Life-course patterning of MLTC clusters and their patient-centred burden in depression: a population-based study using real-world data.

backgroundUnderstanding how multiple long-term conditions (MLTCs) develop and cluster across the life-course remains a key public health priority. Evidence on how these patterns relate to patient-centred outcomes among patients with mental health disorders remains limited.

objectiveTo evaluate patient-centred outcomes in patients with a history of depression and MLTC.

methodsA retrospective matched cohort study design was implemented in the Clinical Practice Research Datalink, a UK primary care database. Patients aged ≥10 years with a first recorded diagnosis of depression between 1 January 2003 and 31 December 2023 were matched (1:1) to a comparator without a depression history on age, sex and practice. MLTC was defined using 256 long-term conditions. Clusters of MLTC were derived for each life decade via multiple correspondence analysis and k-means clustering. Patient-centred burden was assessed across eight patient-prioritised 'work themes': symptoms, emotions, investigation and monitoring, health service and administration, learning and adapting, medication, finance and accumulation and complexity. Poisson and logistic regression models assessed associations between decade-specific MLTC clusters and burden indicators.

findingsAmong 1 962 757 patients with incident depression, 1 563 065 (78%) presented with MLTC over a mean follow-up of 14 years (SD=6), compared with 987 138 (50%) of comparators (mean follow-up=10, SD=7). Patients with depression history had a higher median number of LTCs (5 (IQR 3-9) vs 4 (3-8)). Clusters of MLTCs varied by life decade: atopic triad and liver clusters predominated in early life, metabolic and cardiovascular (CVD) clusters in midlife and musculoskeletal (MSK) and respiratory in later life. The burden patterning varied between cases and their comparators, who showed lower overall burden and distinct early-life patterns. The strongest prognostic cluster for increased burden in patients with depression changed across life. In early life, the metabolic cluster had the highest risk (incidence rate ratio=1.02 (95% CI 1.01 to 1.03)), followed by respiratory and CVD during midlife and MSK/functional in later life.

conclusionsMLTC in patients with depression is common, shows dynamic temporal patterning and leads to high patient-centred burden from adolescence onwards. CLINICAL IMPLICATIONS: The study identified specific burdensome clusters of MLTC at each life decade which could inform MLTC prevention for patients with depression.