Predictors of opioid overdose during the COVID-19 pandemic: The role of relapse, treatment access and nonprescribed buprenorphine/naloxone.

Introduction: Emerging data indicate a disproportionate increase in overdose deaths since the onset of COVID-19. Speculation about causes for the increase center on rising drug use, illicit drug supply changes, and reduced treatment access. Possible overdose mitigation factors include reduced federal MOUD prescribing restrictions, naloxone distribution programs, and increased use of telehealth. Similarly, nonprescribed buprenorphine (NPB) use, increasingly described as a harm reduction strategy in the absence of treatment, may have moderated overdose risk. This study explored factors associated with pandemic-related overdose in people who use opioids (PWUO) in New Jersey.

Methods: We surveyed 342 PWUO from March to May 2021. Approximately 50 % of our sample was treated at some time since the COVID-19 emergency declaration in March 2020. The risk and protective factors associated with overdose were identified using Pearson's chi square test and ANOVA and tested in a series of multivariable logistic regression models for the full sample and the subsample of PWUO treated during the pandemic.

Results: Forty-eight percent of respondents increased their drug use during the pandemic, including 32 % who relapsed after previous abstinence. Fifteen percent overdosed at least once since March 2020. In the full sample, overdose was associated with Hispanic ethnicity (AOR = 3.51; 95 % CI = 1.22-10.11), pre-pandemic overdose (AOR = 6.75; 95 % CI = 3.03-15.02), lack/loss of medical insurance (AOR = 3.02; 95 % CI = 1.01-9.02), relapse (AOR = 2.94; 95 % CI = 1.36-6.36), and nonprescribed use of buprenorphine/naloxone (AOR = 3.16; 95 % CI = 1.49-6.70). The study found similar trends in the treatment sample, with the exceptions that heroin/fentanyl use also predicted overdose (AOR = 3.43; 95 % CI = 1.20-9.78) and the association of overdose with nonprescribed buprenorphine/naloxone was stronger (AOR = 4.91; 95 % CI = 2.01-12.03). Potential mitigating factors, such as take-home methadone and telehealth, were not significant.

Conclusions: Relapse during the pandemic was widespread and a significant contributor to overdose. Lack/loss of medical insurance further exacerbated the risk. Despite the growing literature reporting "therapeutic" use of NPB, people using nonprescribed buprenorphine/naloxone in the current study experienced up to five times the risk of overdose as nonusers. This finding suggests that, despite therapeutic intent, PWUO may be using NPB in ways that are ineffectual for addiction management, especially in the context of changing buprenorphine induction protocols in the context of fentanyl.