Key PointsQuestion
Do individuals with first-episode schizophrenia already demonstrate evidence of glucose dysregulation?
Findings
In this meta-analysis of 14 case-control studies comprising 1345 participants, individuals with first-episode schizophrenia had elevated fasting plasma glucose levels, elevated plasma glucose levels after an oral glucose tolerance test, and elevated fasting plasma insulin levels, as well as greater insulin resistance compared with healthy individuals serving as controls.
Meanings
Glucose homeostasis is altered from illness onset in schizophrenia, indicating that patients are at increased risk for type 2 diabetes as a result; this finding has implications for the monitoring and treatment of patients with schizophrenia.
Importance
Schizophrenia is associated with an increased risk of type 2 diabetes. However, it is not clear whether schizophrenia confers an inherent risk for glucose dysregulation in the absence of the effects of chronic illness and long-term treatment.
Objective
To conduct a meta-analysis examining whether individuals with first-episode schizophrenia already exhibit alterations in glucose homeostasis compared with controls.
Data Sources
The EMBASE, MEDLINE, and PsycINFO databases were systematically searched for studies examining measures of glucose homeostasis in antipsychotic-naive individuals with first-episode schizophrenia compared with individuals serving as controls.
Study Selection
Case-control studies reporting on fasting plasma glucose levels, plasma glucose levels after an oral glucose tolerance test, fasting plasma insulin levels, insulin resistance, and hemoglobin A1c (HbA1c) levels in first-episode antipsychotic-naive individuals with first-episode schizophrenia compared with healthy individuals serving as controls. Two independent investigators selected the studies.
Data Extraction
Two independent investigators extracted study-level data for a random-effects meta-analysis. Standardized mean differences in fasting plasma glucose levels, plasma glucose levels after an oral glucose tolerance test, fasting plasma insulin levels, insulin resistance, and HbA1c levels were calculated. Sensitivity analyses examining the effect of body mass index, diet and exercise, race/ethnicity, and minimal (≤2 weeks) antipsychotic exposure were performed.
Data Synthesis
Of 3660 citations retrieved, 16 case-control studies comprising 15 samples met inclusion criteria. The overall sample included 731 patients and 614 controls. Fasting plasma glucose levels (Hedges g = 0.20; 95% CI, 0.02 to 0.38; P = .03), plasma glucose levels after an oral glucose tolerance test (Hedges g = 0.61; 95% CI, 0.16 to 1.05; P = .007), fasting plasma insulin levels (Hedges g = 0.41; 95% CI, 0.09 to 0.72; P = .01), and insulin resistance (homeostatic model assessment of insulin resistance) (Hedges g = 0.35; 95% CI, 0.14 to 0.55; P = .001) were all significantly elevated in patients compared with controls. However, HbA1c levels (Hedges g = −0.08; CI, −0.34 to 0.18; P = .55) were not altered in patients compared with controls.
Conclusions and Relevance
These findings show that glucose homeostasis is altered from illness onset in schizophrenia, indicating that patients are at increased risk of diabetes as a result. This finding has implications for the monitoring and treatment choice for patients with schizophrenia.
Large-scale epidemiologic studies have established that people with schizophrenia die 15 to 30 years earlier than the general population and that 60% or more of this premature mortality is due to causes not related to the central nervous system,1-5 predominantly cardiovascular.6 Rates of type 2 diabetes are estimated to be 2 to 3 times higher in schizophrenia than in the general population, with a prevalence of 10% to 15%.7,8 Although antipsychotic use may contribute to this association, a link between schizophrenia and diabetes was already observed in the 19th century, long before the introduction of antipsychotics and in an era when diets did not have such a propensity to induce metabolic derangements.9,10 For over a decade, there has been a drive to identify whether schizophrenia confers an inherent risk for the development of type 2 diabetes by investigating patients at illness onset before the potentially confounding effects of chronic illness and long-term antipsychotic treatment. Several studies have focused on the presence or absence of type 2 diabetes in patient cohorts compared with controls. The results from meta-analyses of these studies examining the prevalence of type 2 diabetes in individuals with first-episode psychosis and controls have found no significant differences between the 2 groups.11,12 However, there are 2 limitations with restricting analyses to an established diagnosis of type 2 diabetes. The first limitation is that patients may be less likely to seek medical attention, so there is the risk of underreporting. The second is that the development of type 2 diabetes takes time, with peak onset in middle age, and so may not have had time to develop in patients with first-episode schizophrenia. Type 2 diabetes shows a progression through a period of insulin resistance, elevated insulin levels, and impaired glucose tolerance (prediabetes) before the development of symptoms and a patient eventually receiving a diagnosis of type 2 diabetes. If a study’s outcome is whether criteria are met for a diagnosis of type 2 diabetes, significant alterations in glucose homeostasis between patient and control groups may be missed. In view of these limitations, we performed a meta-analysis of studies that focused on measures of glucose control in individuals either at risk for psychosis or in their first episode of psychosis. The aim of our meta-analysis was to test the hypothesis that individuals with first-episode schizophrenia exhibit alterations in glucose homeostasis compared with matched controls.
A systematic review was performed according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses)13 and MOOSE (Meta-analysis of Observational Studies in Epidemiology)14 guidelines (eTables 1 and 2 in the Supplement). Two of us (T.P. and K.B.) independently searched MEDLINE (from 1946 to week 2 of April 2016), EMBASE (from 1947 to April 25, 2016), and PsycINFO (from 1806 to week 2 of April 2016). The following key words were used: (schizophrenia or schizoaffective or psychosis or psychotic) and (early onset or first episode or at risk or ultra high risk or prodrome) and (medication or drug or antipsychotic) and (glucose or diabetes or type 2 or prediabetes or intolerance or oral glucose tolerance test or OGTT) or fasting or random or insulin or insulin resistance or hemoglobin [Hb] A1c or homeosta* or homeostatic model assessment of insulin resistance [HOMA-IR]). Studies in any language were considered, although all the included articles were published in English. The search was complemented by hand-searching of meta-analyses and review articles. Abstracts were screened and the full texts of relevant studies were retrieved. If full texts or abstracts were not available, authors were contacted and articles requested. Two of us (T.P. and K.B.) selected the final studies for review and meta-analysis.
Inclusion criteria were (1) a DSM or International Statistical Classification of Diseases and Related Health Problems diagnosis of schizophrenia, schizoaffective disorder, schizophreniform disorder, schizophrenia spectrum or psychotic disorder not otherwise specified, or an at-risk mental state for psychosis according to research criteria15,16; (2) first episode of illness (defined either as first treatment contact [inpatient or outpatient] or duration of illness up to 5 years following illness onset17); (3) antipsychotic naive or minimal exposure (≤2 weeks of antipsychotic treatment); (4) a healthy control group; (5) glucose homeostasis assessment including 1 or more fasting plasma glucose concentration, random plasma glucose concentration, oral glucose tolerance test (OGTT), percentage of hemoglobin A1 that is glycated (HbA1c), or insulin resistance as measured using the homeostatic model assessment (HOMA). The OGTT was required to meet the American Diabetes Association (ADA)18 and World Health Organization (WHO) criteria,19 namely, serum glucose concentration measured 2 hours after a 75-g oral glucose load following an overnight fast. Fasting serum glucose and insulin concentrations were defined as concentrations of either measure taken after an overnight fast in accordance with the ADA and WHO criteria. HOMA measurements of insulin resistance were required to follow either the original HOMA-IR formula20 (fasting plasma insulin (mU/L) × fasting plasma glucose (mmol/L)/22.5) or the updated HOMA2 formula21 via the University of Oxford Diabetic Trials Unit HOMA2 calculator, version 2.2 (http://www.dtu.ox.ac.uk).
Exclusion criteria were (1) studies only assessing diagnosis of type 1 or type 2 diabetes, (2) patients with multiple episodes of schizophrenia, (3) chronic antipsychotic treatment (>2 weeks’ lifetime exposure), (4) substance- or medication-induced psychotic disorder, (5) physical comorbidity that may affect glucose homeostasis (eg, prior diagnoses of type 1 or type 2 diabetes, other endocrine disorders [eg, Cushing syndrome or acromegaly], pancreatitis, congenital disorders known to increase risk of type 2 diabetes [eg, Klinefelter or Turner syndrome], and other systemic illnesses that may affect pancreatic function [eg, cystic fibrosis, hemochromatosis, or any chronic systemic inflammatory illness]), and (6) absence of measures in a healthy control group.
A small proportion of articles included patients with a limited duration of antipsychotic use (2 weeks maximum). In these cases, authors were contacted to obtain access to data concerning patients who were drug naive. If these data were not available, sensitivity analyses were performed examining only studies of patients who had no antipsychotic exposure.
WHO identifies obesity as the strongest risk factor for type 2 diabetes from evidence based on studies across 188 countries.22 In view of this observation, sensitivity analyses were performed examining studies in which patients and controls were matched on body mass index (BMI) to determine whether failure to match BMI influenced results. The matching was confirmed by either review of study methods or by confirmation of no significant difference between mean BMI levels of the patient and control groups (a 2-tailed P value <.05 was deemed significant). WHO recognizes several other risk factors for type 2 diabetes relating to BMI, including unhealthy diet and physical inactivity.22 Individuals with schizophrenia engage in significantly less physical exercise than controls, with even lower levels of physical activity observed in early stages of the illness.23 In addition, the prodrome is associated with decreased physical activity and poor eating habits.24,25 To address whether differences in diet and exercise between patient and control groups influenced the results, sensitivity analyses examining groups matched for diet and exercise were performed. Diet and exercise matching was confirmed either by review of study methods or by confirmation of no significant difference between mean diet and exercise parameters of the patient and control groups (a 2-tailed P value <.05 was deemed significant). Nonmodifiable risk factors for type 2 diabetes, such as ethnicity, are also recognized,22 and in this context, sensitivity analyses were also performed examining studies in which participants were matched for ethnic background.
For every study, data were extracted according to the following model: author, year of publication, country, design (ie, prospective, cross-sectional, case-control, and retrospective), matching criteria for patients and controls (confirmed by review of study methods or by confirmation of nonsignificance between mean parameter levels of patient and control groups; a 2-tailed P value <.05 was deemed significant), whether or not patient groups were antipsychotic naive (and if not, duration of treatment), and mean (SD) measure of glucose homeostasis in patient and control groups. If there were multiple publications for the same data set, data were extracted from the study with the largest data set. The Table demonstrates this data extraction with the exception of raw glucose homeostasis measurements (mean and SDs), which are documented in eTables 3-7 in the Supplement. The parameters of glucose homeostasis available in the studies described in the Table but not included in meta-analysis, along with the rationale behind exclusion, are documented in the eAppendix in the Supplement.
A 2-tailed P < .05 was deemed significant. A random-effects model was used in all analyses owing to an expectation of heterogeneity of data across studies. Standardized mean differences in glucose homeostasis measurements between patient and control cohorts were used as the effect size, determined using Hedges adjusted g. The 95% CI of the effect size was also calculated. The direction of the effect size was positive if individuals with schizophrenia demonstrated higher values of glucose homeostatic measurements compared with controls. Heterogeneity across studies was assessed using the Cochran Q statistic.42 Inconsistency across studies was assessed with the I2 statistic,43 with an I2 value of less than 25% deemed to have low heterogeneity; 25% to 75%, medium heterogeneity; and greater than 75%, high heterogeneity. Publication bias and selective reporting were assessed using the Egger test of the intercept44 (although this factor was not calculated when <10 studies were analyzed as recommended by the Cochrane Collaboration45) and represented diagrammatically with funnel plots, again as recommended by the Cochrane Collaboration45 (eFigures 1-5 in the Supplement). CMA, version 3.0 (Comprehensive Meta-analysis Software) was used in all analyses.
After exclusion of studies reporting on overlapping data sets, 16 case-control studies26-41 comprising 15 samples met inclusion criteria and were analyzed. The search process is demonstrated in Figure 1, and the final studies selected are summarized in the Table. The overall sample included 731 patients and 614 controls.
Fasting Plasma Glucose Concentration
Fasting plasma glucose concentration in patients and controls was analyzed using data from 14 studies comprising 718 patients and 599 controls.26-39 Fasting plasma glucose concentration was significantly elevated in patients compared with controls (Hedges g = 0.20; 95% CI, 0.02 to 0.38; P = .03) (Figure 2). There was significant between-sample heterogeneity, with an I2 value of 58.29% (Cochran Q = 31.17; P = .003). Findings of the Egger test (P = .07) suggested that publication bias was not significant. Restricting the analyses to antipsychotic-naive patients by excluding the 3 studies that included patients with up to 2 weeks of antipsychotic treatment37-39 demonstrated that fasting plasma glucose concentration remained significantly elevated in patients compared with controls (Hedges g = 0.30; 95% CI, 0.11 to 0.48; P = .002). A sensitivity analysis examining studies in which patients and controls were matched for diet and exercise parameters29,31-33,35,36 demonstrated that fasting plasma glucose concentration remained significantly elevated in patients compared with controls (Hedges g = 0.25; 95% CI, 0.07 to 0.43; P = .007) (eFigure 6 in the Supplement). However, after restricting the analyses to BMI-matched studies,26-33,37-39 there was no longer a significant difference in fasting plasma glucose concentration in patients compared with controls (Hedges g = 0.20; 95% CI, −0.03 to 0.44; P = .08). A sensitivity analysis examining studies in which patients and controls were matched for ethnicity26,31,34,36,37,39 demonstrated that fasting plasma glucose concentration remained significantly elevated in patients compared with controls (Hedges g = 0.19; 95% CI, 0.03 to 0.35; P = .02).
Plasma Glucose Concentration After OGTT
Plasma glucose concentration after OGTT was analyzed using data from 4 studies comprising 271 patients and 237 controls.34-37 Plasma glucose concentration was significantly elevated in patients compared with controls (Hedges g = 0.61; 95% CI, 0.16-1.05; P = .007) (Figure 2). Between-sample heterogeneity was significant, with an I2 value of 82.40% (Cochran Q = 17.05; P = .001). A sensitivity analysis examining studies in which patients and controls were matched for ethnicity34,36,37 demonstrated that fasting plasma glucose concentration after OGTT remained significantly elevated in patients compared with controls (Hedges g = 0.78; 95% CI, 0.40-1.17; P < .001). In the context of low study numbers, sensitivity analyses to assess the impact of BMI, antipsychotics, or diet and exercise were not performed.
Fasting Plasma Insulin Concentration
Fasting plasma insulin concentration in patients and controls was analyzed using data from 11 studies26,28,30-34,37-40 comprising 512 patients and 448 controls. Fasting plasma insulin concentration was significantly raised in patients compared with controls (Hedges g = 0.41; 95% CI, 0.09-0.72; P = .01) (Figure 3). Between-sample heterogeneity was significant, with an I2 value of 80.80% (Cochran Q = 52.09; P < .001). Findings of the Egger test (P = .12) suggested that publication bias was not significant. Excluding the 3 studies that included patients with up to 2 weeks of antipsychotic treatment37-39 to restrict the analyses to antipsychotic-naive patients demonstrated that fasting plasma insulin concentration remained significantly elevated in patients compared with controls (Hedges g = 0.47; 95% CI, 0.03-0.91; P = .04). Exclusion of 1 study that examined non–BMI-matched patients and controls34 demonstrated that fasting plasma insulin concentration remained significantly elevated in patients compared with controls (Hedges g = 0.38; 95% CI, 0.04-0.72; P = .03). A sensitivity analysis examining studies in which patients and controls were matched for ethnicity26,31,34,37,46 demonstrated that fasting insulin concentration remained significantly elevated in patients compared with controls (Hedges g = 0.49; 95% CI, 0.30-0.68; P < .001). In the context of low study numbers, a sensitivity analysis to assess the impact of diet and exercise was not performed.
Insulin resistance as measured using the HOMA-IR tool in patients and controls was analyzed using data from 10 studies26,28-32,34,38,39,41 comprising 485 patients and 400 controls. HOMA-IR was significantly raised in patients compared with controls (Hedges g = 0.35; 95% CI, 0.14-0.55; P = .001) (Figure 3). Between-sample heterogeneity was moderate but significant, with an I2 value of 55.40% (Cochran Q = 20.18; P = .02). Findings of the Egger test (P = .10) suggested that publication bias was not significant. Excluding the 2 studies that included patients with up to 2 weeks of antipsychotic treatment38,41 to restrict the analyses to antipsychotic-naive patients demonstrated that HOMA-IR remained significantly elevated in patients compared with controls (Hedges g = 0.44; 95% CI, 0.23-0.65; P < .001). Exclusion of 1 study that examined non–BMI-matched patients and controls34 demonstrated that HOMA-IR remained significantly elevated in patients compared with controls (Hedges g = 0.31; 95% CI, 0.09-0.53; P = .005). A sensitivity analysis examining studies in which patients and controls were matched for ethnicity26,31,34,39 demonstrated that HOMA-IR remained significantly elevated in patients compared with controls (Hedges g = 0.66; 95% CI, 0.43-0.88; P < .001). In the context of low study numbers, a sensitivity analysis to assess the impact of diet and exercise was not performed.
The HbA1c levels were analyzed using data from 4 studies27,34,38,41 comprising 166 patients and 164 controls. The HbA1c levels were not altered in patients compared with controls (Hedges g = −0.08; 95% CI, −0.34 to 0.18; P = .55) (eFigure 7 in the Supplement). Between-sample heterogeneity was moderate as indicated by an I2 value of 31.50%, but a Cochran Q value of 4.38 (P = .22) suggested nonsignificant heterogeneity. Of these 4 studies, 2 studies examined patients with up to 2 weeks of antipsychotic use,38,41 and 1 study examined non–BMI-matched patients and controls.34 In the context of low study numbers, sensitivity analyses were not performed.
Our main findings are that patients with schizophrenia show raised fasting plasma glucose levels, reduced glucose tolerance, raised fasting plasma insulin levels, and increased insulin resistance at illness onset. With the exception of fasting glucose levels, these alterations were also seen when analyses were restricted to antipsychotic-naive and BMI-matched samples. When analysis was restricted to diet and exercise–matched samples, significance was maintained for raised fasting glucose levels in patients. All results remained significant when analyses were restricted to samples matched for race/ethnicity. No significant differences were demonstrated in HbA1c levels, although this result should be interpreted with caution owing to the small sample size used in this analysis. The results of our meta-analysis extend recent studies showing high rates of diabetes in patients with chronic schizophrenia by demonstrating that altered glucose homeostasis is present from illness onset.
Strengths and Limitations
By focusing our analysis on patients with first-episode schizophrenia, an attempt was made to limit the duration of secondary illness-related factors known to affect glucose homeostasis. However, individuals in the prodromal state and those with first-episode schizophrenia already have poorer dietary habits, decreased physical activity, and an increased likelihood of smoking compared with age-matched controls.23-25,47 Our search did not find any studies that examined glucose homeostasis in individuals at risk for developing psychosis that matched our inclusion criteria, and the duration of untreated psychosis was documented in only 5 of the 16 studies analyzed.27,28,32,36,38 Since our definition of first-episode schizophrenia was broad, ranging from first clinical contact to duration of illness up to 5 years following illness onset,17 quantification of the duration of poor lifestyle habits for the overall sample was not possible, and the small number of studies that specifically documented duration of untreated illness prevented a meta-regression from examining the influence of chronicity of illness on glucose homeostasis. Although a sensitivity analysis examining studies in which participants were matched for diet and exercise remained significant for raised fasting glucose levels in the patient cohort, there was no significant elevation in the BMI-matched sensitivity analysis. However, the sensitivity analyses of fasting insulin levels and insulin resistance showed significant dysregulation for these variables in the patient cohort that was matched by BMI to controls. Thus, although it is a limitation that we were not able to examine their potential influences in all instances, when we were able to examine them, differences in BMI, diet, and exercise did not account for our findings with the exception of BMI for fasting glucose levels.
Although all participants used in the meta-analysis were described as physically healthy with no illnesses that would affect glucose homeostasis, only 8 studies defined use of over-the-counter and prescription medication as a specific exclusion criterion,28-31,34,36,37,41 and only 4 studies defined psychotropic use other than antipsychotics as an exclusion criterion29,30,32,33 (full details in eTable 8 in the Supplement). Thus, the potential use of medication other than antipsychotics that might disturb glucose homeostasis is a possible confounding factor in our meta-analysis. We also acknowledge that 4 of the 16 studies used in this meta-analysis analyzed patients with schizophrenia as well as individuals with schizophreniform disorder, brief psychotic disorder, and psychosis not otherwise specified,27,37,38,41 which may contribute to heterogeneity in the sample. There was also variability in matching criteria for patients and controls, which might be significant given the effect of demographic variables on risk for type 2 diabetes.22 Nevertheless, 8 studies documented that participants were matched for race/ethnicity,26,31,34,36-39,41 and our sensitivity analyses suggest that differences in ethnicities between groups were not responsible for the overarching findings of the meta-analysis. Two studies failed to match for sex,28,33 1 study failed to match for age,30 and only 8 studies documented that participants were matched for smoking status26-28,31,33,34,39,41 (Table and eTable 8 in the Supplement). Other limitations of our analyses include between-sample heterogeneity in glucose homeostasis parameters tested, including the use of either the original HOMA-IR equation20 or the HOMA2 equation.21 Nevertheless, the random-effects model that we used is robust to heterogeneity, and the fact that findings were consistent across different methods suggests that they are robust to technical variation.
In view of the findings of our meta-analysis, prospective studies investigating the effect of lifestyle factors on the glucose dysregulation seen in patients with first-episode schizophrenia would help to determine the degree to which alterations are intrinsic to schizophrenia or the consequences of emerging symptoms. Longitudinal studies examining the efficacy of early interventions targeting a reduction in diabetic risk (both lifestyle based and pharmacologic) in individuals with schizophrenia who exhibit subtle early aberrances in glucose homeostasis would be useful.
Although the findings of this meta-analysis may in part reflect poorer lifestyle habits in patients compared with controls, other mechanisms may also contribute to the link between schizophrenia and altered glucose regulation. Both schizophrenia and type 2 diabetes are associated with early developmental risk factors, such as low birth weight, preterm birth, gestational diabetes, and maternal malnutrition or obesity. The increased risk of impaired glucose homeostasis and schizophrenia in the context of early developmental insults is demonstrated by studies examining survivors from the 1944-1945 Dutch famine and the 1959-1961 Chinese famine. These studies demonstrate a relative risk of approximately 2 for developing schizophrenia in individuals conceived or in early gestation during a period of famine,48-50 as well as an increased risk of impaired glucose tolerance later in life.51 Stress and hypercortisolemia may also contribute to this association between the 2 conditions, with antipsychotic-naive individuals with first-episode psychosis exhibiting higher baseline cortisol levels and blunted cortisol wakening response compared with controls.52 There is also evidence for a shared genetic vulnerability. Relatives of individuals with schizophrenia experience higher rates of type 2 diabetes,53-55 and genome-wide association studies have revealed shared susceptibility genes between schizophrenia and type 2 diabetes.56,57 Evidence to support the existence of pleiotropy between these genes has been demonstrated by a network analysis examining common signaling pathways involved in both schizophrenia and type 2 diabetes, with identification of proteins that play a role in calcium signaling, adipocytokine signaling, Akt signaling, and γ-secretase–mediated ErbB4 signaling.57 Thus, dysfunction in common signaling pathways may drive central neurologic dysfunction as well as peripheral metabolic dysfunction.
Regardless of the mechanism, this meta-analysis has demonstrated an association between schizophrenia and early derangements in glucose homeostasis. The OGTT is a more sensitive measure of abnormalities in glucose metabolism than fasting plasma glucose level58,59 and is recognized by WHO as the only means of identifying individuals with impaired glucose tolerance. The use of fasting plasma glucose measurement alone as a screen for type 2 diabetes results in approximately 30% of type 2 diabetes cases being missed.60 Indeed, the OGTT has been recommended for screening and monitoring of patients with schizophrenia spectrum disorders owing to its increased sensitivity.61 This sensitivity lends further significance to the large effect size for raised glucose concentrations after OGTT seen in patients with schizophrenia compared with controls. Although predominantly used in research, HOMA-IR is well validated as a surrogate marker of insulin resistance, with its results correlating well with standard tests of insulin resistance, such as the hyperinsulinemic-euglycemic clamp.62 Therefore, the results from this analysis have major clinical implications. They indicate that individuals with schizophrenia present at the onset of illness with an already vulnerable phenotype for the development of type 2 diabetes. Given that several antipsychotic drugs may worsen glucose regulation,63,64 there is thus a responsibility placed on the treating clinician to select an appropriate antipsychotic at an appropriate dose so as to limit the metabolic impact of treatment. Furthermore, the association between schizophrenia and glucose dysregulation suggests that patients should be educated regarding diet and physical exercise, as well as diabetic screening, and offered early lifestyle and pharmacologic interventions to combat the risk of progression to type 2 diabetes.
Corresponding Author: Toby Pillinger, MRCP, Institute of Psychiatry, Psychology and Neuroscience, King’s College London, De Crespigny Park, London SE5 8AF, England (toby.pillinger@kcl.ac.uk).
Accepted for Publication: November 11, 2016.
Published Online: January 11, 2017. doi:10.1001/jamapsychiatry.2016.3803
Author Contributions: Dr Pillinger had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Pillinger, Beck, Donocik, Jauhar, Howes.
Acquisition, analysis, or interpretation of data: Pillinger, Beck, Gobjila, Jauhar, Howes.
Drafting of the manuscript: All authors.
Critical revision of the manuscript for important intellectual content: Pillinger, Beck, Howes.
Statistical analysis: Pillinger, Beck, Jauhar.
Administrative, technical, or material support: Pillinger, Beck, Gobjila, Donocik.
Study supervision: Pillinger, Jauhar, Howes.
Conflict of Interest Disclosures: Dr Howes has received investigator-initiated research funding from and/or participated in advisory/speaker meetings organized by AstraZeneca, Autifony, BMS, Eli Lilly, Heptares, Janssen, Lundbeck, Lyden-Delta, Otsuka, Servier, Sunovion, Rand, and Roche. No other disclosures were reported.
Funding/Support: This study was funded by grants MC-A656-5QD30 from the Medical Research Council-UK, 666 from the Maudsley Charity 094849/Z/10/Z from the Brain and Behavior Research Foundation, and Wellcome Trust (Dr Howes) and the National Institute for Health Research Biomedical Research Centre at South London and Maudsley National Health Service Foundation Trust and King’s College London.
Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Contributions: Tony Cohn, MBChB, MSc, FRCPC (Psychiatry and Nutritional Sciences, University of Toronto), provided data. There was no financial compensation.
1.Beary
M, Hodgson
R, Wildgust
HJ. A critical review of major mortality risk factors for all-cause mortality in first-episode schizophrenia: clinical and research implications.
J Psychopharmacol. 2012;26(5)(suppl):52-61.
PubMedGoogle ScholarCrossref 2.Crump
C, Winkleby
MA, Sundquist
K, Sundquist
J. Comorbidities and mortality in persons with schizophrenia: a Swedish national cohort study.
Am J Psychiatry. 2013;170(3):324-333.
PubMedGoogle ScholarCrossref 3.Hoang
U, Goldacre
MJ, Stewart
R. Avoidable mortality in people with schizophrenia or bipolar disorder in England.
Acta Psychiatr Scand. 2013;127(3):195-201.
PubMedGoogle ScholarCrossref 4.Laursen
TM, Nordentoft
M, Mortensen
PB. Excess early mortality in schizophrenia.
Annu Rev Clin Psychol. 2014;10:425-448.
PubMedGoogle ScholarCrossref 5.Brown
S, Kim
M, Mitchell
C, Inskip
H. Twenty-five year mortality of a community cohort with schizophrenia.
Br J Psychiatry. 2010;196(2):116-121.
PubMedGoogle ScholarCrossref 6.Osby
U, Correia
N, Brandt
L, Ekbom
A, Sparén
P. Mortality and causes of death in schizophrenia in Stockholm county, Sweden.
Schizophr Res. 2000;45(1-2):21-28.
PubMedGoogle ScholarCrossref 7.De Hert
M, Schreurs
V, Vancampfort
D, Van Winkel
R. Metabolic syndrome in people with schizophrenia: a review.
World Psychiatry. 2009;8(1):15-22.
PubMedGoogle ScholarCrossref 8.Mitchell
AJ, Vancampfort
D, Sweers
K, van Winkel
R, Yu
W, De Hert
M. Prevalence of metabolic syndrome and metabolic abnormalities in schizophrenia and related disorders—a systematic review and meta-analysis.
Schizophr Bull. 2013;39(2):306-318.
PubMedGoogle ScholarCrossref 9.Maudsley
H. The Pathology of Mind: A Study of Its Distempers, Deformities and Disorders. London, England: Julian Friedman Publishers; 1979.
11.Vancampfort
D, Correll
CU, Galling
B,
et al. Diabetes mellitus in people with schizophrenia, bipolar disorder and major depressive disorder: a systematic review and large scale meta-analysis.
World Psychiatry. 2016;15(2):166-174.
PubMedGoogle ScholarCrossref 12.Mitchell
AJ, Vancampfort
D, De Herdt
A, Yu
W, De Hert
M. Is the prevalence of metabolic syndrome and metabolic abnormalities increased in early schizophrenia? a comparative meta-analysis of first episode, untreated and treated patients.
Schizophr Bull. 2013;39(2):295-305.
PubMedGoogle ScholarCrossref 13.Moher
D, Liberati
A, Tetzlaff
J, Altman
DG; PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement.
J Clin Epidemiol. 2009;62(10):1006-1012.
PubMedGoogle ScholarCrossref 14.Stroup
DF, Berlin
JA, Morton
SC,
et al; Meta-analysis of Observational Studies in Epidemiology (MOOSE) Group. Meta-analysis of Observational Studies in Epidemiology: a proposal for reporting.
JAMA. 2000;283(15):2008-2012.
PubMedGoogle ScholarCrossref 15.Miller
TJ, McGlashan
TH, Woods
SW,
et al. Symptom assessment in schizophrenic prodromal states.
Psychiatr Q. 1999;70(4):273-287.
PubMedGoogle ScholarCrossref 16.Yung
AR, Yuen
HP, McGorry
PD,
et al. Mapping the onset of psychosis: the Comprehensive Assessment of At-Risk Mental States.
Aust N Z J Psychiatry. 2005;39(11-12):964-971.
PubMedGoogle ScholarCrossref 17.Breitborde
NJ, Srihari
VH, Woods
SW. Review of the operational definition for first-episode psychosis.
Early Interv Psychiatry. 2009;3(4):259-265.
PubMedGoogle ScholarCrossref 18.Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus.
Diabetes Care. 2003;26(suppl 1):S5-S20.
PubMedGoogle ScholarCrossref 19.World Health Organization. Definition, Diagnosis and Classification of Diabetes Mellitus and Its Complications: Report of a WHO Consultation; Part 1: Diagnosis and Classification of Diabetes Mellitus. Geneva, Switzerland: World Health Organization; 1999.
20.Matthews
DR, Hosker
JP, Rudenski
AS, Naylor
BA, Treacher
DF, Turner
RC. Homeostasis model assessment: insulin resistance and β-cell function from fasting plasma glucose and insulin concentrations in man.
Diabetologia. 1985;28(7):412-419.
PubMedGoogle ScholarCrossref 21.Levy
JC, Matthews
DR, Hermans
MP. Correct homeostasis model assessment (HOMA) evaluation uses the computer program.
Diabetes Care. 1998;21(12):2191-2192.
PubMedGoogle ScholarCrossref 22.Roglic
G; World Health Organization. Global Report on Diabetes. Geneva, Switzerland: World Health Organization; 2016.
23.Stubbs
B, Firth
J, Berry
A,
et al. How much physical activity do people with schizophrenia engage in? a systematic review, comparative meta-analysis and meta-regression.
Schizophr Res. 2016;176(2-3):431-440.
PubMedGoogle ScholarCrossref 24.Juutinen
J, Hakko
H, Meyer-Rochow
VB, Räsänen
P, Timonen
M; Study-70 Research Group. Body mass index (BMI) of drug-naïve psychotic adolescents based on a population of adolescent psychiatric inpatients.
Eur Psychiatry. 2008;23(7):521-526.
PubMedGoogle ScholarCrossref 25.Koivukangas
J, Tammelin
T, Kaakinen
M,
et al. Physical activity and fitness in adolescents at risk for psychosis within the Northern Finland 1986 Birth Cohort.
Schizophr Res. 2010;116(2-3):152-158.
PubMedGoogle ScholarCrossref 26.Zhang
XY, Chen
DC, Tan
YL,
et al. Glucose disturbances in first-episode drug-naïve schizophrenia: relationship to psychopathology.
Psychoneuroendocrinology. 2015;62:376-380.
PubMedGoogle ScholarCrossref 27.Petrikis
P, Tigas
S, Tzallas
AT, Papadopoulos
I, Skapinakis
P, Mavreas
V. Parameters of glucose and lipid metabolism at the fasted state in drug-naïve first-episode patients with psychosis: evidence for insulin resistance.
Psychiatry Res. 2015;229(3):901-904.
PubMedGoogle ScholarCrossref 28.Enez Darcin
A, Yalcin Cavus
S, Dilbaz
N, Kaya
H, Dogan
E. Metabolic syndrome in drug-naïve and drug-free patients with schizophrenia and in their siblings.
Schizophr Res. 2015;166(1-3):201-206.
PubMedGoogle ScholarCrossref 29.Dasgupta
A, Singh
OP, Rout
JK, Saha
T, Mandal
S. Insulin resistance and metabolic profile in antipsychotic naïve schizophrenia patients.
Prog Neuropsychopharmacol Biol Psychiatry. 2010;34(7):1202-1207.
PubMedGoogle ScholarCrossref 30.Arranz
B, Rosel
P, Ramírez
N,
et al. Insulin resistance and increased leptin concentrations in noncompliant schizophrenia patients but not in antipsychotic-naive first-episode schizophrenia patients.
J Clin Psychiatry. 2004;65(10):1335-1342.
PubMedGoogle ScholarCrossref 31.Ryan
MC, Collins
P, Thakore
JH. Impaired fasting glucose tolerance in first-episode, drug-naive patients with schizophrenia.
Am J Psychiatry. 2003;160(2):284-289.
PubMedGoogle ScholarCrossref 32.Venkatasubramanian
G, Chittiprol
S, Neelakantachar
N,
et al. Insulin and insulin-like growth factor-1 abnormalities in antipsychotic-naive schizophrenia.
Am J Psychiatry. 2007;164(10):1557-1560.
PubMedGoogle ScholarCrossref 33.Cohn
TA, Remington
G, Zipursky
RB, Azad
A, Connolly
P, Wolever
TM. Insulin resistance and adiponectin levels in drug-free patients with schizophrenia: a preliminary report.
Can J Psychiatry. 2006;51(6):382-386.
PubMedGoogle ScholarCrossref 34.Spelman
LM, Walsh
PI, Sharifi
N, Collins
P, Thakore
JH. Impaired glucose tolerance in first-episode drug-naïve patients with schizophrenia.
Diabet Med. 2007;24(5):481-485.
PubMedGoogle ScholarCrossref 35.Wani
RA, Dar
MA, Margoob
MA, Rather
YH, Haq
I, Shah
MS. Diabetes mellitus and impaired glucose tolerance in patients with schizophrenia, before and after antipsychotic treatment.
J Neurosci Rural Pract. 2015;6(1):17-22.
PubMedGoogle ScholarCrossref 36.Saddichha
S, Manjunatha
N, Ameen
S, Akhtar
S. Diabetes and schizophrenia—effect of disease or drug? results from a randomized, double-blind, controlled prospective study in first-episode schizophrenia.
Acta Psychiatr Scand. 2008;117(5):342-347.
PubMedGoogle ScholarCrossref 37.Garcia-Rizo
C, Kirkpatrick
B, Fernandez-Egea
E, Oliveira
C, Bernardo
M. Abnormal glycemic homeostasis at the onset of serious mental illnesses: a common pathway.
Psychoneuroendocrinology. 2016;67:70-75.
PubMedGoogle ScholarCrossref 38.Sengupta
S, Parrilla-Escobar
MA, Klink
R,
et al. Are metabolic indices different between drug-naïve first-episode psychosis patients and healthy controls?
Schizophr Res. 2008;102(1-3):329-336.
PubMedGoogle ScholarCrossref 39.Chen
S, Broqueres-You
D, Yang
G,
et al. Relationship between insulin resistance, dyslipidaemia and positive symptom in Chinese antipsychotic-naive first-episode patients with schizophrenia.
Psychiatry Res. 2013;210(3):825-829.
PubMedGoogle ScholarCrossref 40.Sun
HQ, Li
SX, Chen
FB,
et al. Diurnal neurobiological alterations after exposure to clozapine in first-episode schizophrenia patients.
Psychoneuroendocrinology. 2016;64:108-116.
PubMedGoogle ScholarCrossref 41.Fernandez-Egea
E, Bernardo
M, Donner
T,
et al. Metabolic profile of antipsychotic-naive individuals with non-affective psychosis.
Br J Psychiatry. 2009;194(5):434-438.
PubMedGoogle ScholarCrossref 42.Bowden
J, Tierney
JF, Copas
AJ, Burdett
S. Quantifying, displaying and accounting for heterogeneity in the meta-analysis of RCTs using standard and generalised Q statistics.
BMC Med Res Methodol. 2011;11:41.
PubMedGoogle ScholarCrossref 44.Egger
M, Davey Smith
G, Schneider
M, Minder
C. Bias in meta-analysis detected by a simple, graphical test.
BMJ. 1997;315(7109):629-634.
PubMedGoogle ScholarCrossref 45.Higgins
JPT, Green
S. Cochrane Collaboration: Cochrane Handbook for Systematic Reviews of Interventions. Chichester, England; Wiley-Blackwell; 2008.
46.Chen
DC, Du
XD, Yin
GZ,
et al. Impaired glucose tolerance in first-episode drug-naïve patients with schizophrenia: relationships with clinical phenotypes and cognitive deficits.
Psychol Med. 2016;46(15):3219-3230.
PubMedGoogle ScholarCrossref 47.Myles
N, Newall
HD, Curtis
J, Nielssen
O, Shiers
D, Large
M. Tobacco use before, at, and after first-episode psychosis: a systematic meta-analysis.
J Clin Psychiatry. 2012;73(4):468-475.
PubMedGoogle ScholarCrossref 48.St Clair
D, Xu
M, Wang
P,
et al. Rates of adult schizophrenia following prenatal exposure to the Chinese famine of 1959-1961.
JAMA. 2005;294(5):557-562.
PubMedGoogle ScholarCrossref 49.Xu
MQ, Sun
WS, Liu
BX,
et al. Prenatal malnutrition and adult schizophrenia: further evidence from the 1959-1961 Chinese famine.
Schizophr Bull. 2009;35(3):568-576.
PubMedGoogle ScholarCrossref 50.Susser
ES, Lin
SP. Schizophrenia after prenatal exposure to the Dutch Hunger Winter of 1944-1945.
Arch Gen Psychiatry. 1992;49(12):983-988.
PubMedGoogle ScholarCrossref 51.Ravelli
AC, van der Meulen
JH, Michels
RP,
et al. Glucose tolerance in adults after prenatal exposure to famine.
Lancet. 1998;351(9097):173-177.
PubMedGoogle ScholarCrossref 52.Borges
S, Gayer-Anderson
C, Mondelli
V. A systematic review of the activity of the hypothalamic-pituitary-adrenal axis in first episode psychosis.
Psychoneuroendocrinology. 2013;38(5):603-611.
PubMedGoogle ScholarCrossref 54.Foley
DL, Mackinnon
A, Morgan
VA,
et al. Common familial risk factors for schizophrenia and diabetes mellitus.
Aust N Z J Psychiatry. 2016;50(5):488-494.
PubMedGoogle ScholarCrossref 55.van Welie
H, Derks
EM, Verweij
KH, de Valk
HW, Kahn
RS, Cahn
W. The prevalence of diabetes mellitus is increased in relatives of patients with a non-affective psychotic disorder.
Schizophr Res. 2013;143(2-3):354-357.
PubMedGoogle ScholarCrossref 56.Lin
PI, Shuldiner
AR. Rethinking the genetic basis for comorbidity of schizophrenia and type 2 diabetes.
Schizophr Res. 2010;123(2-3):234-243.
PubMedGoogle ScholarCrossref 57.Liu
Y, Li
Z, Zhang
M, Deng
Y, Yi
Z, Shi
T. Exploring the pathogenetic association between schizophrenia and type 2 diabetes mellitus diseases based on pathway analysis.
BMC Med Genomics. 2013;6(suppl 1):S17.
PubMedGoogle ScholarCrossref 58.Tai
ES, Lim
SC, Tan
BY, Chew
SK, Heng
D, Tan
CE. Screening for diabetes mellitus—a two-step approach in individuals with impaired fasting glucose improves detection of those at risk of complications.
Diabet Med. 2000;17(11):771-775.
PubMedGoogle ScholarCrossref 60.World Health Organization. Definition and Diagnosis of Diabetes Mellitus and Intermediate Hyperglycaemia: Report of a WHO/IDF Consultation. Geneva, Switzerland: World Health Organization; 2006.
61.van Winkel
R, De Hert
M, Van Eyck
D,
et al. Screening for diabetes and other metabolic abnormalities in patients with schizophrenia and schizoaffective disorder: evaluation of incidence and screening methods.
J Clin Psychiatry. 2006;67(10):1493-1500.
PubMedGoogle ScholarCrossref 63.Newcomer
JW, Haupt
DW, Fucetola
R,
et al. Abnormalities in glucose regulation during antipsychotic treatment of schizophrenia.
Arch Gen Psychiatry. 2002;59(4):337-345.
PubMedGoogle ScholarCrossref 64.Howes
OD, Bhatnagar
A, Gaughran
FP, Amiel
SA, Murray
RM, Pilowsky
LS. A prospective study of impairment in glucose control caused by clozapine without changes in insulin resistance.
Am J Psychiatry. 2004;161(2):361-363.
PubMedGoogle ScholarCrossref