English healthcare study identifies nine distinct groups among highest-cost patients

An observational study in BMC Medicine examined primary and secondary care costs for 10,119,490 adults in England during 2018/19, representing approximately 23% of the adult population. Researchers linked Clinical Practice Research Datalink records with Hospital Episode Statistics and applied unsupervised machine learning to patients in the top 1% of total care costs, using diagnoses, age, sex, deprivation and ethnicity.

These high-cost patients accounted for 26.8% of healthcare expenditure in the study population. Their mean annual costs were £17,485, compared with £653 across the overall population, in 2018 prices. Hierarchical clustering performed best and identified nine groups, with mean ages ranging from 56 to 79 years and mean annual costs from £15,792 to £19,107.

Recurring profiles across clustering methods and subgroup analyses included younger people with liver disease and mental health conditions, patients with nodal metastases, patients with prostate cancer and nodal metastases, and older people with cardiovascular disease and dementia. These findings highlight the heterogeneity of patients with high healthcare expenditure.

Cluster-based population management could inform case management and resource allocation, but this analysis did not test whether such targeting improves outcomes or reduces costs. The findings reflect 2018/19 care patterns. Funding included the Health Foundation and NIHR; the authors declared no competing interests.