Abstract
The number of unintentional deaths involving opioid and/or benzodiazepine use continues to increase in Australia. This study examined patterns of concurrent prescribing of opioids and benzodiazepines/Z-drugs (BZDs) for long-term use in Australian general practice. A cross-sectional analysis was undertaken using MedicineInsight, a national database of de-identified general practice electronic health records. We estimated the proportion of patients (per 1000, ‰) in 2017 receiving concurrent prescriptions for opioid and BZD medications for long-term use (≥ 3 prescriptions within 90 days). Poisson regression models were used to estimate the marginal adjusted prevalence ((adj)P) and adjusted prevalence ratios ((adj)PR) were used to compare concurrent long-term prescribing according to sociodemographic characteristics, rurality, smoking status, and diagnosis of mental health or musculoskeletal conditions. The sample included 1,207,671 individuals (41.3% males; mean age 50.6 ± 18.6 years) regularly attending 544 general practices. The prevalence of concurrent long-term opioid and BZD prescribing was 7.0‰, and the median duration of prescribing overlap was 611 days (p25-p75 348-952). The prevalence was higher for patients aged over 65 years ((adj)PR = 3.62 95% CI 3.30, 3.98), females ((adj)PR = 1.33 95% CI 1.27, 1.39), those living in more disadvantaged ((adj)PR = 1.70 95% CI 1.49, 1.93) or rural/remote areas ((adj)PR = 1.13 95% CI 1.00, 1.28), smokers ((adj)PR = 4.10 95% CI 3.87, 4.35), and those with mental health ((adj)PR = 3.23; 95% CI 2.83, 3.69) or musculoskeletal conditions ((adj)PR = 2.74; 95% CI 2.47, 3.04). In patients with both mental health and musculoskeletal conditions, the prevalence was 32.1‰. Interventions to reduce concurrent long-term prescribing could be targeted to the identified vulnerable groups.