Primary healthcare utilisation among individuals with multimorbidity in deprived communities; a modelling study

Background Almost one in four adults in England have two or more long-term health conditions (LTCs). Patients living in deprived areas develop multimorbidity seven years earlier than those in the least deprived areas; this puts significant pressure on our healthcare system. Some conditions share similar characteristics and can commonly occur together. Aim We conducted a modelling study to cluster patients based on shared characteristics and understand the healthcare utilisation of these different multimorbidity clusters. Design & setting A modelling study using routinely collected clinical data from general practices in a highly deprived London borough (IMD quintiles 1-2). Method We analysed a large database of demographic and healthcare records. Adults (≥18 years) registered between 2018 and 2022 with at least two long-term conditions (LTCs) were included. Latent class analysis was used to identify patient clusters, adjusting for four covariates. Results 1 182 972 adults were registered with 40 general practices in the borough; 19.7% ( n =170,128) were living with ≥2 LTCs, and over 65% ( n =111,251) in the two most deprived quintiles. Ten clusters were developed and considered the most clinically appropriate. The Neuro-Psychiatric cluster was the largest, including 26.2% ( n =44,492) of patients. Over 98% ( n =23,306) of patients in the Autoimmune cluster were female, whereas 93.3% ( n =41,516) of patients in the Neuro-psychiatric cluster were male. Most multimorbid patients in the Inflammatory (84.7% ) and Mental Health (83.3% ) clusters were aged between 18-50. Most patients in the Behavioral (90.3% ) clusters were between 50-90; this cluster demonstrated the highest likelihood of healthcare utilization. Conclusion Individual-based clustering can provide an in-depth understanding of clinical profiles and healthcare utilization of multimorbid patients living in deprived regions.