The finding: This was a pragmatic multicentre RCT (n= 428) which evaluated the efficacy of interventions to reduce 10-year risk of cardiovascular disease among patients with schizophrenia and abdominal obesity. Participants were randomized 1:1:1 to either (12-months of) lifestyle modification coaching + care coordination + usual care; or care coordination + usual care; or usual care. There were no intervention effects for either coaching or care coordination compared to usual care in reducing cardiovascular risk.

Strengths: This large multicentre trial had multiple strengths: randomized controlled trial design; intervention components were manualized evidence-based interventions; design allowed the comparison of relative impact of lifestyle coaching and care coordination; primary outcome was a composite measure that integrates multiple heterogeneous risk factors.
Weaknesses: The primary outcome (10-year cardiovascular risk) is a surrogate outcome for outcomes that may have greater clinical significance (such as mortality). Only 60% of intervention participants attended at least half of the weekly meetings.

Relevance: These findings suggest that given the major barriers to health behavior change in this vulnerable population, interventions which address environmental or structural changes may be more effective than those which focus on individual health behavior change (such as lifestyle modification and care coordination).

The finding: This rigorous meta-analysis of 118 (observational and clinical trials) studies involving more than 430,000 people with serious mental illness (defined as schizophrenia, bipolar disorder or major depressive disorder) found a prevalence of type 2 diabetes (T2DM) of 11.3% (95% CI: 10.0%-12.6%). The prevalence of type 2 diabetes was higher among women with SMI than men with SMI, and the relative risk of T2DM among patients with multiple episodes of psychiatric illness was almost double that of matched general population controls. There were higher T2DM prevalences for patients taking all antipsychotic medications, except aripiprazole and amisulpride.

Strengths: This was a very large sample, and rigorous methods for meta-analysis were utilized.
Weaknesses: In most studies, diabetes diagnoses were based on chart review, which may limit the reliability and validity. Studies were based on cross-sectional, not longitudinal data, which does not allow determination of causality of association between medication use and T2DM risk. Insufficient reporting on lifestyle behaviors in the included studies precluded the evaluation of these as potential mediators.

Relevance: This meta-analysis of studies involving more than 400,000 patients with SMI provides the strongest evidence to date about the increased risk of diabetes, and the need for routine screening and integrating (or at least coordinating) diabetes treatment with mental health treatment.