Summary: A new analysis of UK Biobank data shows that people living with multiple long-term physical conditions face a significantly higher risk of developing depression. Researchers who examined health records from more than 142,000 adults found that certain combinations of chronic illnesses—especially cardiometabolic conditions such as heart disease combined with diabetes—more than doubled the likelihood of a depression diagnosis within a decade.
Women with joint or bone conditions and people with chronic lung, liver, or bowel diseases were also identified as having elevated risk. The study underlines the importance of integrated care approaches that treat physical and mental health together, rather than in isolation.
Key Facts:
- High-risk combinations: Cardiometabolic conditions, particularly heart disease paired with diabetes, are strongly linked to higher depression risk.
- Gender differences: Women with arthritis or other joint and bone problems showed especially high vulnerability.
- Care implications: Coordinated, integrated care models are needed to address both mental and physical needs in patients with multimorbidity.
Source: University of Exeter
People with multiple long-term physical health conditions have a substantially increased chance of developing depression, according to a new study.
The researchers report that specific patterns of multimorbidity—defined as living with two or more chronic conditions—are associated with considerably higher rates of subsequent depression. Combinations involving cardiometabolic disease, chronic respiratory illness, and digestive or liver conditions were among the clusters linked to the greatest risk.
The study team at the University of Edinburgh analyzed records from 142,005 UK Biobank participants who had linked primary care data and at least one baseline physical condition. Participants were aged 37 to 73 and had no history of depression at the start of the study. Using clustering methods to identify common multimorbidity profiles, the researchers then followed participants to see which groups went on to receive a depression diagnosis.
Researchers applied multiple statistical clustering approaches and selected the best-performing model to define groups of people with similar physical health profiles. They then examined how membership in these clusters related to the time until a new diagnosis of depression, using survival analysis to account for follow-up time.
One cluster characterized by a high overall burden of physical illness—without a single dominant condition but a complex mixture of problems—showed the highest subsequent risk of depression. Clusters with a clear overrepresentation of cardiometabolic conditions (for example, people with both heart disease and diabetes) were among the largest and were also linked to higher depression rates. Chronic respiratory diseases such as asthma and COPD, as well as liver and bowel conditions, were similarly associated with increased depression risk in both sexes. Notably, women with joint and bone disorders, including arthritis, experienced elevated risk, a pattern less pronounced in men.
Quantitatively, the highest-risk groups experienced roughly one case of depression for every 12 people over ten years, compared with about one case for every 25 people who had no recorded physical chronic conditions. Hazard ratios comparing cluster membership to those without physical conditions ranged across models, with several clusters showing more than double the risk.
While biological mechanisms—such as inflammation or metabolic dysregulation—may contribute to the link between multimorbidity and depression, the authors emphasize that social and systemic factors are also likely important. Living with multiple chronic illnesses can increase social isolation, reduce functional capacity, and complicate access to coordinated care, all of which may raise the risk of poor mental health.
Lead author Lauren DeLong, a PhD student at the University of Edinburgh’s School of Informatics, noted that the study reveals clear associations between physical multimorbidity and later depression and called for further research to unpack causal pathways and modifiable factors. Co-author Bruce Guthrie, Professor of General Practice, highlighted the study’s implications for clinical practice, saying health services must do more to anticipate and manage mental health needs in patients with physical illness.
Professor Mike Lewis, NIHR’s Scientific Director of Innovation, commented that harnessing large health datasets can reshape how patients are treated by revealing the combined impact of multiple conditions, rather than focusing on single diseases in isolation.
About this health and depression research news
Author: Guy Atkinson
Source: University of Exeter
Contact: Guy Atkinson – University of Exeter
Image: The image is credited to Neuroscience News
Original Research: Open access. “Cluster and survival analysis of UK Biobank data reveals associations between physical multimorbidity clusters and subsequent depression” by Lauren DeLong et al. Nature Communications Medicine. DOI: 10.1038/s43856-025-00825-7
Abstract
Cluster and survival analysis of UK Biobank data reveals associations between physical multimorbidity clusters and subsequent depression
Background
Multimorbidity—the presence of two or more chronic conditions in the same individual—is an increasing challenge for health services and research. Coexisting physical and mental health problems are particularly consequential for patients’ quality of life and for health systems. This study set out to investigate whether specific patterns of physical multimorbidity are associated with a higher risk of later developing depression.
Methods
The analysis used physical morbidity data from UK Biobank participants aged 37–73. Of the total cohort, 142,005 participants had linked primary care data and at least one baseline physical condition. After stratifying by sex (77,785 women; 64,220 men), four clustering methods were evaluated and the best-performing approach was selected based on clustering performance metrics. Fisher’s Exact test identified conditions over- or under-represented in each cluster. Among participants with no prior history of depression, survival analysis estimated associations between cluster membership and time to a subsequent depression diagnosis.
Results
The k-modes clustering models performed best for these categorical multimorbidity data. Resulting clusters reflected clinically plausible groupings, including several large clusters with cardiometabolic overrepresentation (15.5% of the whole cohort; 19.7% of women; 24.2% of men). Associations between cluster membership and depression varied, with hazard ratios spanning approximately 1.29 to 2.67, and most multimorbidity clusters showing higher depression risk than participants without physical conditions.
Conclusions
The study demonstrates that certain groups of co-occurring physical conditions are linked with an increased risk of subsequent depression. These findings support the need for further research into the biological, social, and health-system factors that connect physical multimorbidity and mental health, and they argue for integrated care strategies to better support patients living with multiple chronic diseases.