Worldwide, an estimated 62 million people use opioids for non-medicinal purposes (World Drug Report, 2021). Unregulated prescription-type opioids, such as fentanyl and other synthetic opioids, continue to be major contributors to the opioid crisis. In North America, fentanyl and its analogues are now involved in the majority of opioid-related deaths (World Drug Report 2021, Special Advisory Committee on the Epidemic of Opioid Overdoses, 2021). Curbing the opioid crisis likely requires a multi-faceted approach that targets the problem through both improvements in prevention measures as well as improvements in access to and efficacy of treatment and harm reduction strategies.

Treatment for prescription-type opioid use disorder (POUD) typically involves the use of an opioid agonist to help ease symptoms of withdrawal, reduce craving, and ultimately decrease substance use and improve other outcomes. However, individuals may continue to use opioids despite receiving such treatment (Fischer et al., 2012). Reasons for continued drug use while receiving opioid agonist therapy (OAT) are varied and not always clear; some individuals indicate that a lack of control over craving contributed to their continued opioid use while on OAT (Fischer et al., 2012), which is a common theory that clinicians also use to explain ongoing opioid consumption during treatment.

Research examining craving as a predictor of opioid use while on treatment has been faced with mixed results. While numerous studies have failed to link drug craving with subsequent relapse (Serre et al., 2015), other studies using ecological momentary assessment have demonstrated a link between cravings and subsequent heroin use (Johnson et al., 2009, Marhe et al., 2013, Preston et al., 2018), perhaps due to the shorter time frames between craving measurements and outcomes. Among treatment-seeking people with a prescription opioid use disorder, McHugh et al., found that cravings predicted greater odds of next-week opioid use (McHugh et al., 2014). A more recent study showed that, while cravings predicted next-week opioid use among patients with chronic pain, the relationship between craving and opioid use was reduced when participants were provided with pain coping counselling (Messina and Worley, 2019). Together, these findings suggest that the relationship between cravings and opioid use is complex and potentially dynamic, such that the extent that craving is associated with opioid use may not remain stable across different contexts or periods of time, among other factors.

Our recent findings have shown that, relative to methadone, buprenorphine/naloxone (BUP/NX) treatment for POUD was associated with fewer opioid-positive urine drug tests over a 24-week period (Jutras-Aswad et al., 2022) and a decrease in past 24-hour craving (McAnulty et al., 2022). The extent to which decreased craving was responsible for the reduction in opioid use has yet to be explored. To address this question, we first aimed to determine if past 24-hour craving would be associated with next-week opioid use. We chose to use next-week opioid use based on findings that craving could predict opioid use in the next week (McHugh et al., 2014, Messina and Worley, 2019). We then aimed to determine to what extent the reduction of opioid use during OAT could be explained by a reduction in craving. We hypothesized that lower self-reported opioid use during BUP/NX treatment relative to methadone would be mediated at least in part by a reduction of craving.