Woebot Reduced Problematic Substance Use During Pandemic
When Covid-19 first hit, most U.S. states were quick to designate liquor stores and other alcohol sales outlets as “essential businesses.” Little wonder that with all the isolation and uncertainty the pandemic brought on, Nielson reported a 21 percent rise in brick-and-mortar alcohol dollar sales and a 234 percent increase in online sales of alcohol compared to the year prior. Where did all that alcohol go? A recent RAND survey of 1,540 adults showed that in 2020, three of four adults drank more per month, and women indulged in 41 percent more heavy drinking compared to 2019.
The National Institutes of Drug Abuse (NIDA) recently awarded Woebot Health grant funding to collaborate with Stanford University’s School of Medicine to study Woebot among adults with substance use disorder. Addiction expert and Principal Stanford Investigator Judith (Jodi) Prochaska, PhD, MPH, and Woebot Health’s Chief Clinical Officer and Principal Investigator Athena Robinson, PhD, talked about the first study results, which were just published in The Journal of Medical Internet Research (JMIR) and highlighted in this press release, showing Woebot significantly reduced problematic substance use during the pandemic.
Substance use is a hot topic. What makes this study unique?
Athena: Most studies focused on substance use disorder interventions have been about web-based solutions. This is one of the first studies, maybe even the very first, to explore an intervention for Substance Use Disorders (SUDs) delivered via a therapeutic relational agent readily available through a mobile app.
Jodi: The research is timely. A search of “alcohol” or “addiction” in apps stores brings up many options — ranging from alcohol home delivery and bartender tutorials to sobriety counters and education on addiction. Yet, evidence that mobile apps work for reducing problematic alcohol and drug use is limited. Our plans to evaluate Woebot for substance use disorders were developed pre-pandemic. Notably, the need for novel in-hand treatment options became amplified during a time when people were isolating at home and restricted from accessing addiction counseling or self-help services. Nearly all (98 percent) of participants were sheltering-in-place at study start due to the COVID-19 pandemic. This first study tested program feasibility, acceptability, and efficacy.
What did the study find?
Jodi: Participants used Woebot regularly, sending 75 messages on average per week; they found the app acceptable; and they reported a 30 percent decrease in substance use occasions from their levels at baseline. Woebot helped the study participants manage their substance use over an eight week period, and through the very unusual conditions of the pandemic.
Athena: We also saw significant improvements in self-reported measures of craving and confidence. Craving ratings were reduced by half and confidence to resist urges rose by more than 36 percent. Participants also reported significant reductions in symptoms of depression and anxiety.
Did any of the results surprise you?
Jodi: While participants’ ratings of acceptability were encouraging overall, Woebot scored highest on effective bond formation, and that surprised me.
What’s effective bond formation?
Jodi: The measure of effective bond formation reflects the rapport that developed between participants and Woebot. In the field, close rapport between therapist and client is a big predictor of the level of change achieved in therapy. It’s incredibly novel that participants developed an affinity for and bond with Woebot, which fully disclosed being a non-human coach. It supports the idea that a relational agent, when well designed, can be engaging and effectively empathetic.
Athena: One other thing that surprised me was the increase in participants’ confidence to manage cravings. Self-efficacy is a well understood phenomenon within psychological research, and it has been shown in multiple domains to impact outcomes. That we were able to impact self-efficacy and confidence in such a brief, self-guided intervention is fascinating. There was also a lot of interest in this study. The sheer number of people who wanted to participate helped me understand just how many people in the real world are struggling with substance use on a day-to-day basis.
Is substance misuse the same thing as addiction?
Jodi: Repeated misuse of a substance can lead to addiction. Addiction is characterized by compulsive drug seeking and continued use despite harmful consequences. Addiction causes long-lasting changes in the brain, especially in the brain’s natural inhibition and reward centers.
Tell us a bit more about the NIDA grant that made this study possible.
Athena: We’re incredibly honored to have received a grant from the National Institute on Drug Abuse (NIDA) and to be part of the government’s investment in innovation to address important societal issues. The grant was for $1.5 million to study, with Stanford, the feasibility, acceptability and efficacy of our products among adults with substance use disorders.
What’s next?
Athena: This was an initial, non-controlled study that examined pre/post changes in a single group design. We’re doing an iterative program of research with Stanford. The next phase of our NIDA work will investigate Woebot’s efficacy compared to an active control condition in a fully-powered randomized clinical trial and will also explore associations between pandemic-related stress and substance use problems.
