Patients who received longer initial benzodiazepine prescriptions were less likely to discontinue treatment than those prescribed 7 days or fewer, according to findings published in PLOS Medicine.
Researchers analyzed linked administrative health care data from 1.8 million adults in Ontario, Canada, who received a new oral benzodiazepine prescription from 2013 to 2020, with follow-up through 2021. New use was defined as no benzodiazepine prescription in the prior 182 days. Z-drugs, midazolam, and clobazam were excluded from the cohort, as were patients who had received palliative care in the 6 months prior to cohort entry. Lorazepam was the most commonly prescribed benzodiazepine at initiation, followed by clonazepam and diazepam.
The primary exposure was the duration of the index prescription, categorized as 7 days or fewer, 8 to 14 days, 15 to 30 days, or more than 30 days. The primary outcome was time to benzodiazepine discontinuation, defined as no subsequent benzodiazepine dispensing within 1.5 times the days supplied, with a minimum look-forward window of 30 days.
Following adjustment for demographic, clinical, and prescribing factors, longer initial prescriptions were associated with a lower likelihood of discontinuation. Compared with prescriptions of 7 days or fewer, prescriptions of 8 to 14 days were associated with about half the likelihood of discontinuation, prescriptions of 15 to 30 days with about one-quarter the likelihood, and prescriptions of more than 30 days with about one-seventh the likelihood.
The median time to discontinuation was 19 days overall, 16 days among female patients, and 19 days among male patients.
Patients initially dispensed 2 or more benzodiazepines were also less likely to discontinue treatment than those dispensed 1 benzodiazepine. Long-acting benzodiazepines and combinations of short- and long-acting benzodiazepines were associated with delayed discontinuation compared with short-acting agents alone.
Results were generally consistent across sex-stratified and sensitivity analyses for prescription duration, benzodiazepine type, and number of benzodiazepines prescribed. Findings involving initial dose were smaller and less consistent, with some differences in direction by sex and across sensitivity analyses.
The researchers noted that current benzodiazepine prescribing guidance generally defines short-term prescribing as lasting less than 4 weeks. They suggested their findings may indicate that threshold is too liberal, because prescriptions lasting more than 7 days were associated with a greater likelihood of prolonged use.
The study had several limitations. Dispensing records could not confirm whether patients took medications as prescribed. Some medications, including z-drugs and certain psychiatric medications, were not fully captured. The analysis also could not account for disease severity, nonprescribed benzodiazepine use, social stability, or other unmeasured factors. Because the study was observational, the findings should be interpreted as associations rather than evidence that initial prescribing choices caused prolonged benzodiazepine use.
Disclosures: The study was funded by a Womenmind grant. Several researchers reported consulting fees, research support, honoraria, travel support, academic awards, or stock ownership unrelated to the study. The funders had no role in the study design, data collection, analysis, publication decision, or manuscript preparation.
Source: PLOS Medicine
