A ten-society guideline on tapering benzodiazepines, and data on who is starting them

Benzodiazepines are among the few drug classes where stopping badly is more dangerous than continuing. That asymmetry shapes a joint clinical practice guideline published in the Journal of General Internal Medicine on 17 June, and it explains why the document's central recommendation is phrased as a prohibition.

Who wrote it, and how

The American Society of Addiction Medicine partnered with nine other medical societies and professional associations, representing a range of clinical settings and patient populations, to produce guidance on evidence-based strategies for tapering benzodiazepine medication [s1].

The guideline was developed using modified GRADE methodology together with a clinical consensus process, incorporating a systematic literature review and several targeted supplemental searches, and was revised following external stakeholder review [s1].

The phrase "clinical consensus process" alongside GRADE is a signal about the evidence base rather than about the authors. Where randomised evidence on tapering schedules is sparse — and for benzodiazepines it is — guidelines fall back on structured expert agreement. The recommendations should be read as the considered position of ten organisations, not as a synthesis of trial results.

The recommendations

The published key takeaways are five [s1]:

Clinicians should engage in ongoing risk-benefit assessment of benzodiazepine use and of tapering. Clinicians should use shared decision-making in collaboration with patients. Clinicians should not discontinue benzodiazepines abruptly in patients who are likely to be physically dependent and at risk of withdrawal. Clinicians should tailor tapering strategies to each patient and adjust based on patient response. And clinicians should offer adjunctive psychosocial interventions to support successful tapering [s1].

The third is the one that carries the most weight. Abrupt benzodiazepine discontinuation in a physically dependent person can produce a withdrawal syndrome that is medically serious, and the guideline's framing — a "should not" rather than a "should" — reflects that the harm being guarded against comes from stopping, not from starting.

The title's qualifier matters too: "considerations when risks outweigh benefits" [s1]. The document is about how to taper when a decision to taper has been made. It is not an instruction to taper everyone.

Nothing in this article is guidance for any individual. Benzodiazepine tapering is a clinical decision made with a prescriber, and the guideline itself is directed at clinicians.

Who is being started on these drugs

A longitudinal study published in The Lancet Regional Health - Western Pacific on 10 June supplies the other half of the picture: not how people come off these drugs, but how many are going on them [s2].

Using territory-wide electronic health record data from Hong Kong covering 2014 to 2023, the authors analysed prevalence, incidence and duration of benzodiazepine and Z-drug prescriptions in adults, with long-term use defined as prescriptions exceeding 90 days [s2]. Joinpoint regression assessed trend changes across four age groups: 18-25, 26-49, 50-64 and 65 and over [s2].

The number of patients prescribed these drugs rose across the decade, with an average annual percentage change of 3.44 (95% CI 3.26-3.61) in prevalence and 1.51 (0.64-2.45) in incidence [s2].

The age breakdown is where the finding sits. The sharpest increases were in adults aged 18-25 — prevalence AAPC 9.43 (8.36-10.51) and incidence AAPC 7.56 (6.19-8.89) — while incidence in those aged 65 and over declined after 2019, though it remained the highest of any group [s2].

Long-term prescribing rose consistently, and again most steeply in young adults: an AAPC of 13.43 (11.98-14.62) for benzodiazepines and 12.88 (7.85-18.24) for Z-drugs [s2]. Depression and dementia were the most common psychiatric diagnoses recorded within 180 days before and after treatment initiation [s2].

The authors conclude that long-term prescribing practices need review and that clear guidelines for safe use are needed, particularly among young adults [s2].

What the two documents do not settle together

They describe different health systems and cannot be joined into a single claim. The guideline is a US-led, ten-society consensus document; the prescribing data are territory-wide records from Hong Kong [s1][s2]. Whether the age pattern observed in Hong Kong holds elsewhere is not addressed by either.

The prescribing study is also descriptive. It reports what was prescribed, not whether it was appropriate, and an increase in prescriptions to young adults could reflect greater recognition of treatable conditions as easily as it could reflect over-prescribing. The study does not adjudicate that, and neither can this article.

What the two share is a premise: that the number of people on these drugs long-term is large enough, and growing fast enough in some groups, that the question of how to stop safely is now a routine clinical problem rather than a specialist one.

What to watch

Whether comparable prescribing trend data emerge from other health systems with territory-wide records, and whether the age-group divergence — falling incidence in the elderly, rising in young adults — appears elsewhere [s2].

Sources

• [s1] Joint Clinical Practice Guideline on Benzodiazepine Tapering: Considerations When Risks Outweigh Benefits. Journal of General Internal Medicine, 17 June 2025. https://doi.org/10.1007/s11606-025-09499-2
• [s2] A decade of Benzodiazepine and Z-drug use in Hong Kong: a longitudinal study. The Lancet Regional Health - Western Pacific, 10 June 2025. https://doi.org/10.1016/j.lanwpc.2025.101591