Perspectives on a contingency management intervention for alcohol use among consumers with serious mental illness

Abstract

Objective:

This study describes the perspectives of outpatients with serious mental illness (SMI) and alcohol dependence on their participation in a contingency management (CM) intervention for alcohol use.

Methods:

Thirty-five adults with SMI and alcohol dependence participated in a randomized trial of CM for alcohol use, where they were rewarded with prizes contingent on abstinence from alcohol. All participants were interviewed regarding their participation in CM with a consistent structure that included nine open-ended questions. Favored and disliked aspects of CM, perception of alcohol biomarker accuracy, and interest in participating in similar CM interventions provided by treatment centers, rather than researchers, were explored.

Results:

Participants spoke enthusiastically about receiving prizes, as well as how CM increased their awareness of drinking and helped support their abstinence from alcohol. Most participants felt the ethyl glucuronide biomarker urine tests used to measure alcohol use were accurate, and were interested in enrolling in CM if it was offered as a clinical program. Research staff who implemented the intervention were well-regarded by participants, and interactions with research staff were perceived positively.

Conclusions and Implications for Practice:

Adults with SMI and alcohol dependence participating in a trial of CM for alcohol use reported overall positive perceptions of and experiences with CM. Receiving small tangible prizes and having positive interpersonal interactions with study staff were reported as especially impactful. These findings indicate that CM is well-received by consumers, in addition to its empirical and practical benefits as an evidence-based, low-cost intervention.

Introduction

Contingency management (CM) is one of the few behavioral interventions for substance use that is associated with reductions in substance use in adults with serious mental illness (SMI) (McDonell et al., 2013; Roll, Chermack, & Chudzynski, 2004; Tracy et al., 2007; Weinstock, Alessi, & Petry, 2007). Prior studies have found that consumers in addiction treatment without SMI were generally positive about CM. Participants in these studies noted that reinforcers made them happy and proud of themselves, and that participating in the CM intervention was worthwhile (Hartzler, Jackson, Jones, Beadnell, & Calsyn, 2014; Kirby, Benishek, Dugosh, & Kerwin, 2006; Neal, Tompkins, & Strang, 2016; Sinclair, Burton, Ashcroft, & Priebe, 2011).

Given their cognitive, psychiatric, and clinical differences, individuals with co-occurring SMI may have a unique perspective about the strengths and weaknesses of CM due to their lived experience with mental illness and other challenges, such as homelessness, which is more common among individuals with co-occurring disorders. CM provides positive reinforcement for individuals who may be accustomed to a more penalizing system. The reward system of CM may provide this vulnerable population with much needed positive support. Additionally, the reward system provides participants with positive anticipation. Evidence that participants enjoy and find CM worthwhile may aid in success and motivation to complete a CM intervention for alcohol use.

Only one study has explored the perspectives of consumers receiving CM for addiction treatment who also have co-occurring SMIs (Srebnik et al., 2013). In this study, 29 consumers with SMI were interviewed to supplement a more comprehensive study of provider perspectives of CM (Srebnik et al., 2013). Consumers in this study cited that CM was “an incentive to continue to do good” and “[CM] gave me accountability to keep clean and sober” (Srebnik et al., 2013). No consumer-reported themes relating to disliked aspects of CM emerged. Limitations of this study included being conducted at one clinic, only about one third of CM participants completed the qualitative interviews, and the interviews included only three questions.

The present qualitative study seeks to expand on prior research with qualitative interviews to more extensively explore the experiences and perspectives on CM of consumers living with SMI and alcohol dependence who are receiving co-occurring disorder treatment within a community mental health center. Specifically, this qualitative study investigates aspects of CM that were favored and disliked by participants with SMI and alcohol dependence, perceived accuracy of ethyl glucuronide (EtG) tests (a urine test for alcohol to assess abstinence), and willingness to enter a CM program if it were offered at a community agency.

Methods

Participants

Participants were 35 adults diagnosed with SMI and alcohol dependence engaged in services at an urban community mental health and addiction treatment agency. Participants had a mean age of 45.69 (SD=9.84), 54% identified as white, 40% as Black, 3% as American Indian, and 3% as Multiracial. Sixty-six percent of participants were male. Nearly three-quarters (71%) of the participants were diagnosed with a mood disorder (major depression, bipolar I, or bipolar II) and 29% were diagnosed with a psychotic disorder (schizophrenia-spectrum or schizoaffective) via administration of the Mini International Neuropsychiatric Interview (Sheehan et al., 1998). They participated in a larger randomized controlled trial of CM for alcohol use and completed a qualitative interview upon completing the 12-week CM intervention (McDonell et al., 2017). All participants included in the qualitative analysis were randomized to CM for alcohol use.

The study involved consumers providing urine samples three times per week for 12 weeks which were analyzed for the presence of the alcohol biomarker, EtG, by conducting an immunoassay test using a Thermo Fisher Indiko analyzer (Fremont, CA). Participants were reinforced with prizes, ranging from small practical items (i.e., deodorant, toothpaste, etc.) to larger personal items (i.e., crockpot, portable DVD player) for submitting EtG-negative urine samples and for treatment attendance. A more detailed description of the procedures can be found in the study outcome paper (McDonell et al., 2017). The University of Washington Institutional Review Board approved this study.

Procedures and Materials

Research assistants conducted semi-structured qualitative interviews. The research assistants were trained in qualitative interviewing techniques by the Principal Investigator, a clinical psychologist and research professor with experience in conducting qualitative research with individuals with SMI. Questions focused on aspects of the study that participants liked or disliked, if they thought the study helped them in any way, if they thought the EtG urine test results were accurate, and if they would participate in another similar CM intervention. Questions about EtG were asked due to concerns by some about EtG providing false positives and identifying non-beverage alcohol use. The nine interview questions included the following: “What did you like about the study?”, “What did you like best about the study?”, Do you think being in the study helped you? If so, how?”, “Did earning prizes for not using alcohol help you stay sober?”, “What did you like about the study? Why?”, If an agency like [agency the CM study was conducted through] offered rewards to stay sober and show up to treatment, would you sign up or be part of the rewards program? Why or why not?”, Did you feel like the results of the alcohol UAs were accurate?”, and, “Would you like to tell us anything else?”. Interviews were transcribed as they were conducted. Audio and video content was not documented.

Analysis

A conventional content analysis framework was employed (Hsieh & Shannon, 2005). The first and second authors read and reflected upon the interviews and identified major themes during data analysis. The first author coded the transcripts and developed a draft codebook based on the raw data. The first and second authors collaboratively reviewed the codebook together and reached consensus. The second author independently coded the transcripts, adding codes as needed. When coding was completed, the first and second authors compared codes and reached consensus where disagreements were identified. All coding was conducted using QDA Miner qualitative coding software (Provalis Research, Montreal, Canada).

Results

Results follow from the four content areas of the interview: 1) aspects of the study favored or enjoyed by participants, 2) aspects of the study disliked by participants, 3) willingness to participate in CM through an agency, and 4) accuracy of urine analyses for alcohol.

The first and most extensively developed question regarded aspects of the study the participants liked. All thirty-five participants identified receiving prizes as a primary aspect of the study that they appreciated. For example, one participant described why they liked receiving prizes, “When I’m at home and see them I think ‘hey I got this [the prizes] for staying sober” [1067]. Nineteen (54%) liked the accountability and awareness of their drinking that they felt during the study. Eight (23%) cited experiencing a positive change in their perspective or attitude because of their participation. Sixteen (46%) participants cited interactions with study staff as a significant positive aspect of the study. Twenty-seven (77%) participants asserted that participating in the study and winning prizes helped them abstain from alcohol. One participant in the CM condition described wanting to abstain from alcohol, “I don’t know why, but I still wanted to be clean, at least for alcohol, even though I knew it wouldn’t be held against me and it wouldn’t be shared. I was conscious of that” [2026]. Twelve (34%) participants described the study as giving them something to look forward to, a reason to get out of the house and come to the agency (see Table 1).

Table 1.

Positive/liked aspects of the study (N=35)

Theme Domain	Example
Prizes (n=35)	“Gift cards, I’ve been rewarded for my choices.” [1078]
	“Keeping track of myself and getting rewards.” [2025]
	“I thought it was a lot of fun. Both talking about addiction and getting prizes.” [1025]
	“They’re [the prizes] fun! It’s like being a kid at a carnival.” [1075]
	“I got to replace things I’ve lost” [1053]
	“When I’m at home and see them [the prizes] I think ‘hey I got this for staying sober.’ I got the bars the bars of soap, the pens, the deodorant, the shampoo, the tweezers, the hygiene kit you gave me.” [1067]
Positive change in thoughts and/or perspective (n=8)	“Something to do besides thinking about everything wrong with the world, and being negative… it gave me a little peace of mind” [1001]
	“It was support to help me quit and it taught me I didn’t have to make the decision on my own, it’s what was best for my health and my future and why I’m here today now” [1078]
	“It caused me to look at my drinking from a different angle.” [1018]
	“It got me conscious about what I need to do to get my life back in order.” [1053]
	“I got to think about rewarding myself with the prizes and gift cards and it felt good.” [2015]
Accountability/Awareness of self and drinking (n=19)	“Yeah, being accountable.” [2035]
	“Well you know, it helped me in that I promised my own self throughout the study that I wouldn’t use so that it wouldn’t show up here. So I haven’t used.” [2033]
	“The prizes I don’t care. It’s accountability.” [1077]
	“I think it gave me UAs [urine tests] to live up to. I got something to work for.” [1025]
	“Being able to observe myself more.” [2025]
	“Yeah, just to understand my feelings and addictions, pay more attention to them.” [1056]
	“Trying to learn more about my alcohol use, and having fun without drinking.” [1023]
	“It was nice to come in and get reports on how I was doing with smoking drinking and doing drugs.” [1069]
	“I don’t care about the prizes, seeing myself getting clean it helped me because I thought more and took more time to think about using than you would.” [2020]
	“It gives me incentive to slow down and think about me and how I’m doing.” [1004]
Study staff (n=16)	“It was fun, sharing with you, uplifting even when I was feeling bummed out at first. You’ve been really patient with me I’m gonna miss working with you.” [1075]
	“Our conversations.” [2015]
	“I like talking to the staff.” [1031]
	“Meeting you! Giving me something to look forward to” [1067]
Helping to attain abstinence (n=27)	“It inspires to stay sober.” [1057]
	“It makes me stop, I want my result to be clean.” [1077]
	“I felt punished when I did something bad and I felt rewarded when I did something good, when I did my first dirty UA I said I quit I didn’t want to do this class that was my fault for the choices I make! It’s my fault I didn’t get the Grammy award cause I came to the awards drunk.” [1078]
	“I don’t know why, but I still wanted to be clean, at least for alcohol, even though I knew it wouldn’t be held against me and it wouldn’t be shared. I was conscious of that.” [2026]
	“It kept me clean and sober from the time I came in, and it still does.” [2016]
	“Being clean and sober and learning stuff.” [1041]
Something to look forward to/motivation to attend agency (n=12)	“It gave me a reason to come in with IOP.” [1012]
	“Kept me with something to do.” [1041]
	“It gave me something to look forward to, a schedule…” [1057]
	“A chance to get out of the house.” [2024]

The second question focused on aspects of the study participants did not like. While twenty (57%) participants maintained there were no aspects of the study that they did not like, five (14%) mentioned that they did not enjoy days where they did not earn prize draws (because they tested positive for alcohol as assessed by the EtG biomarker), and one (3%) participant added that the questionnaires brought up difficult feelings (see Table 2).

Table 2.

Remaining themes identified (N=35)

Aspects of the study participants disliked
Theme Domain	Example
Liked everything (n=20)	“No, I thought the study was great.” [1025]
	“No, I liked everything. All of it.” [2016]
Not getting prize draws (n=5)	“Losing draws due to my drinking.” [1053]
	“When I didn’t win prizes.” [1023]
	“Losing, not getting prizes.” [2038]
Questionnaires (n=1)	“There’s a lot of the questions that made me depressed because it brought up old bad things and behavior.” [1067]
Perceived accuracy of EtG tests
Theme Domain	Example
Accurate all the time (n=27)	“In my opinion they are more accurate than other tests I’ve taken.” [2025]
	“Well, I didn’t use and they all showed I was negative. You can’t get any more accurate than that.” [2033]
	“They didn’t give false or negative ones, I don’t think.” [2038]
	“I know I drank.” [1077]
	“They were accurate.” [2017]
Accurate most of the time (n=6)	“Well, they were for the most part. We had those two problems.” [1025]
	“Mostly, but it seemed like there was some trouble.” [1043]
	“Yes, except for once.” [1016]
	“Most of them. A couple times there were errors.” [1053]
Concern about false positive (n=6)	“Positive when it wasn’t supposed to.” [1016]
	“Yeah, as soon as I stopped using hand sanitizer.” [2015]
Participating in CM through an agency
Theme Domain	Example
Yes, unequivocally (n=30)	“Oh yeah you bet I would, I get UAs all the time at [participating agency].” [1067]
	“I would absolutely. It would be helpful when I’m ready to quit.” [1031]
	“Yes, because if I got gift cards I could buy more vegetables and treats for my dog and cat and Propel water.” [1023]
	“Yeah, why not?” [1036]
	“Oh sure, especially if you’re on SSI or food stamps there are a lot of items you can’t get. So it’s a big help to get soap and stuff here.” [2015]
Confidentiality concerns (n=2)	“Yeah, if it was 100% confidential.” [1053]
	“I would think about it. It would be through the agency and I might not want them to know a lot about it [the UA results].” [2020]
Depends on the agency (n=4)	“Maybe, depends on who’s running it.” [1004]
	“Maybe if it weren’t too judgmental.” [1043]

The third question asked participants’ their view regarding accuracy of the EtG tests to determine alcohol abstinence, and subsequently whether prize draws were earned at each study visit. Almost all (n=34, 97%) participants specifically stated that they believed the tests were accurate all or most of the time, with 27 (77%) participants endorsing that the tests were always accurate and six (17%) stating they were accurate almost every time. Six (17%) participants believed that at least one of their EtG tests displayed a false positive (see Table 2). We reviewed the EtG results for these participants and did not find evidence of false positives (a false positive is defined as an EtG-positive sample occurring when there has been no self-reported drinking for at least five days prior).

The fourth question examined participants’ willingness to enroll in CM as a treatment if it was offered as a clinical program, rather than as part of a research study. Nearly all (86%; n=30) participants stated that they were open to participating in CM. One participant replied, “Oh sure, especially if you’re on SSI or food stamps there are a lot of items you can’t get. So it’s a big help to get soap and stuff here” [2015]. However, two (6%) gave the caveat that confidentiality of urine tests results would need to be guaranteed, and four (11%) qualified that their participation would be dependent on the particular agency that was offering the CM program (see Table 2).

Discussion

Overall, participants in the CM intervention found their participation to be a positive experience. The most-cited favored aspect of the study was earning prizes and gift cards. The variety of prizes available, ranging from practical items to prizes that had more personal entertainment value, could have positively impacted the attractiveness of this particular CM intervention. Another frequently occurring theme was that the study helped participants feel accountable and aware of their drinking, and experience a positive change in their thoughts or treatment perspective. These cognitive benefits are an undoubtedly beneficial byproduct of CM, and have been demonstrated in prior research (Petry, Alessi, & Ledgerwood, 2012a; Petry, Alessi, & Ledgerwood, 2012b; Petry, Alessi, Ledgerwood, & Sierra, 2010). Most participants also enjoyed interacting with study staff, indicating the importance of respectful, non-judgmental interaction. While most participants liked the study without qualification, some did not like study visits when they tested positive for alcohol and thus did not earn prize draws, and one participant found the interview questions to be unpleasant. Most participants thought the EtG tests used to verify alcohol abstinence were sufficiently accurate; however, a few felt that they experienced some false positive tests. While the perception and reality of false positives during the study were low, the use of a slightly higher cutoff level (for example, 300ng/mL as opposed to 100ng/mL used in this study) may help to alleviate these concerns and possible rare false positive occurrences. All participants said they would consider participating in a CM intervention if offered by an agency, though some noted that it would depend on the agency and if confidentiality of test results could be guaranteed.

Findings support two insights into the impact of a positive reinforcement behavioral intervention for individuals with the lived experience of SMI and alcohol dependence. First, the unanimous appreciation for prizes indicates the importance of tangible reinforcers for adults receiving treatment at an urban community mental health setting. As most individuals in the study were not employed, the intrinsic reward of earning prizes for either alcohol-abstinent urine samples or attendance was psychologically motivating and rewarding. Second, the positive feedback to interpersonal interactions suggests that participants valued the non-judgmental and encouraging support offered by research staff. This suggests that interactions with those delivering CM might be another source of motivation and support for those receiving CM. Consistent and brief positive interactions with study staff might be particularly important for those with co-occurring disorders, many of whom have interacted with punitive, over-burdened, and reactionary systems (e.g. the justice system, crisis services, community mental health and addiction treatment agencies). As consumers continue to move through these systems due to the often chronic nature of co-occurring SMI and alcohol use disorders, CM could offer an important opportunity for low-barrier, rewards-based behavior change. It also offers a unique opportunity for consumers to build ongoing relationships with the staff administering CM outside of the typical provider/consumer paradigm.

Limitations of this study include the post hoc addition of the qualitative interview questions. The initial study design did not include collection of qualitative data. These questions were added because it was determined qualitative data might add insight into the acceptance and feasibility of the intervention. Future studies could conduct interviews throughout the treatment period, rather than following it, to explore whether participants looked forward to study visits, how research staff could have facilitated a better experience, and participants’ reasons for missing visits despite feeling positively about the CM intervention. Future studies could also assess changes in perceptions of CM by administering interviews at multiple timepoints. Similarly, it could be informative to interview individuals who were not willing to participate in a CM intervention to explore negative attitudes and possible misconceptions about CM. Furthermore, the semi-structured interview questions presented in this study focused disproportionately on potentially positive aspects of CM. Future interviews should be structured with more open-ended questions to provide participants with substantive opportunities to express negative feedback on CM. An additional limitation was that because interviews were conducted with participants following the final study visit, participants who were absent from this visit did not have the opportunity to take part in the interview. While the results include data from 35 participants in the CM condition, there were five additional participants in the CM condition who were not able to be interviewed due to loss of contact. If these participants who dropped out of the study had negative views of CM, the remaining results could be somewhat positively biased. Finally, as with much of qualitative research, interviewer bias or demand characteristics may have affected participants’ responses during the interview. This possibility was hopefully lessened by the research team’s background in qualitative data collection with this population. Interviews were conducted at the end of the CM phase of the study, so there were no further opportunities to earn rewards, eliminating a situation that could have pressured participants toward a positive bias. Because participants and research staff developed a comfortable, open rapport over the course of the 16-week study, these interviews are likely to represent participants’ honest feedback.

Conclusion

This study provides a unique perspective on CM for alcohol use from consumers with SMI and alcohol dependence. The overwhelmingly positive response to CM in this population is encouraging and speaks to the acceptability of CM as a possible treatment option for alcohol use among individuals with SMI. The results also highlight that consumers trust the accuracy of the EtG biomarker to assess alcohol abstinence. The qualitative data support the positive outcomes obtained in the study and indicate that participants like both the tangible rewards provided in the study as well as the positive interpersonal interactions they received from study staff. Given the findings, coupled with earlier work showing that CM is an effective low-cost (Murphy, et al., 2015), easily-administered intervention, pragmatic community behavioral health agencies should be encouraged to include CM in their repertoire of evidence-based practice for individuals with co-occurring SMI and alcohol use disorders.

Impact

Adults with serious mental illnesses experience unique challenges in substance abuse treatment. Contingency management is one of few evidence-based behavioral interventions for substance use that has shown promising results for this population. The perspectives of adults with serious mental illnesses and on contingency management for substance abuse has been only minimally explored. This study offers a more extensive qualitative look at the experiences of adults with serious mental illness and alcohol dependence engaging in a contingency management intervention for alcohol use, and presents findings in the participants’ own words. Contingency management was well-received by participants in this study, indicating this intervention may be appealing and acceptable to consumers in clinical practice, as well as empirically valid in research settings.