What is meant when we say we are clustering multimorbidity?

Abstract

Clustering multimorbidity has been a global research priority of late. Existing studies usually identify these clusters using one of several popular clustering methods such as latent class analysis or hierarchical clustering and then explore various characteristics of these clusters (eg, their genetic underpinning or their sociodemographic drivers) as downstream analysis. These studies make several choices during clustering that are often not explicitly acknowledged in the literature, for example, whether they are clustering conditions or clustering individuals, and thus, lead to different clustering solutions. We observe that, in general, clustering multimorbidity might mean different things in different studies, and argue that making these choices more explicit and, more importantly, letting the downstream analysis or the purpose of identifying multimorbidity clusters guide these choices, might lead to more transparent and operationalisable multimorbidity clusters. In this Personal View, we discuss various purposes of identifying multimorbidity clusters and build a case for how different purposes can justify the different choices in data and methods.