Send to

Choose Destination
Lancet Psychiatry. 2016 Nov;3(11):1049-1058. doi: 10.1016/S2215-0366(16)30262-0. Epub 2016 Oct 5.

The association between first-episode psychosis and abnormal glycaemic control: systematic review and meta-analysis.

Author information

Coventry and Warwickshire Partnership NHS Trust, Coventry, UK; Department of Mental Health and Wellbeing, University of Warwick, Coventry, UK. Electronic address:
University Hospitals Birmingham, Birmingham, UK.
Department of Mental Health and Wellbeing, University of Warwick, Coventry, UK.
Department of Mental Health and Wellbeing, University of Warwick, Coventry, UK; Birmingham and Solihull Foundation Mental Health Trust, Birmingham, UK.
Warwick Medical School, Division of Health Sciences, Statistics and Epidemiology Unit, University of Warwick, Coventry, UK.



Schizophrenia might share intrinsic inflammatory disease pathways with type 2 diabetes. We aimed to assess whether first-episode psychosis, which could be described as developing schizophrenia, is associated with prediabetic markers, or developing diabetes, to determine whether intrinsic disease links could cause the disorders to develop in unison. We hypothesised that biochemical measures of prediabetic states would be more common in antipsychotic naive patients with first-episode psychosis than in healthy matched controls.


For this systematic review and meta-analysis, using PRISMA criteria, we searched Embase, MEDLINE, PsycINFO, Web of Science, and Google Scholar for clinical studies published between database inception and Jan 6, 2016. We assessed case-control studies with biochemical assessment of prediabetic states in patients with first-episode psychosis alongside matched controls. We sought data at the summary estimate level. Several measurements were used to test for prediabetes, including fasting plasma glucose, insulin resistance (measured by the Homeostatic Model Assessment), and impaired glucose tolerance. We calculated standardised mean differences for each outcome. We used the inverse variance method, for which the weight given to each study was the inverse of the variance of the effect estimate. For dichotomous outcomes, we entered the number of events and number in each group into RevMan 5.3 and used the Mantel-Haenszel method to pool studies.


We identified 1436 studies, of which 12 were included in final analysis, including 1137 participants. Pooled analyses found first-episode psychosis to be related to insulin resistance (mean difference 0·30 [95% CI 0·18 to 0·42]), impaired glucose tolerance (mean difference 1·31 [0·37 to 2·25]), and the number of patients with impaired glucose tolerance (odds ratio 5·44 [2·63 to 11·27]), but not fasting plasma glucose (mean difference 0·03 mmol/L [-0·04 to 0·09]).


Our findings suggest a potential link between prediabetic markers, in particular impaired glucose tolerance and insulin resistance, and first-episode psychosis. However, we cannot establish causality, and the studies contributing to this review were at some risk of bias. Nevertheless, the findings might help to explain the increased prevalence of type 2 diabetes in patients with schizophrenia and could have implications for the management of patients with schizophrenia.



[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center