Mapping inequalities in dementia diagnosis in England: a Bayesian spatial analysis of disease prevalence and case ascertainment

Bayesian spatial models using Local Authority District data across England show substantial geographic variation in dementia diagnosis persists after accounting for underlying dementia prevalence, with spatial effects adding residual differences.

• Why it matters: The NHS estimate that only about 65% of people with dementia receive formal diagnosis, but the geographic and socioeconomic drivers of this inequality are not well separated from true disease burden.
• What they did: The study jointly modelled dementia prevalence (using Lancet Commission risk factors) and diagnosis rates (as functions of healthcare provision and sociodemographic factors), adding spatial random effects and comparing case ascertainment with regional dementia subtype coding.
• The result: Urbanicity, deprivation, ethnicity, and healthcare provision are linked to case ascertainment, and regions with lower ascertainment tend to have higher proportions of undifferentiated dementia diagnoses, enabling targeted regional intervention planning.

AI summary of the abstract · may contain errors · not medical advice · read the paper.