Abstract
INTRODUCTION: Insomnia is a recognised risk factor for type 2 diabetes (T2DM). Although benzodiazepines (BDZs) are commonly prescribed for the treatment of insomnia, evidence regarding their use and the onset of T2DM among individuals with insomnia remains limited. This study aimed to assess the association between BDZ use and incident T2DM in a population with insomnia.
METHODS: Within the UK Biobank prospective cohort of participants reporting insomnia, we examined the association between BDZ use and the risk of incident T2DM using multivariable adjusted Cox regression models. Robustness was evaluated through propensity score matching, inverse probability of treatment weighting, and competing risk analyses. Drug target Mendelian randomisation was conducted to explore potential causal relationships between BDZ pharmacological targets and T2DM.
RESULTS: Among 125,789 participants with insomnia, 119 of 1,524 BDZ users (7.8 per cent) and 6,833 of 124,265 non-users (5.5 per cent) developed incident T2DM over a median follow-up of 14.3 years. The primary multivariable Cox regression and all sensitivity analyses showed no association between BDZ use and T2DM risk. Drug target Mendelian randomisation further demonstrated no significant causal effect of BDZ target gene expression on T2DM development.
DISCUSSION: These findings indicate that BDZ use does not materially influence T2DM risk in individuals with insomnia, despite theoretical concerns regarding metabolic effects. The consistency across observational and genetic approaches supports the robustness of this null association and may ease concerns about long-term metabolic safety in clinical settings.
CONCLUSION: Evidence from both cohort analyses and drug target Mendelian randomisation suggests that BDZ use does not significantly increase the risk of T2DM among UK Biobank participants with insomnia.