Benzodiazepine Receptor Agonist Deprescribing Principles for Harm Minimization and Discontinuation: Modified Delphi Recommendations from a Multi-Disciplinary Expert Panel

Background Deprescribing of benzodiazepine receptor agonists (BZRA) has garnered significant attention in recent years. A popularised focus on deprescribing, which has become prominent within some guidelines, is commonly justified for BZRA by both a lack of compelling long-term effectiveness data and the neurocognitive and psychomotor hazards of these drugs in aging. With trends in clinical practice, harms may also arise from improper utilization of pharmacologic principles and the lack of patient shared decision making or recognition of withdrawal associated phenomena. Methods A modified Delphi method facilitated 3 rounds of anonymized voting among clinical BZRA experts and patient representatives to iteratively improve recommendations. An 80% agreement (‘agree’ or ‘strongly agree’ on a 5 item Likert-type scale) was required for recommendation confirmation. Recommendations which failed to reach the agreement threshold, were modified based on participant feedback and were downgraded to ‘moderate’ (> 80% agreement) or ‘weak’ consensus (50–80% agreement) based on results in subsequent rounds. Results 35 of 48 invitees participated (73% response rate) which included 7 family physicians, 9 psychiatrists, 5 pharmacists, 6 patient advocates, 2 nurse practitioners, 2 licensed clinical social workers, 2 health service policy researchers, a physician assistant and a psychotherapist. Strong consensus was achieved for attaining informed consent prior to deprescribing (recommendation 1), using a flexible and gradual tapering approach (recommendation 2) characterised by patient shared-decision making (recommendation 3) with hyperbolic dose reductions or micro-tapering (recommendation 4) facilitated via novel preparation techniques or compounded pharmaceutical formulations when possible and safe (recommendation 7). Reversion to previous doses may also occur if necessary to reduce the incidence of withdrawal (recommendation 6). Strong consensus was also reached for the potential value of adjunctive psychosocial interventions (recommendation 8) and/or peer-support resources (recommendation 9). Moderate and weak consensus was achieved, respectively, for the utility of step-wise conversion to a longer-acting BZRA (recommendation 5) or the use of an adjunctive non-BZRA pharmacotherapy (recommendation 10). Conclusions This consensus guidance document for primary care providers, mental health clinicians and long-term users of BZRA outlines ten principles/recommendations intended for improving deprescribing outcomes with an emphasis on minimizing acute and protracted withdrawal risk.