Abstract
Objectives To identify, describe and validate clusters of patients based on their multimorbidity, to allow better design of health services and highlight groups that may require tailored interventions.
Design Retrospective cohort study. Patients with multimorbidity were stratified by four age strata and clustered using latent class analysis. Associations between multimorbidity clusters, demographics and outcomes were quantified using generalised linear models.
Setting 382 general practices in England contributed primary care health record data to the Clinical Practice Research Datalink (CPRD).
Participants All multimorbid adults (18 years old or more, with two or more long-term conditions) whose diagnoses are defined in 2012 (N=113,211), from a random sample of CPRD-GOLD (N=391,669). Cluster identification used a random set of 80% of the multimorbid patients (N=90,571), with consistency of results checked in the remaining 20% of multimorbid patients (N=22,640).
Main outcome measures NHS service utilisation was measured by three variables: primary care consultations, hospitalisations and repeat prescriptions in one year after January 2012. All-cause mortality was recorded within two and five years.
Results Clinically distinct and meaningful clusters were identified using robust latent class analysis for 38 long-term conditions within each age strata. Associated patient profiles and outcomes were derived. “Physical-mental health co-morbidity” and “Respiratory disease and multimorbidity” clusters were common across all age strata. In under 65 year olds, “Substance misuse and mental illness co-morbidity” cluster (<7% prevalence, consisted mostly of male, current smokers and patients from derived areas) was found to have the highest demographic-adjusted 5-year mortality rate (in the 45-64 age strata, adjusted odds ratio =1.08 (95% CI 1.07 to 1.10), p<0.01) despite low health service utilisation.
Cardiovascular-related clusters were prevalent in over 65 year olds where patients in the cluster “Chronic pain, cardiovascular disease and mental illness” (65-84 age strata) had the highest primary care consultations in one year (median=23, interquartile range [14-35]) and 5-year mortality (39%). In the 85+ age strata, patients in the “Low service use multimorbidity” cluster had the lowest number of morbidities (median=3 [2-4]), service use and mortality. Consistency of results across identification and validation data was confirmed.
Conclusions Across age strata, a clear distinction between morbidity clusters was uncovered, both in the prevalence of long-term conditions within them, and in their associations with outcomes (service use and mortality). Interventions and policies to improve the care of multimorbid patients may be more effective when targeted on the distinct clusters of multimorbidity we have highlighted.
What is already known on this topic
Multimorbidity is common and is associated with greater healthcare service utilisation and treatment burden.
Multimorbidity is prevalent in the older population, and in areas with greater socioeconomic deprivation.
Three main multimorbidity groups have been identified in the literature, involving: (1) cardiovascular-metabolic conditions, (2) mental health related conditions, and (3) musculoskeletal disorders.
What this study adds
Distinct, validated and clinically meaningful clusters of multimorbid patients have been identified using routine primary care data, with patients broadly representative of the English population.
Among younger and middle-age individuals, those with substance misuse, alcohol problems, chronic pain and depression (low prevalence) had the highest mortality but relatively low service utilisation.
Over half of the oldest old multimorbid patients can be grouped in a cluster with relatively low service use.
Competing Interest Statement
The authors have declared no competing interest.
Funding Statement
SJK and YZ are supported by SJK?s MRC Career Development Award (MR/P021573/1).
Author Declarations
All relevant ethical guidelines have been followed and any necessary IRB and/or ethics committee approvals have been obtained.
Yes
All necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived.
NA
Any clinical trials involved have been registered with an ICMJE-approved registry such as ClinicalTrials.gov and the trial ID is included in the manuscript.
NA
I have followed all appropriate research reporting guidelines and uploaded the relevant Equator, ICMJE or other checklist(s) as supplementary files, if applicable.
Yes
Footnotes
Competing interests: All authors have completed the Unified Competing Interest form(available on request from the corresponding author) and declare: SJK reports grants from Medical Research Council, during the conduct of the study.
Transparency declaration: The lead author affirms that the manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned have been explained (see protocol and amendments in supplemental file 1).
Ethical approval: This study was approved by the CPRD ISAC, and so is covered by their ethics approval.
Funding: SJK and YZ are supported by SJK’s MRC Career Development Award (MR/P021573/1).
Data Availability
CPRD-GOLD is available to approved researchers for projects approved by the CPRD Independent Scientific Advisory Committee