were mostly accounted by pre-existing health conditions.