Community prevalence of multimorbidity in Aotearoa New Zealand based on linked general practice and hospitalisation...

The aim of this study was to quantify the community prevalence of multimorbidity from both hospital and general practice (GP) records.

Using encrypted health identifiers, electronic health record data from 374,071 patients aged ≥18 years, enrolled in ProCare primary health organisation in January 2014, were linked to national hospitalisation records. We identified the prevalence of multimorbidity (two or more long-term conditions [LTCs]) recorded within hospitalisation data (based on 61 ICD-10-AM LTC categories from the M3 multimorbidity index) and the same conditions matched to Read codes, Read terms or free text from GP records.

Multimorbidity was identified in 31% of patients; under half of this group, 52,867 (14%), were identified from hospital records and 62,453 (17%) exclusively from GP records. Multimorbidity increased markedly with age, affecting over 60% of both sexes by 65-74 years. Under-estimation using hospitalisation data alone was greater for women, older people, Chinese and Europeans.

Multimorbidity prevalence is substantially under-estimated if relying solely on hospital discharge data, capturing under half of the total multimorbidity burden found when GP-level data are included. These data sources need to be combined to properly recognise the complexity of care and health-system resources required to meet patient needs.