Antidepressant and antipsychotic drug prescribing in people with type 2 diabetes

Abstract

BACKGROUND: Prescribing of antidepressant and antipsychotic drugs is increasing worldwide for a range of indications including mental illness. Prescribing of these drugs among people with type 2 diabetes (T2D) is higher relative to the general population, due to the higher prevalence of mental illness in people with diabetes as well as the use of some types of antidepressants in treating diabetic neuropathy. In addition, there is concern that the prescribing of antidepressant and antipsychotic drugs may lead to poorer outcomes in people with diabetes. However, the literature on this topic is limited.

METHODS: This project was completed in three stages: (1) repeated annual cross-sectional analyses describing prevalence of antidepressant and antipsychotic prescribing, overall and by sociodemographic factors and drug subtype, in people with T2D in Scotland from 2004 to 2021; (2) a systematic review of observational studies examining the association between antidepressant or antipsychotic drugs and mortality, vascular morbidity, and cardiometabolic risk factors in people with T2D, and; (3) a retrospective cohort study using Cox proportional hazards models to estimate hazard ratios (HRs) for each of all-cause mortality, cardiovascular disease (CVD)-specific mortality, and major cardiovascular events (MCEs, comprising stroke and myocardial infarction) in people diagnosed with T2D in Scotland between 2006 and 2021, prescribed each of antidepressant and antipsychotic drugs in the five years prior to diagnosis of diabetes relative to people without the prescribing of interest. Fully adjusted models included covariates for age, sex, ethnicity, deprivation, smoking status, comorbidities, history of CVD, history of alcohol use disorders, hospital admission diagnosis of mental illness, and other psychotropic drug prescribing. Stages (1) and (3) were conducted using a population-based diabetes registry with 99% coverage of people with a diagnosis of diabetes in Scotland.

RESULTS: Between 2004 and 2021, the number of people with a diagnosis of T2D in Scotland nearly doubled, from 161,915 to 309,288. Age-sex standardised annual prescribing prevalence of antidepressants increased from 20.0 per 100 person-years (95% CI 19.8-20.2) in 2004 to 33.3 per 100 person-years (95% CI 33.1-33.6) in 2021, while prescribing prevalence of antipsychotics increased from 2.8 per 100 person-years (95% CI 2.7-2.8) in 2004 to 4.7 per 100 person-years (95% CI 4.6-4.8) in 2021. This trend was consistent across most drug subtypes except first-generation antipsychotics, prescribing of which remained largely stable. Prescribing patterns also differed by age, sex, and socioeconomic status. The systematic review included 18 studies, 14 reporting on antidepressants and four on antipsychotics. While some evidence suggested that antidepressant prescribing may be associated with increased risk of CVD morbidity, and that both antidepressant and antipsychotic prescribing may impair glycaemic control, findings were mixed, and with significant methodological limitations. The retrospective cohort study of 240,657 people with T2D showed that antidepressant prescribing prior to diabetes diagnosis was associated with statistically significant or marginally statistically significant higher risk of each of allcause mortality, CVD-specific mortality, and MCEs in crude or partially adjusted models. However, these associations were not significant in fully adjusted models. By contrast, antipsychotic prescribing was associated with significantly higher risks of all-cause mortality (HR 1.31, 95% CI 1.24-1.37) and CVD-specific mortality (HR 1.19, 95% CI 1.09-1.30), but a lower risk of MCEs (HR 0.89, 95% CI 0.83-0.97) in fully adjusted models. Sensitivity analyses comparing second-generation antipsychotic prescribing with no antipsychotic prescribing yielded results consistent with the primary analyses.

CONCLUSION: Increased prevalence of antidepressant and antipsychotic prescribing among people with T2D in Scotland between 2004 and 2021 potentially reflects the growing burden of multimorbidity in this population. Though the association between antidepressant prescribing prior to diabetes diagnosis and higher risk of each of all-cause mortality, CVDspecific mortality, and MCEs was not significant in fully adjusted models, the broader evidence base is mixed, and questions remain about potential long-term risks. Findings that antipsychotic prescribing prior to diabetes diagnosis was associated with significantly higher risk of all-cause and CVD-specific mortality raise concerns about the safety of these medications in people with T2D. Given the substantial public health burden of T2D and its associated complications, and the increasingly widespread use of antidepressant and antipsychotic drugs for a range of indications, it remains a priority to further advance our understanding of the prescribing of antidepressant and antipsychotic drugs in people with diabetes.