Disparities in diabetes treatment and monitoring for people with and without mental disorders: a systematic review and meta-analysis

Abstract

Background: people with mental disorders have an increased risk of diabetes, yet conflicting evidence exists regarding the quality of diabetes care. To address this evidence gap, we did a systematic review and meta-analysis to assess and compare diabetes quality of care (QOC) in people with diabetes with versus without mental disorders.

Methods: we conducted a PRISMA-2020/MOOSE-compliant systematic review and random-effects meta-analysis of cohort and case-control studies published in any language (Scopus/Embase/MEDLINE/PsycINFO; 08/02/2025; protocol https://osf.io/cuysp), estimating summary odds ratios (ORs) for diabetes QOC indicators in individuals with any versus without mental disorders, in order to investigate the association between the presence of a mental disorder and diabetes QOC indicators, including overall diabetes monitoring and treatment. Studies were excluded for which it was not possible to generate pooled quantitative data. We screened the study titles and abstracts identified, and we extracted data from published studies after full-text review. If full data were not available, we requested data from study authors twice. The primary outcome was a binary composite measure of diabetes QOC, i.e., the percentage of people receiving any diabetes monitoring as well as treatment. Secondary outcomes were study-specific diabetes QOC indicators which were matched to the nine NICE diabetes monitoring indicators (i.e., urine albumin-creatinine ratio (uACR) test, HbA1c test, blood pressure measured, foot surveillance, serum creatinine test, serum cholesterol test, BMI recorded, smoking status recorded, retinal monitoring) and specific diabetes interventions as well as antidiabetes medications. We analysed primary and secondary outcomes according to any mental disorder and to specific diagnostic (sub-)groups. Study quality was evaluated using the Newcastle-Ottawa-Scale (NOS).

Findings: data from 49 studies (42 cohort and 7 case-control) were included, comprising 5,503,712 individuals with diabetes, of whom 838,366 (15.2%) had a diagnosed mental disorder (defined using ICD-9 or ICD-10 criteria in 40 studies). The sample included 1,956,506 (of 4250666 = 46.0%) females and 2,294,160 (54.0%) males since sex was reported in only 35/49 studies. The mean age was 61.4 years (SD = 8.7; range = 47–82 years), and 2,020,253 (93%) participants had type 2 diabetes since the type of diabetes was only reported in 22 out of 49 studies. Overall, studies reported on various mental disorders (k=38), on mood disorders spectrum (k=21), major depressive disorder (k=21), schizophrenia (k=20), bipolar disorder (k=11), substance use disorder spectrum including alcohol use disorder (k=11), dementia (k=6), anxiety disorder spectrum (k=5) and personality disorder spectrum (k=1). Most studies were high-quality (84% scored ≥7 on the NOS, median NOS score=8, IQR 7-9) and spanned all continents, except Africa and South America. 19 studies provided adjusted estimates. Data were collected between 1990 and 2020.

Statistically significant negative associations were observed between mental disorder status and the likelihood of receiving any recommended diabetes monitoring (k=28, OR=0.81, 95% CI=0.70–0.94, p=0.004), HbA1c measurements (k=24, OR=0.81, 95% CI=0.68–0.97, p=0.024), retinal screening (k=21, OR=0.77, 95% CI=0.63–0.95, p=0.013), lipid/cholesterol measurements (k=20, OR=0.83, 95% CI=0.69–0.99, p=0.043), foot investigations (k=11, OR=0.85, 95% CI=0.76–0.95, p=0.004), and renal investigations (k=16, OR=0.78, 95% CI=0.63–0.96, p=0.022). GLP-1 agonist prescriptions were negatively, while insulin prescriptions were positively associated with certain mental disorders, suggesting disparities in access to novel treatments. There was a positive association between comorbid mental disorders in people with diabetes and physical healthcare utilisation (k=17, OR=1.59, 95% CI=1.30-1.94, p<0.001). Additionally, male sex was negatively associated with receiving diabetes treatment. Sensitivity analyses confirmed findings when excluding low-quality, unadjusted, or veteran-only studies. There was no evidence of publication bias, and disparities did not change over time for the period from 1990 to 2020.

Interpretation: mental disorders are negatively associated with receiving adequate diabetes monitoring and GLP-1 agonist therapy. Addressing these disparities has the potential to address the increased mortality associated with mental disorders.

Funding: none.

More information

Identifiers

ORCID for Samuele Cortese: orcid.org/0000-0001-5877-8075

Catalogue record

Export record

Altmetrics