Has the increased risk of cardiovascular mortality with Severe Mental Illness (SMI) compared to the general population changed over time?

Talk Code: 
Amanda Lambert
Prof Tom Marshall, Dr Helen Parretti
Author institutions: 
University of Birmingham: Institute of Applied Health Research, University of East Anglia, Norwich Medical School


Previous reviews of the association between severe mental illness (SMI) and cardiovascular disease (CVD) have mainly focussed on single diagnoses (schizophrenia, bipolar disorder (BD)) or outcomes (coronary heart disease, stroke) and either mortality or morbidity. Results have been inconsistent although SMI appears to increase risk of CVD. Our study aimed to conduct a comprehensive up-to-date systematic review of the relationship between SMI and CVD and explore changes in the relationship over time. Here we report effects on CVD mortality.


We searched for studies comparing CVD incidence or mortality in SMI compared to controls in a systematic search of MEDLINE, EMBASE, PsycINFO, and CINAHL, Web of Science and ZETOC conference abstracts and Cochrane CENTRAL database. We used SMI and CVD search terms, without language or date restriction. Bibliographies of included studies were also searched. Two independent reviewers screened titles and abstracts for initial selection, then reviewed full texts of selected studies and decided on inclusion. Data on risk of CVD for SMI versus controls from included studies was extracted. Effects were pooled using random effects meta-analysis and sources of heterogeneity between studies explored with meta-regression.


Reviewers screened 9591 titles and abstracts, initially selected 303 full texts and included 58 studies in quantitative analysis of CVD mortality. CVD mortality risk was increased for both schizophrenia (SMR: 1.96, 95%CI: 1.72-2.23) and bipolar disorder (SMR: 1.65, 95%CI: 1.53, 1.78). CVD mortality risk for schizophrenia was higher than for BD (RR: 1.19, 95%CI: 1.04-1.36). Results were similar for CHD and stroke mortality when considered separately. However, there was significant heterogeneity between studies, partly explained by an increased risk in studies conducted since the 1990s.More recent studies reported larger associations between SMI and cardiovascular mortality than earlier studies: 1990s vs 1950s-70s, RR: 1.44 (95% CI: 1.13-1.83); 2000s vs 1950s-70s, RR: 1.30 (95% CI: 1.05-1.62); 2010s vs 1950s-70s, RR: 2.41 (95% CI: 1.41-4.13).


SMI is associated with a significantly increased risk of CVD-related mortality which appears to be increasing over time. Although improved physical health for people with mental health conditions is reflected in UK health policy, more needs to be done to understand the reasons for the higher risk in CVD, to identify people with SMI at risk of dying from CVD and to put appropriate interventions in place to reduce the risk.

Submitted by: 
Amanda Lambert
Funding acknowledgement: 
This systematic review is conducted as part of a PhD funded by the National Institute of Health Research (NIHR).