Characteristics and outcomes of clusters of multimorbid patients in UK general practice. (13th November 2019)

• Record Type: Journal Article
• Authors: Zhu, Y; Edwards, D; Kiddle, S; Payne, R
• Is Part Of: European journal of public health. Volume 29(2019)Supplement 4
• Publication Date: 2019-11-13
• Subjects: Epidemiology -- Europe -- Periodicals; Public health -- Europe -- Periodicals; 362.109405
• DOI: 10.1093/eurpub/ckz185.662
• Languages: English
• ISSNs: 1101-1262

Abstract

Abstract: Background: Current clinical specialities, guidelines and quality of care metrics are organised around single diseases and treatments of multiple conditions are rarely coordinated, resulting in insufficient or even conflicting care. This study uses large scale English general practice (GP) records to identify and characterise clusters of patients based on their multimorbidity to allow better design of health services and highlight groups that require additional interventions. Methods: This is a retrospective cohort study that includes multimorbid adult patients (N = 113, 211), from a random sample of 391, 669 English patients with valid GP records in 2012 where 38 long-term conditions were defined. Latent class analysis, stratified by age groups, was used to identify multimorbidity clusters. Class solutions are validated and associations between multimorbidity clusters, patient characteristics, public health service utilisation and mortality are assessed. Results: Poor socioeconomic status is associated with clusters with higher service use and mortality risk. Physical-mental health co-morbidity is a major component of multimorbidity across all age strata. The clusters with highest age-stratified mortality risk in under 65 year olds were linked to alcohol and substance misuse, whereas in over 65 year olds they were linked to cardiovascular disease. The largest cluster in the 85+ years strata (58%) has the lowest number of morbidities, a low degree of service use and mortality. Consistency was seen across identification and validation data. Conclusions: We find a clear distinction between morbidity clusters, both in the prevalence of long term conditions within them, and in their associations with outcomes (service use and mortality). Specific health services and interventions might be more effective when targeted on the distinct types of multimorbidity we have identified, with a particular focus on the morbidity clusters associated with the worst patient outcomes. Key messages: The first study to derive age stratified multimorbidity clusters from a large GP record system, whose patients are representative of the English population. Knowledge about particularly dangerous clusters of multimorbidity, such as those involving alcohol and drug use in 18–64 years old, and cardiovascular disease in those 65 years or older.