🟠 Moderate Evidence
A population-based cohort study of nearly 1.8 million adults in Ontario has identified a critical relationship between initial benzodiazepine prescription duration and the likelihood of long-term use, raising concerns about prescribing practices that may inadvertently perpetuate dependence. The study, published in PLOS Medicine, found that prescriptions longer than 7 days were significantly associated with reduced rates of benzodiazepine discontinuation.
Key takeaways
- Prescriptions of 8–14 days reduced discontinuation likelihood by 46%, with longer durations showing even steeper reductions
- Lorazepam accounted for 63.9% of initial prescriptions, despite guidelines favouring shorter durations and lower doses
- Median time to benzodiazepine discontinuation was 16 days for women and 19 days for men, suggesting early interventions could prevent long-term use
Study at a Glance
| Source | PLOS Medicine |
| Study type | Population-based retrospective cohort study |
| Sample size | N = 1,820,808 adults with incident benzodiazepine prescriptions |
| Population | Adults aged 18+ in Ontario, Canada with new benzodiazepine prescriptions |
| Study period | January 1, 2013 to December 31, 2021 (follow-up) |
Impact of initial prescription duration on benzodiazepine discontinuation
Adjusted Hazard Ratios comparing longer prescriptions to the referent group (≤7 days), Ontario 2013–2021
Source: Bozinoff et al., PLOS Medicine, 2024 | Georgian Medical Journal News
Shorter prescriptions linked to better long-term outcomes
The research team, led by investigators at the University of Toronto and affiliated institutions, analysed linked health administrative data from Ontario for adults with new benzodiazepine prescriptions between January 2013 and December 2020. The study followed participants through December 2021 to determine discontinuation rates. Researchers adjusted their statistical models for age, sex, anxiety, insomnia, substance use disorders, and other comorbidities to isolate the effect of initial prescription duration.
The findings present a stark dose-response relationship: adults prescribed benzodiazepines for 8–14 days had a 46% lower likelihood of discontinuation compared to those receiving 7-day prescriptions (adjusted Hazard Ratio [aHR] 0.54, 95% CI [0.54–0.54]), while those receiving prescriptions exceeding 30 days showed a 74% reduction (aHR 0.26, 95% CI [0.25–0.26]), according to the PLOS Medicine study.
Prescriptions exceeding 30 days reduced the likelihood of benzodiazepine discontinuation by 74% compared to 7-day prescriptions, suggesting that initial duration substantially shapes long-term use patterns.
— Bozinoff et al., University of Toronto (PLOS Medicine, 2024)
Lorazepam dominates prescribing despite guideline concerns
The population profile revealed important prescribing patterns that warrant scrutiny. The median age at the index prescription was 53 years (IQR 38–67), with 62.6% of patients female. Lorazepam accounted for 63.9% of initial benzodiazepine prescriptions, followed by clonazepam (17.3%) and diazepam (5.8%). This distribution is notable because UK National Institute for Health and Care Excellence (NICE) guidelines and similar international recommendations generally favour short-acting benzodiazepines prescribed for the shortest possible duration to minimise dependence risk.
The median time to benzodiazepine discontinuation was 16 days for women (IQR 6–29) and 19 days for men (IQR 6–29), according to the study. These figures suggest that the majority of patients discontinue benzodiazepines relatively quickly; however, a substantial minority continue use well beyond recommended periods. The relationship between initial prescription characteristics and long-term use suggests that clinical decisions made at the point of prescribing may have cascading effects on dependency outcomes.
Implications for prescribing practice and policy
This study aligns with growing evidence that benzodiazepine prescribing patterns significantly influence discontinuation outcomes. The World Health Organization and major international medical societies have emphasised the importance of limiting benzodiazepine use to acute anxiety, insomnia, and specific medical indications, with treatment duration typically capped at 2–4 weeks to reduce the risk of tolerance, dependence, and adverse effects including cognitive decline and increased fracture risk in older adults.
The Ontario findings suggest that prescribers may inadvertently signal a longer anticipated duration of therapy by issuing longer initial prescriptions, potentially anchoring patient expectations and clinical inertia that sustains benzodiazepine use. Secondary exposures in the study—including benzodiazepine class (short-acting versus long-acting), number of agents dispensed, and daily dose—further refined these associations, though the duration effect remained the strongest predictor of discontinuation difficulty. See our Clinical Updates section for additional guidance on benzodiazepine tapering strategies.
Research gaps and future directions
While this study provides robust population-level evidence from a jurisdiction of 14 million residents, several questions remain unanswered. The research does not establish whether shorter initial prescriptions reduce long-term use because patients genuinely require only brief therapy, or whether prescription duration serves as a psychological or behavioural signal that shapes ongoing demand and renewal patterns. Additionally, the study did not measure clinical outcomes such as symptom resolution, relapse rates, or patient satisfaction, limiting inference about whether shorter prescriptions represent optimal clinical practice or merely reflect patient or provider preference.
Future prospective studies incorporating qualitative data from patients and prescribers could illuminate the mechanisms driving these associations. Trials testing structured interventions—such as explicit time-limited prescriptions paired with patient education about benzodiazepine risks—may help establish whether this prescribing modification actually prevents long-term use and improves health outcomes. In the meantime, the Ontario cohort data provide a compelling rationale for revisiting benzodiazepine prescribing protocols to ensure alignment with best-evidence guidelines emphasising minimal duration and dose. Read more on Pharmacy & Prescribing policy updates.
What this means
Frequently asked questions
Why is long-term benzodiazepine use considered a public health concern?
Long-term benzodiazepine use is associated with cognitive impairment, increased fall and fracture risk (particularly in older adults), and potential dependence or withdrawal syndromes. Meta-analyses cited in the medical literature have linked chronic use to increased all-cause mortality. Guidelines emphasize short-term prescribing to mitigate these risks.
Does this study prove that shorter prescriptions cause better outcomes?
No. This is a retrospective observational study showing an association between initial prescription duration and time to discontinuation. It does not prove causation, and other factors (patient preference, symptom severity, provider experience) may explain both the prescription duration and discontinuation patterns. Randomised controlled trials would be needed to establish causal effects.
What should I do if I have been taking benzodiazepines long-term?
Discuss benzodiazepine discontinuation with your doctor. Abrupt cessation can trigger withdrawal symptoms; a slow, medically supervised taper is standard. Your clinician can develop a safe tapering schedule, address underlying anxiety or insomnia with alternative therapies, and monitor for relapse. Do not stop benzodiazepines without medical guidance.
The Ontario findings underscore a fundamental principle in prescribing: initial clinical decisions reverberate through treatment trajectories. By aligning benzodiazepine prescribing duration with evidence-based recommendations for brevity and time-limited use, clinicians may substantially reduce the incidence of unintended long-term dependence. As health systems refine prescribing protocols and electronic health records integrate duration-based safeguards, population-level data like these provide the empirical foundation for meaningful change.
Was this article helpful?
Disclaimer. This article is health journalism intended for general information and education. It is not medical advice and is not a substitute for professional diagnosis or treatment. Always consult a qualified healthcare provider about your individual circumstances. Full disclaimer →
Related Coverage
Medically reviewed by Prof. Giorgi Pkhakadze, MD, MPH, PhD. Spotted an error? Contact the editorial team.
