Alcohol use disorder (AUD), associated with significant morbidity and mortality, continues to be a major public health problem. The COVID-19 pandemic exacerbated the impact of AUD, with a 25% increase in alcohol-related mortality from 2019 to 2020. As such, innovative treatments for AUD are urgently needed. Inpatient alcohol withdrawal management (ie, detoxification) accounts for over 20% of the annual admissions to publicly funded facilities to treat substance use disorders [1]. Such admissions are often an entry point to recovery for individuals with AUD, and a successful linkage to ongoing treatment has been shown to improve the outcomes [2]. The types of treatment to which patients are linked can vary considerably and may include additional intensive interventions such as partial hospital, residential, and intensive outpatient services, or a range of outpatient treatments such as outpatient counseling and attendance at 12-step programs. Although estimates vary, only 30%-35% of patients receive some form of addiction treatment after discharge [3]. One of the major barriers for successful linkage is the fragmentation of detoxification treatment from outpatient addiction treatment services [4].

Peer recovery coaches are deployed increasingly in health care settings, often providing services outside the scope of traditional clinical services [5]. Recovery coaches have the “lived experience” of sustained recovery who obtain training to become certified to provide support for individuals in recovery [6]. Recovery coaches’ roles vary tremendously depending on the setting and patient population, ranging from single encounters in the emergency department to provide referral to services, to providing ongoing support to patients longitudinally over time [7,8]. What is consistent across these various roles, however, is that peer recovery coaches often provide social support, build trust, respect all pathways to recovery, instill hope, and assist in navigating community recovery resources [6].

Additionally, mobile technologies are increasingly studied to improve treatment outcomes for a variety of illnesses, including AUD [9]. Prior trials that have combined smartphones with peer recovery coaches have included primary care patients with hazardous drinking [10,11] or individuals in the emergency department with opioid use disorder [12]. However, there are no prior studies that have attempted to leverage technology to increase engagement with peer recovery coaches to specifically improve linkage to care following inpatient alcohol withdrawal management treatment.

To fill this gap, we conducted a proof-of-concept study to evaluate the feasibility of an existing care coordination platform and smartphone app (Lifeguard) in facilitating recovery coach support for patients being discharged from detoxification. Lifeguard is a smartphone app that was designed to assist patients, their caregivers, and their providers by giving reminders about upcoming appointments and medications, allowing caregivers to remotely support the patient, tracking symptoms such as vital signs, giving providers patient-reported measures between visits, and creating care plans. We aimed to conduct a small proof-of-concept trial to assess the feasibility of a peer recovery coach using this smartphone app to provide the support to patients undergoing inpatient withdrawal treatment, with the goal of improving linkage to ongoing addiction treatment after discharge.

In total, we approached 24 individuals, of whom 12 were excluded for the following reasons: declined participation (n=3), discharged before enrollment could proceed (n=2), unable to contact participant for the screening visit (n=2), required inpatient psychiatric care for suicidal ideation (n=1), lacked capacity to provide consent (n=2), and did not have a phone (n=2). Of the 12 who provided informed consent, 2 were not enrolled due to technical difficulties with the app which the study staff could not resolve. A total of 10 individuals were found to be eligible and provided informed consent to participate in the study. The participants were all men with AUD admitted for inpatient withdrawal management; averaged 50.5 years old; and were mostly of White racial backgrounds (n=6), non-Hispanic (n=9), and single (n=8).

Given that this proof-of-concept study aimed to evaluate the feasibility, our outcomes of interest were primarily with the proportion of participants engaging with the coach before and after discharge, responding to BAM prompts, and successfully linking with addiction treatment by 30-day follow-up. Overall, 8 participants successfully engaged with the coach prior to discharge while still on the inpatient unit. Following discharge, 6 participants continued to engage with the coach, doing so on an average of 5.3 days (SD 7.3, range 0-20 days). During the 30 days post discharge, 5 participants responded to the BAM prompts, doing so on an average of 4.6 days (SD 6.9, range 0-21 days). Half of the participants were successfully linked with ongoing addiction treatment during the follow-up. The participants who engaged with the coach post discharge, compared to those who did not, were significantly more likely to link with treatment (83% vs 0%, χ²=6.67, P=.01). Responding to the BAM prompts was not associated with successful linkage (80% vs 20%, χ²=3.60, P=.06).

This proof-of-concept study evaluated the feasibility of a digitally assisted peer recovery coach in engaging patients undergoing inpatient alcohol withdrawal management. We additionally evaluated if coach engagement might facilitate linkage to ongoing treatment. While still preliminary, the results were promising in the feasibility outcomes examined, with most of the participants engaging with the coach and responding to the prompts during the follow-up period. The results also indicated that engagement with the coach was associated with a greater likelihood of linking to ongoing addiction treatment after discharge, suggesting possibly that the support provided by peers can positively impact postdischarge outcomes. Nevertheless, the total number of days that the participant engaged with the coach was limited, including the total number of days responding to BAM prompts, pointing to the challenges when in-person contact is absent. While in-person interventions are likely to have a greater magnitude of effect on the outcomes, there is a growing recognition for the role of digital technologies in improving clinical outcomes of individuals in recovery [15,16]. Taken together, the results suggest that further research is warranted to evaluate the benefits of recovery coaches to facilitate postdetoxification linkage to care through a smartphone app.

It remains unclear if providing support during the transition after inpatient treatment is the critical ingredient in enhancing continuing care for those in recovery. Prior trials to improve postdetoxification linkage to care have produced mixed results. In a randomized trial, a brief family treatment during the admission was superior to treatment as usual (TAU) in increasing entry to treatment at 30 days [17]. In another trial, a staff escort to the aftercare appointment along with a small incentive improved linkage with treatment compared to incentives alone or TAU [18]. In another trial that randomized participants to either a peer- or physician-led referral to 12-step programs or no active referral at all, those receiving a peer-led referral were significantly more likely to attend a 12-step meeting (64%) compared to physician-led referral (48%) or to controls (33%). Finally, patients receiving a visit from a nurse from the outpatient clinic to which patients were referred were significantly more likely to link with care compared with TAU [19]. By contrast, in a randomized trial comparing motivational enhancement therapy, peer-delivered 12-step facilitation, and TAU, linkage at 30 days was no different between the arms [20]. In another trial, patients receiving an enhanced telephone monitoring—consisting of a 50-minute in-person individual session and 12 weekly 15-minute phone calls after discharge—were no more likely to engage with outpatient treatment at 3-months compared to TAU [21]. These results suggest that more research is needed to identify how to improve postdetoxification linkage to care.

The BAM used for this trial was designed to deliver patient-reported measures to the peer recovery coach, who can then intervene if needed by providing support and referral to services. However, the overall response to BAM was limited (40% never responded), and of those who responded, the frequency of responding was low (mean of 4.6 out of 30 days). This may suggest that daily requests to complete the BAM may be too burdensome to participants. Nevertheless, the results suggested that those who responded to the BAM prompts may be more likely to engage with aftercare, although the results were not statistically significantly (P=.06). As such, further research is warranted to assess the optimal frequency for requesting patient-reported measures, as well as whether such prompts can enhance engagement with the peer recovery coach or improve linkage to care.

There are several limitations to this study. This was a small proof-of-concept pilot study, with no control group, and used a convenience sample at a single institution. Although we did not restrict enrollment by gender, only men chose to enroll. Most of the follow-up postdetoxification occurred at programs outside of our institution, which prevented confirmation of linkage via chart review. Finally, the lack of any in-person contact may have hampered the degree of engagement between the participant and the coach.

A recovery coach using a smartphone app may feasibly support patients following admission for inpatient withdrawal management. Further research is warranted to evaluate the potential role for peer recovery coaches in improving postdischarge linkage to treatment.