37% risk reduction
Of the original cohort of 304,676 patients with OUD, the study’s final analytic sample included 23,036 patients (mean age, 30 years; 51% men), representing 14,213,075 person-days of insurance coverage. Of these, 2,210,927 person-days (15.6%) entailed claims for buprenorphine (mean daily dose, 15.4 mg; SD, 7.31 mg).
A total of 474,181 person-days included claims for benzodiazepines or Z-drugs with concurrent buprenorphine treatment. The mean daily dose of any benzodiazepine or Z-drug was 23.4 diazepam-milligram equivalents. The mean daily dose of short-acting benzodiazepines, long-acting benzodiazepines, and Z-drugs was 25.3, 31.3, and 4.9 diazepam-milligram equivalents, respectively.
Buprenorphine treatment days were associated with a 37% lower chance of drug-related poisoning (95% confidence interval, 0.60-0.66) in comparison with nontreatment days. On the other hand, the odds of poisoning increased by 81% on days on which patients were treated with Z-drugs or benzodiazepines (95% CI, 1.73-1.91).
Interestingly, individual analyses showed that benzodiazepine and Z-drug treatment days were associated with increased odds of poisoning events (odds ratio, 1.29; 95% CI, 1.19-1.39). Odds of poisoning events on benzodiazepine-only treatment days, on the other hand, were markedly lower (OR, 1.88; 95% CI, 1.78-1.98).
Subgroup analyses revealed that both short-acting and long-acting benzodiazepine treatment days were associated with comparably elevated odds of drug-related poisoning (OR, 1.86 and 1.68, respectively). High-dose benzodiazepine treatment days were associated with higher increased odds of a poisoning event (122%) in comparison with low-dose treatment days (78%).
High-dose, but not low-dose, benzodiazepine or Z-drug treatment was linked to increased poisonings when the drug was taken concurrently with buprenorphine (OR, 1.64; 95% CI, 1.39-1.93). However, the risk was still lower than the risk associated with taking the agents without concurrent treatment with buprenorphine (low-dose OR, 1.69; high-dose OR, 2.23).
‘Not all benzodiazepines are bad’
Dr. Xu noted that the findings have potentially important implications for clinical practice, beginning with the dose-dependent relationship between benzodiazepine/Z-drug use and drug-related poisonings among individuals with OUD. This indicates that lowering doses or shortening treatment duration may reduce risk, he said.
Similarly, the lower risk associated with long-acting benzodiazepines relative to short-acting beonzodiazepines – as well as the substantially lower risk associated with Z-drugs, compared with either short- or long-acting benzodiazepines – suggests that switching from benzodiazepines to long-acting agents or Z-drugs may lower the risk for overdose, he added.
“Clinicians are often challenged by patients with opioid use disorder who are also on benzodiazepines or Z-drugs. There’s an inclination to say no to them, because they’re too high risk to start buprenorphine maintenance, or abruptly taper the benzodiazepines, which can be very destabilizing,” he noted.
“Our data show that people on benzodiazepines can absolutely receive buprenorphine and still get some benefit,” Dr. Xu said. “In addition, not all benzodiazepines are bad for these individuals. There are safer formulations and safer doses, too.”
However, he added, he would not initiate benzodiazepine treatment if he didn’t have to, especially long-term treatment.
“One of the messages from our data is that this clearly contributes to higher overdose risk. But we often inherit patients who already have benzodiazepines on board, so we need to figure out what to do. That is the question that nobody had really clearly addressed prior to this study,” Dr. Xu concluded.