Clinical Intervention Following Detection of Incident Diabetes Within a Large Psychiatric Service System: A Chart...

Individuals with mental illness have a markedly reduced life expectancy, partly due to a significantly increased risk of developing type 2 diabetes, driven by lifestyle factors and the adverse metabolic effects of psychotropic medications. This study examined whether abnormal laboratory values indicating incident diabetes, when ordered within psychiatric services, were followed by documented interventions.

We used electronic health record data from the Psychiatric Services of the Central Denmark Region and drew a random sample of patients with incident diabetes detected between 2019 and 2024. Incident diabetes was defined by the first laboratory value ordered by the psychiatric services indicating diabetes: glycated haemoglobin ≥ 48 mmol/mol, fasting plasma glucose ≥ 7.0 mmol/L, plasma glucose ≥ 11.1 mmol/L during a 2-h oral glucose tolerance test or random plasma glucose. Patients with pre-existing diabetes, identified through ICD-10 codes or antidiabetic medication history, were excluded. Manual chart review assessed whether incident diabetes was followed, within 3 months, by documented clinical intervention (such as referral, repeat testing, or medication initiation), or was merely acknowledged without further action. Descriptive statistics and adjusted logistic regression analysis were used to characterise the cohort and examine characteristics associated with lack of intervention.

Among 416 patients with incident diabetes (57% male, median age 48 years [IQR: 35-59], 33% with schizophrenia or other psychotic disorders), only 37% received a clinical intervention. The most frequent intervention was referral to a general practitioner (27%), while antidiabetic medication was initiated in only 6% of those who received an intervention. The only characteristic associated with lack of clinical intervention was body mass index < 25 kg/m² (adjusted odds ratio 3.82, 95% CI: 1.92-8.20).

This study suggests that appropriate clinical intervention following detection of incident diabetes is lacking. This may inform future strategies to reduce metabolic and cardiovascular risks among individuals with mental illness.