Cost-utility analysis of benzodiazepine deprescription in primary care: a cohort study

Long-term benzodiazepine (BZD) use is associated with adverse clinical outcomes and increased healthcare utilization. Although deprescription programs aim to reduce these risks, the net economic impact of BZD discontinuation—particularly in terms of healthcare resource use—remains unclear. This study aimed to evaluate the cost-utility of BZD deprescription intervention in primary care, assessing its impact on direct healthcare costs, service utilization, and quality-adjusted life years (QALYs). We conducted a retrospective cohort study using data from 333 patients enrolled in the BenzoStopJuntos deprescription program in Andalusia, Spain. Participants were classified according to BZD status at six-month follow-up: continuation vs. discontinuation. Healthcare resource use (primary care, hospital specialist consultations, primary care emergency, and hospital emergency visits) was extracted from routine electronic health records and monetized using standardized national tariffs (2024 euros). Health outcomes were estimated using the COOP/WONCA charts, from which utility scores and QALYs were derived. We conducted a cost-utility analysis (CUA) from the public payer perspective and calculated the incremental cost-effectiveness ratio (ICER). Probabilistic sensitivity analysis (PSA) was performed using Monte Carlo simulations. Covariate-adjusted Poisson regression models were used to analyze the association between BZD status and healthcare utilization. Subgroup analyses were conducted by comorbidity level (Charlson Comorbidity Index) and sex. Of the 333 participants, 45.6% discontinued BZD use at follow-up. The discontinuation group incurred lower mean total healthcare costs (€753.2 vs. €826.6), with a non-significant difference of –€73.4, and higher mean QALYs, resulting in an incremental QALY gain of 0.033. The ICER was –€2,185 per QALY, indicating that deprescription was a dominant strategy. Adjusted Poisson models showed a significant 45% reduction in primary care emergency visits in the discontinuation group (IRR = 0.55, 95% CI: 0.37–0.82; p = 0.003), with no significant differences in routine primary care, hospital specialist consultations, or hospital emergency visits. PSA confirmed the robustness of these findings, with the majority of simulations in the southeast quadrant of the cost-effectiveness plane. Subgroup analyses showed consistent cost-effectiveness, with greater dominance observed in participants with low to moderate comorbidity. BZD deprescription through a structured primary care intervention is a cost-effective and clinically safe strategy. It does not result in increased short-term healthcare utilization and may reduce emergency care demands. These findings support the integration of deprescription protocols into routine clinical practice and the expansion of similar interventions in public health systems. ClinicalTrials.gov NCT06209827. 01/12/21