Patient-centric characterization of multimorbidity trajectories in patients with severe mental illnesses: A temporal bipartite network modeling approach

Overview

People with severe mental illness have a lower life expectancy and a higher risk of physical conditions. To improve how these comorbidities can be detected and predicted, researchers have used linked health records to determine how they appear over time and relate to each other.

The challenge

People with severe mental illness, such as schizophrenia or bipolar disorder, have a 10-20 year shorter life expectancy. Most of these early deaths are due to physical comorbidities, like cardiovascular disease, diabetes and smoking-related lung disease. Understanding how these conditions arise and contribute to life expectancy is therefore a vital part of improving care for this population.

Previous research in this area has focused on how often people are diagnosed with other conditions in various settings, such as locations or time periods. There are only a few studies that have aimed to understand how conditions appear over the lifetime of a person with severe mental illness, partly because of the lack of long-term healthcare data.

The solution

Researchers supported by HDR UK used electronic health records from the South London and Maudsley NHS Foundation Trust, one of the largest secondary mental health care providers in the UK, to identify over 9,000 patients with severe mental illness. These records were linked to an English national hospital database to collect details of all admitted care of these people.

Using an approach called temporal bipartite networks, the team mapped out the relationships between individual patients and diagnoses of comorbidities over time. This helps to identify differences between subgroups of people that may be otherwise missed when looking at an average of a large group. This analysis also allowed them to look at the relationship between the comorbidities rather than studying them individually.

Impact and outcomes

The patterns identified showed that patients with few comorbidities tend to have more common diseases. This suggests that prevention strategies targeting these common issues could have a big impact on the well-being of people with severe mental illness.

Demographics and medical history were identified as strong influencing factors, and people with similar attributes tended to be diagnosed with the same conditions. The researchers plan to extend this approach to predict a person’s risk of disease.

Dr Tao Wang, the first author of the paper published in the Journal of Biomedical Informatics, said:

“To effectively prevent these physical comorbidities and improve health for this population, a key prerequisite is to understand how and when different diseases occur during the course of severe mental illness. Previous studies on multimorbidity either focus on estimating the prevalence of a disease in a population without considering relationships between diseases or ignore the heterogeneity of individual patients in examining disease progression by looking merely at aggregates across a whole cohort.”

Impact committee

The Impact committee considered that this work was world-leading in originality, significance and rigour. They particularly noted that this research would help improve the healthcare of an underserved group.