Associations of socioeconomic status and healthy lifestyle with incident early-onset and late-onset dementia: a prospective cohort study

Background: Modifiable risk factor estimates are sparse for early-onset dementia incidence. This study aimed to estimate and compare the risk factor profiles of early-onset dementia and late-onset dementia, and to explore the complex relationships between socioeconomic status, lifestyles, and early-onset dementia risk. Methods: In this prospective cohort study, we used data from the UK Biobank for analysis of early-onset dementia and late-onset dementia. For early-onset dementia analyses, data were collected on those aged younger than 60 years without prevalent dementia at baseline. For late-onset dementia analyses, data were collected on those aged 65 years or older at the end of follow-up. Participants with missing information on socioeconomic factors were excluded. Two models were used to test associations between early-onset dementia incidence and socioeconomic status. The first model tested associations between socioeconomic status and early-onset and late-onset dementia incidence, adjusting for covariates. Participant socioeconomic status was defined using education level, income, and employment status via latent class analysis. The second model additionally included a healthy lifestyle score, which was constructed using data on smoking, alcohol consumption, physical activity, and the Healthy Diet Index. Incident early-onset dementia was defined as a dementia case diagnosed before 65 years of age. Multivariable-adjusted Cox proportional hazard regression models were used to estimate the hazard ratio (HR) for risk of dementia. We used multivariable-adjusted Cox proportional-hazard regression models to estimate the HR for risk of both early-onset dementia and late-onset dementia. Findings: Between 2007 and 2010, 257 345 individuals were included in the analysis of early-onset dementia, and 294 133 older individuals were included in the analysis of late-onset dementia. During a mean follow-up of 11·9–12·5 years, 502 early-onset dementia cases and 5768 late-onset dementia cases were documented. Risk factor profiles were typically dissimilar between early-onset dementia and late-onset dementia. For instance, the age and sex adjusted HR for low socioeconomic status (vs high) was 4·40 (95% CI 3·43–5·65) for early-onset dementia and 1·90 (1·74–2·07) for late-onset dementia, yielding a ratio of HRs of 2·32 (1·78–3·02). After adjusting for various risk factors, participants with low socioeconomic status (vs high) had increased risk for early-onset dementia (3·38, 2·61–4·37), and overall lifestyle mediated 3·2% (1·8–5·7) of the association. Individuals with both low socioeconomic status and unhealthy lifestyles had a higher risk of early-onset dementia (5·40, 3·66–7·97). No significant interaction was observed between lifestyle and socioeconomic status. The association between socioeconomic status and early-onset dementia seemed to be more pronounced in individuals with type 2 diabetes (HR 11·21, 95% CI 2·70–46·57). Interpretation: Early-onset dementia and late-onset dementia might have different risk factor profiles; although risk factors were similar, the magnitude of associations between risk factors and dementia incidence was greater for early-onset dementia. Only a small proportion of the socioeconomic inequity in dementia risk was mediated by healthy lifestyles, which indicates that measures other than healthy lifestyle promotion to improve social determinants of health are warranted. Funding: The National Key Research and Development Program of China, the National Natural Science Foundation of China, the Hubei Province Science Fund for Distinguished Young Scholars, and the Fundamental Research Funds for the Central Universities.