Long-term opioid use conferred a 40% higher death risk in a prospective study of U.S. veterans with or without HIV. The heightened risk ran to 46% in veterans with HIV compared with 25% in those without HIV. Long-term benzodiazepine use also inflated death risk in this population.
Prescription of opioid analgesics and benzodiazepines, sometimes together, is on the rise in the U.S. Researchers at Yale University and other institutions noted that many patients do not benefit from long-term opioids, which often cause side effects. Long-term benzodiazepines for anxiety and chronic insomnia also carry risks, the authors observed, and could be replaced by safer drugs. Opioid overdose represents a public health crisis in the U.S., they added, while prescription opioids and benzodiazepines may contribute to mortality through myriad mechanisms, including falls, motor vehicle accidents and respiratory infections.
The study, published in the Journal of Acquired Immune Deficiency Syndromes, analyzed data on 64,602 U.S. veterans. To get a better understanding of how long-term (≥ 90-day) opioid and/or benzodiazepine use affect all-cause mortality in people with and without HIV, the investigators conducted a prospective study of the Veterans Aging Cohort Study-Virtual Cohort (VACS-VC) in fiscal years 2009 and 2010. VACS-VC matches HIV-positive and negative veterans 1:2 by age, sex, race/ethnicity and study site.
HIV-positive people in this analysis were all taking antiretroviral therapy, and no veterans analyzed had cancer. The investigators determined opioid and benzodiazepine use through pharmacy data and excluded medications used to treat opioid dependence. They determined how many people died in fiscal year 2010 by checking the Veterans Health Administration vital status file.
The analysis focused on 64,602 veterans, 16,989 of them (26%) with HIV infection. Veterans with HIV had lower rates of long-term opioid use, benzodiazepine use or both opioid and benzodiazepine use. During follow-up, 1,570 veterans died, 539 with HIV and 1,031 without HIV (3.2% versus 2.2%, P < .001).
Because observational studies like this are liable to confounding by indication (for example, mortality may be fundamentally higher in people prescribed certain medications), the researchers used a method called propensity score matching, which in this case adjusted for factors influencing the possibility of receiving long-term opioids or benzodiazepines.
In this analysis, long-term opioid use was associated with a 40% higher death risk (hazard ratio [HR]: 1.40, 95% confidence interval [CI]: 1.22 to 1.61), while long-term benzodiazepine use was associated with a 26% higher death risk (HR: 1.26, 95% CI: 1.08 to 1.48). Veterans prescribed both long-term opioids and long-term benzodiazepines ran a 56% higher death risk (HR: 1.56, 95% CI: 1.26 to 1.92).
Further analysis indicated a significant interaction between HIV status and risk of death associated with long-term opioids: Veterans with HIV had a 46% higher death risk with long-term opioids (HR: 1.46, 95% CI: 1.15 to 1.87), while HIV-negative veterans had a 25% higher risk (HR: 1.25, 95% CI: 1.05 to 1.49). HIV-positive veterans who took both opioids and benzodiazepines over the long term ran a 65% higher death risk (HR: 1.65, 95% CI: 1.15 to 2.38), compared with a 29% higher risk among HIV-negative veterans (HR: 1.29, 95% CI: 0.98 to 1.71).
The researchers also found evidence tying higher opioid doses (≥20 mg morphine-equivalent daily dose) to a greater risk of death from combined long-term opioids and benzodiazepines. This threshold was higher (≥50 mg morphine-equivalent daily dose) for long-term opioid use alone.
The authors believe their study "adds to a nascent understanding of the overlapping harms associated with psychoactive substance coprescribing." They suggested that drug-drug interactions in patients taking multiple agents may account for the increased death risk associated with opioid and benzodiazepine use. The researchers called for caution in prescribing opioids and benzodiazepine together, especially in people with HIV infection, who have an overall increased risk of death.
Mark Mascolini is a freelance writer focused on HIV infection.