Skip to main content

Recovery Coaching, Digital Breathalyzer Boost Retention in Outpatient SUD Treatment

September 23, 2020

High attrition rates have long plagued outpatient behavioral health providers. Attrition rates are especially high within substance use disorder treatment programs1. Retention in treatment has often been a primer for innovation and organizational change in attempts to retain participants for both participant recovery and organizational financial benefit 2,3. This article outlines observed problems within a traditional outpatient substance use disorder treatment program and applied methods for improvement to increase retention and decrease attrition.

Pavillon Outpatient Services offers outpatient substance use disorder (SUD) treatment in Greenville, South Carolina. The primary service offering is an intensive outpatient program (IOP). Alcohol is the primary substance of use for the majority of participants (approximately 75%). The IOP consists of 9 hours of group work per week, individual therapy, a family group component and urine drug screening at random intervals. Although our rates of participant attrition from the service were comparable to national averages, we were concerned about attrition from both a participant-success and business-case perspective.

Problems and changes

Components of program design were identified as points of possible improvement during the change project. Participants generally only engaged with the treatment provider on service days, on average 3 days per week. We found participants generally lacked engagement with local recovery resources. The need to maintain appropriate counselor-participant boundaries impaired the clinician’s ability to make direct connections with some community resources. The time required to obtain laboratory urine drug screen results impeded the clinician’s ability to assist with return to use within an appropriate timeframe. Traditional random urine drug screenings left gaps in time for potential use and were not providing a sufficient level of structure for persons served. Further, these time gaps led to late identification of those who could benefit from changes to their service plan or referral to higher levels of care.

Two changes were identified as potential solutions to the identified problems: 1. addition of recovery coaching and 2. addition of a cellular phone digital breathalyzer. Our benchmark measures for improvement for the change project included: retention rate, attrition rate, days of treatment before attrition, and days of treatment before referral to higher level of care. We defined retention as participant completion of the full IOP treatment episode with an average length of stay being between 6 and 8 weeks. We defined attrition as a participant leaving treatment before completion and lost to contact for referral to an alternative level of care. If successful contact was made and referral facilitated, the participant was considered a transfer. We hypothesized that addition of coaching and a cell-based breathalyzer would have the following impacts:

  • Retention rate would increase
  • Number of treatment days before treatment attrition would increase
  • Staff would detect return to use earlier and refer to a higher level of care more quickly therefore number of treatment days before referral to higher level of care would decrease.

Addition of recovery coaching

Recovery coaching, also known as peer recovery support services, has a long history within substance use disorder treatment. Recovery coaching has proven to increase engagement with treatment providers, increase linkage to community resources, increase treatment retention, and is correlated with decreased substance use and relapse, although limited validated research exists for the exact effect of integrating recovery coaching into a treatment program4,5

A full-time recovery coach (RC) was added to the outpatient staff and was trained at a local recovery community organization. Resources for this new position were reallocated by combining two administrative positions, therefore limited additional staffing costs were incurred. Although the IOP operates from a 12-Step foundation, the RC approached work with participants from an all-recovery perspective. The RC was tasked with making contact with each participant on non-service days, including scheduling individual recovery coaching sessions. The RC participated in IOP groups and occasional individual therapy sessions. The RC’s schedule included 2-3 community recovery support meetings (AA, NA, SMART, Refuge Recovery, All Recovery) per week which were announced to all IOP participants. A cell phone was provided for the RC, and communication via text message was permitted and encouraged. 

Benefits emerged rapidly upon implementation of the recovery coach position. Participants began to engage more frequently in community resources (AA, NA, SMART, Refuge Recovery, All Recovery meetings) when the RC was present. Participants tended to communicate more frequently with the RC when available through text message. Participants would often confide in the RC about return to use before confiding with the clinician. Participants also began to contact the RC for emotional and recovery support during difficult days more frequently than they previously contacted clinicians. Creation of this position also increased the availability of basic engagement with participants from 3 days per week to 5, with improved flexibility compared to formally scheduled clinical contacts. 

Addition of cellular phone digital breathalyzer

Similar to the value of laboratory data with other serious and chronic conditions, monitoring drug and alcohol use with laboratory samples is an important component of wider disease management and recovery management strategies6. Typically, monitoring is achieved by urine drug screening. However, instant urine drug screens have a wide range of false positive and false negative results, while laboratory verification can delay clinician response. Also, urine drug screens can only be administered in-person during service days. A variety of monitoring technologies have emerged in recent years. Cellular phone digital breathalyzers (CPDB) have been an effective method for monitoring early detection of return to use7 and participants have reported positive experiences using CPDB’s8 in outpatient SUD treatment.

A convenient and reliable cellular phone digital breathalyzer provider was chosen for this project. CPDB breathalyzers were given to every client upon admission into the intensive outpatient program, regardless of primary substance of use or history of alcohol use. We considered CPDB use a requirement of the program. Costs for the CPDB were diverted from other programmatic funds, and the participants incurred no additional cost. The device is compact and discrete and connects to the participant’s cellular phone through a Bluetooth connection. Each participant was asked to provide a breath alcohol sample 3-4 times per day at preselected times, 7 days per week throughout the duration of their time in the program. Times were determined with input from the Recovery Coach based on participant drinking patterns. An automated reminder text message was sent for each testing time, and the participant had an hour-long window to test. The test was captured through cellular phone video to heighten contextual validity of the test. An online system was used to track test results. Breath alcohol content test results enabled timely clinical response to positive results as well as positive reinforcement for negative results. Traditional random urine drug screens were also continued.


Results were calculated after 12 months of full implementation of recovery coaching and cellular phone digital breathalyzer. The IOP retention rate, as measured by the percentage of participants that successfully completed the IOP, increased by 9.7 percentage points. The IOP attrition rate, as measured by the percentage of participants that dropped-out of the IOP and were lost to contact, decreased by 9.8 percentage points. The transfer rate, number of treatment days before attrition, and number of treatment days before transfer, remained the same after implementation of recovery coaching and CPDB. 


Overall, the change project led to an increase in retention. We interpret these results as suggesting addition of Recovery Coaching and cellular phone digital breathalyzers have a positive effect on retention rates within an outpatient substance use disorder treatment program.

Some problems surfaced with the implementation of CPDB testing. The primary problem, although rare, was lack of technology proficiency among participants. Some participants had difficulty connecting the CPDB to their cellular phones and downloading the application. A few participants possessed older models of phones and were unable to download the application. Some participants refused to use the CPDB and an organizational decision was made to not discharge a participant for refusal to use the CPDB, although no participants who refused the CPDB completed the program. Few positive test results have been disputed and laboratory-confirmed urine drug screens were used for verification. 

Clinicians were hesitant to adopt the CPDB. They viewed the adoption of required CPDB testing as a potential flashpoint for patient resistance due to the inconvenience and feelings that the testing was intrusive. Overall, very few complaints from participants have emerged. Most participants have found the CPDB to be helpful in preventing return to use, rebuilding trust in their relationships, and a healthy form of structure. Few reported feeling extremely inconvenienced from the testing.

The clinicians now consider the CPDB testing an invaluable component of treatment. Benefits have emerged from the implementation of CPDB testing outside of participant self-reported benefits. The CPDB allows clinicians to intervene within 24 hours of a drinking episode. This allows for more meaningful engagement around the drinking episode and rapid increase in support services (i.e., increased recovery coaching sessions, increased individual therapy, family collaboration). 

Given the results, an organization-level decision was made to retain recovery coaching and cellular phone digital breathalyzer as permanent components to the treatment program.

Bob Hennen, MS, LCMHC, LPC, LCAS, LAC is Outpatient Services Manager for Pavillon. Brian Coon, MA, LCAS, CCS, MAC is Director of Clinical Programs at Pavillon in Mill Spring, N.C. 


1 Loveland D, Driscoll H. Examining Attrition Rates at One Specialty Addiction Treatment Provider in the United States: A Case Study Using a Retrospective Chart Review. Subst Abuse Treat Prev Policy. 2014;9(41).

2 McCarty D, Gustafson DH, Wisdom JP, Ford J, Choi D, Molfenter T, Capoccia V, Cotter F. The Network for the Improvement for Addiction Treatment (NIATx): Enhancing Access and Retention. Drug Alcohol Dep. 2017;88(2-3):138-145.

3 Boyle M, Loveland D, George S. Implementing Recovery Management in a Treatment Organization. In Kelly, J & White, W. L. (Eds): Addiction Recovery Management. 2010:235-258.

4 White W. Recovery Coaching: A Lost Function of Addiction Counseling? Counselor. 2004;5(6), 20-22.

5 Eddie D, Hoffman L, Vilsaint C, Abry A, Bergman B, Hoeppner B, Weinstin C, Kelly JF. Lived Experience in New Models of Care for Substance Use Disorder: A Systematic Review of Peer Recovery Support Services and Recovery Coaching. Front Psychol. 2019;10:1052. doi:10.3389/fpsyg.2019.01052

6 DuPont RL, Compton WM, McLellan AT. Five-Year Recovery: A New Standard for Assessing Effectiveness of Substance Use Disorder Treatment. J Subst Abuse Treat. 2015;58:1-5. doi:10.1016/j.jsat.2015.06.024

7 Hamalainen MD, Zetterstom A, Winkvist M, Soderquist M, Karlberg E, Ohagen P, Andersson K, Nyberg F. Real-time Monitoring Using a Breathalyzer-Based eHealth System Can Identify Lapse/Relapse Patterns in Alcohol Use Disorder Patients. Alcohol Alcohol. 2018;53(4):368-375. doi:10.1093/alcalc/agy011

8 Nehlin C, Carlsson K, Oster C. Patients’ Experiences of Using a Cellular Photo Digital Breathalyzer for Treatment Purposes. J Addict Med, 2017;12(2):107-112. doi:10.1097/ADM.0000000000000373

Back to Top