• EPIDEMIOLOGY
• DEMENTIA
• METHODS

Along with population ageing and the continuing extension of life expectancy, multimorbidity—the coexistence of two or more health conditions—is becoming more commonplace in older populations.1 Major consequences of multimorbidity include functional impairment, disability, poor quality of life and high healthcare costs,2 which bring heavy burden to the family and social care systems. The prevalence of multimorbidity in the population aged 65 and older in England is projected to rise considerably from 54% in 2015 to 68% in 2035.3 Although a number of studies have examined the phenomenon of increasing multimorbidity by counting the number of health conditions an individual has, more and more researchers have recognised that it is also important to advance our understanding of the age of onset, clustering and sequence of morbidities.4 As a result, when investigating health conditions as risk factors for dementia, it is necessary to consider how multiple health conditions cluster and whether the risk of dementia varies between different clusters.5

The paper by Khomdoker et al 6 applied latent class analysis (LCA) to extract four multimorbidity clusters using the baseline data of the UK Biobank, and further examined the association between the multimorbidity clusters and the incidence of dementia during follow-up. The authors showed that, compared with individuals with no multimorbidity, those grouped into the ‘mental health’ and ‘cardiometabolic conditions’ clusters had the highest risk of dementia, …