Abstract
Poverty is common among people who have substance use disorder. The therapeutic workplace addresses some of the interrelated and chronic problems of poverty, such as unemployment, lack of education and job skills, and drug use. A prior controlled trial showed that the therapeutic workplace was effective in promoting drug abstinence and self-reported community employment in unemployed adults in medication-assisted treatment for opioid use disorder. The current study extends this research by providing a detailed and objective analysis of employment outcomes using objective data abstracted from participants’ pay stubs. Secondary analyses examined the types and patterns of employment that participants (N=44) obtained during the trial, and the extent to which participants gained and maintained financially sufficient employment. Despite that most participants had relatively long histories of unemployment and underemployment, many participants (n = 26; 59%) obtained employment at some point during the intervention. Most participants worked part time and were employed in low-wage jobs, however. The mean number of hours worked per week was 20.6 hours (range 5.5 to 41.3 hours per week) and the mean hourly pay was $11.00 per hour (range $9.00 to $15.50 per hour). The most common type of employment was in food preparation and serving-related occupations (e.g., waiters and waitresses, restaurant cooks, and fast food counter workers). Many participants (n = 17; 65%) maintained employment in these jobs over several weeks, while others (n = 9; 35%) were employed sporadically for short durations. Additional supports may be needed for some chronically unemployed adults with substance use disorder to promote consistent employment in well-paying jobs.
1. Introduction
Poverty is a major determinant of poor health (Bor et al., 2017; Chetty et al., 2016; Stringhini et al., 2017) and is common among people who have substance use disorder (Armstrong, 2007; Henkel, 2011; Williams & Latkin, 2007). Employment could serve as a vehicle to alleviate poverty and its deleterious effects among unemployed adults who have substance use disorder (Silverman, Holtyn, & Jarvis, 2016; Silverman, Holtyn, & Morrison, 2016; Silverman et al., 2019). The therapeutic workplace promotes employment and drug abstinence in adults with long histories of unemployment and drug misuse. Therapeutic workplace participants are hired and paid for engaging in job-skills training (in phase 1) and for performing real jobs (in phase 2). To promote drug abstinence, therapeutic workplace participants must provide drug-negative urine samples to earn maximum pay in phases 1 and 2 (Silverman, 2004; Silverman et al., 2018).
In a randomized controlled trial, we evaluated the effectiveness of the therapeutic workplace in promoting community employment and polydrug abstinence in unemployed adults in opioid use disorder treatment (Holtyn et al., 2020). After a 90-day abstinence initiation and training period, we randomly assigned participants to a control group or an “abstinence-contingent wage supplement” group for one year. Abstinence-contingent wage supplement participants could earn stipends for participating in the therapeutic workplace’s employment services and wage supplements for working in a community job. To obtain maximize pay, participants had to provide opiate- and cocaine-negative urine samples. Abstinence-contingent wage supplement participants were significantly more likely to maintain opiate and cocaine abstinence and to obtain employment during the year-long intervention compared to control participants.
We basd the primary employment outcomes of the controlled trial on participant self-reports of employment. Self-reported outcomes can be subject to socially desirable reporting as well as recall bias. The current paper provides a secondary analysis of employment outcomes using objective data abstracted from participant pay stubs. The analyses are restricted to abstinence-contingent wage supplement participants because these participants were required to submit pay stubs to earn wage supplements; control participants pay stubs were not submitted by control participants. Nevertheless, the present study provides a detailed analysis of the type and patterns of employment Abstinence-Contingent Wage Supplement participants obtained during the trial, and the extent to which participants gained and maintained financially sufficient employment.
2. Material and methods
We conducted this secondary analysis using data collected during a randomized controlled trial. The trial evaluated the effectiveness of the therapeutic workplace in promoting opiate and cocaine abstinence and employment in unemployed adults in treatment for opioid use disorder. We describe the methods pertinent to the current analyses below; the primary outcomes and full methodology have been reported elsewhere (Holtyn et al., 2020; Toegel et al., 2020). The Johns Hopkins Medicine Institutional Review Board approved the study and all participants provided written informed consent.
2.1. Setting and participants
We conducted the clinical trial at the Center for Learning and Health on the Johns Hopkins Bayview Campus in Baltimore, Maryland. Individuals who were 18 years or older, were unemployed, were enrolled in or eligible for methadone or buprenorphine medication-assisted treatment, provided an opioid-positive (i.e., morphine concentrations of >300 ng/ml, methadone concentrations of >300 ng/ml, or buprenorphine glucuronide concentrations of >10 ng/ml) urine sample, self-reported an interest in gaining employment, and lived in or near Baltimore City were eligible for the trial. After a 90-day abstinence initiation and training period (Toegel et al., 2020), we randomly assigned participants to a control group (n=47) or an abstinence-contingent wage supplement group (n=44). We include in the analyses reported here the participants assigned to the abstinence-contingent wage supplement group and their baseline characteristics are shown in Table 1.
Table 1.
Participant (N=44) characteristics at study intake.
| Characteristicsa | Percentage (n) |
|---|---|
| Male | 52 (23) |
| Race | |
| Black | 64 (28) |
| Other | 36 (16) |
| Married | 18 (8) |
| High school diploma or General Education Diploma | 66 (29) |
| Usual employment pattern, past 3 years | |
| Unemployed | 66 (29) |
| Employed – Full time | 20 (9) |
| Employed – Part time | 14 (6) |
| Ever incarcerated | 91 (40) |
| Living in povertyb | 98 (43) |
| Urine sample positivec | |
| Opiates | 41 (18) |
| Cocaine | 52 (23) |
| Opiates or cocaine | 66 (29) |
| Enrolled in medication-assisted treatment | |
| Methadone | 93 (41) |
| Buprenorphine | 7 (3) |
Participant characteristics were collected using the Addiction Severity Index-Lite, except where noted.
Poverty status was calculated using income and age data from the Addiction Severity Index-Lite and 2018 Poverty Thresholds from the US Census Bureau for one person with no related children.
Urine samples were tested using Abbott Alere urine test cups (Model No. I-DX-1147–022) for morphine (opiates) and benzoylecgonine (cocaine).
2.2. Therapeutic workplace procedures
We invited abstinence-contingent wage supplement participants to attend the therapeutic workplace for 4 hours every weekday where they could engage in job-seeking activities. A therapeutic workplace employment specialist was available to help participants obtain and maintain community employment. The employment specialist assisted participants in completing job-seeking activities (e.g., resume development, online job searches, submission of job applications) and offered referrals to community employment services (e.g., workforce centers, career fairs, staffing agencies).
Participants could earn stipends for attending the therapeutic workplace and engaging in job-seeking activities. We divided the stipends into base pay (for hours worked) and performance pay (for engaging in job-seeking activities). Participants could earn up to $8 per hour in base pay and about $2 per hour in performance pay for up to 20 hours per week. When employed, participants could earn wage supplements for all verified (by pay stubs) hours that they worked in a community job. Participants could earn up to $8 per hour in wage supplements for up to 40 hours per week. To earn the maximum amount in base pay and wage supplements, participants had to provide opiate- and cocaine-negative urine samples according to a routine, and then progressively more intermittent, random drug-testing schedule, which we have described in detail previously (Holtyn et al., 2020).
Briefly, we considered urine samples opiate and cocaine negative if the concentration of morphine (opiates) and benzoylecgonine (cocaine) was less than 300 ng/ml and 150 ng/ml, respectively. Initially, participants had to provide urine samples every Monday, Wednesday, and Friday. After 30 days of meeting the abstinence criteria, we reduced the probability that a Monday, Wednesday, or Friday would be selected as a mandatory testing day from 1.0 until we dsignated an average of one of every six (0.17) Mondays, Wednesdays, and Fridays randomly as mandatory days [i.e., each Monday, Wednesday, and Friday had a 1 in 2 (0.50) chance of being selected as a mandatory testing day, then a 1 in 3 (0.33) chance, then a 1 in 4 (0.25) chance, then a 1 in 5 (0.20) chance, and then a 1 in 6 (0.17) chance].
If a participant provided a positive sample on a mandatory day or missed a mandatory sample, then we reset that participant’s base pay or wage supplement to $1 per hour, we reinstated routine mandatory testing (i.e., every Monday, Wednesday, and Friday), and the schedule-thinning process repeated from the beginning. After being reset, we could raise the base pay or wage supplement by $1 per hour (to a maximum of $8 per hour) every day that the participant provided a drug (opiates and cocaine) negative urine sample and worked for at least 5 min.
2.3. Data analyses
We based the outcomes for this secondary analysis on data abstracted from participant pay stubs. We examined employment patterns using the number of hours that participants worked per week in a community job over the year-long intervention. We calculated the number of weeks for participants to obtain employment using their randomization date and the start date of the first community job they obtained during the intervention. We calculated the average hourly pay for participants by averaging the hourly pay rates abstracted from all pay stubs. The employment specialist identified the types of jobs participants obtained using the company name on participants’ pay stubs and participants’ self-reports of job type. To examine outcomes based on gender, we compared employment outcomes for participants who identified as male or female using independent samples t-tests or Fisher’s exact test, as appropriate. We used GraphPad Prism (Version 8; GraphPad Software, Inc; La Jolla, CA) to perform these analyses.
3. Results and discussion
Figure 1 shows employment patterns for each participant during the year-long intervention period. Of the 44 participants, 26 (59%) obtained employment during the intervention. The average time to employment was 16.8 weeks (range 1 to 50 weeks). After obtaining employment, some participants (n = 12; 46%; Participants 44, 43, 42, 40, 39, 38, 35, 33, 32, 30, 25, and 20 in Figure 1) maintained employment at the same job for the remainder of their study participation. However, many of these participants did not work every week after they obtained employment. We considered participants to have maintained employment at the same job for the remainder of their study participation if their first pay stub and their last pay stub submitted at the end of their study participation were from the same job. Many participants (n = 17; 65%) were employed over several weeks while other participants (n = 9; 35%) were employed for short durations.
Figure 1.
Participant employment patterns over the 52-week intervention period. Each row of data points indicates the number of hours a participant worked for a given week. The number of hours worked per week is represented by the shading from white (0 hours worked per week) to black (50 hours worked per week).
Participants who obtained employment worked 21.5 (41%) of the 52 study weeks, on average (range 2 to 50 weeks). Most participants worked part time and were employed in low-wage jobs. The mean number of hours worked per week was 20.6 hours (range 5.5 to 41.3 hours per week) and the mean hourly pay was $11.00 per hour (range $9.00 to $15.50 per hour).
Table 2 shows the types of jobs participants obtained during the study. Most participants who became employed obtained employment in food preparation and serving-related occupations (e.g., waiters and waitresses, restaurant cooks, and fast food counter workers). The next most common jobs were in construction and extraction occupations (e.g., building and highway construction), and production occupations (e.g., assembly line workers); followed by building and grounds cleaning and maintenance occupations (e.g., janitors, housekeeping, and landscaping) and transportation and material moving occupations (e.g., drivers). A few participants obtained employment in healthcare support (e.g., healthcare home aides), sales (e.g., retail salesperson and cashiers), and protective service (e.g., security guard) occupations.
Table 2.
The types of jobs participants obtained during the 52-week intervention. Because some participants may have obtained multiple jobs, the percentage is calculated using the total number of jobs obtained (i.e., 41).
| Job Descriptiona | Percentage (Number) |
|---|---|
| Food Preparation and Serving Related Occupations | 37 (15) |
| Construction and Extraction Occupations | 15 (6) |
| Production Occupations | 15 (6) |
| Building and Grounds Cleaning and Maintenance Occupations | 10 (4) |
| Transportation and Material Moving Occupations | 10 (4) |
| Healthcare Support Occupations | 7 (3) |
| Sales and Related Occupations | 5 (2) |
| Protective Service Occupations | 2 (1) |
Job descriptions are based on the standard occupational classification system of the U.S. Bureau of Labor Statistics.
Table 3 shows employment outcomes for participants who identified as male or female. The only significant difference in employment outcomes between male and female participants was mean hourly pay, in which male participants earned an average of about $1.50 more per hour than female participants.
Table 3.
Employment outcomes for males versus females.
| Male (n=23) | Female (n=21) | P value | |
|---|---|---|---|
| Obtained employment, % (n) | 70 (16) | 48 (10) | .220 |
| Time to employment, weeks, mean (range) | 18.7 (1.0–50.0) | 13.7 (3.0–27.0) | .384 |
| Study weeks worked, mean (range) | 19.3 (2.0–50.0) | 25.1 (14.0–37.0) | .336 |
| Hours worked per week, mean (range) | 21.8 (5.6–36.8) | 18.8 (5.5–41.3) | .477 |
| Hourly pay, $, mean (range) | 11.60 (10.00–15.50) | 10.08 (9.00–12.50) | .034 |
Our results suggest that only some participants obtained and maintained employment. It is possible that participants who obtained employment in low-wage jobs may experience wage progression over time so that they attain higher wages and become financially self-sufficient. For those participants, the provision of abstinence-contingent wage supplements for a longer duration-potentially at a lower incentive magnitude-could encourage job maintenance while simultaneously supporting recovery and preventing relapse. For those participants who did not obtain employment or were employed sporadically, education- and training-focused interventions that seek to establish job skills may be useful in helping these individuals obtain and maintain employment. Providing counselors and other substance use disorder treatment providers with the resources necessary to implement education- and training-focused interventions and abstinence-contingent wage supplements could greatly benefit unemployed adults in substance use disorder treatment.
All participants in this study were enrolled in medication-assisted treatment for opioid use disorder (i.e., methadone or buprenorphine treatment). We do not know the extent to which findings from this study will generalize to individuals receiving other types of substance use disorder treatment. However, the therapeutic workplace has been successful in promoting therapeutic behavior change in diverse study populations (for reviews see Silverman, Holtyn, & Jarvis, 2016; Silverman, Holtyn, & Morrison, 2016; Silverman et al., 2019).
4. Conclusion
The therapeutic workplace can promote community employment among chronically unemployed adults with substance use disorder. Employment among our sample was typically part time and in low-wage jobs. Additional interventions should promote long-term employment in high-paying jobs.
Highlights.
A therapeutic workplace promoted employment in adults with substance use disorder
Employment was typically part-time and in low-wage jobs
Employment was stable for some participants and sporadic for others
Additional supports are needed to promote long-term employment in high-paying jobs
Acknowledgements
We would like to thank Jacqueline Hampton, who recruited participants for this study and conducted assessments; Andrew Rodewald, who managed our therapeutic workplace and urinalysis laboratory; and Calvin Jackson, who monitored participants while in the therapeutic workplace.
Funding
The study was supported by the National Institute on Drug Abuse of the National Institutes of Health under grants R01 DA037314 and T32 DA07209. The content is solely the responsibility of the authors and does not represent the official views of the National Institutes of Health. The National Institutes of Health had no part in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- Armstrong GL (2007). Injection drug users in the United States, 1979–2002: an aging population. Archives of Internal Medicine, 167(2), 166–173. [DOI] [PubMed] [Google Scholar]
- Bor J, Cohen GH, & Galea S (2017). Population health in an era of rising income inequality: USA, 1980–2015. The Lancet, 389(10077), 1475–1490. [DOI] [PubMed] [Google Scholar]
- Chetty R, Stepner M, Abraham S, Lin S, Scuderi B, Turner N, Bergeron A, & Cutler D (2016). The association between income and life expectancy in the United States, 2001–2014. JAMA, 315(16), 1750–1766. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Henkel D (2011). Unemployment and substance use: a review of the literature (1990–2010). Current Drug Abuse Reviews, 4(1), 4–27. [DOI] [PubMed] [Google Scholar]
- Holtyn AF, Toegel F, Subramaniam S, Jarvis BP, Leoutsakos J-M, Fingerhood M, & Silverman K (2020). Abstinence-contingent wage supplements to promote drug abstinence and employment: a randomised controlled trial. Journal of Epidemiology and Community Health, 74(5), 445–452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Silverman K (2004). Exploring the limits and utility of operant conditioning in the treatment of drug addiction. The Behavior Analyst, 27(2), 209–230. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Silverman K, Holtyn AF, & Jarvis BP (2016). A potential role of anti-poverty programs in health promotion. Preventive Medicine, 92, 58–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Silverman K, Holtyn AF, & Morrison R (2016). The therapeutic utility of employment in treating drug addiction: Science to application. Translational Issues in Psychological Science, 2(2), 203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Silverman K, Holtyn AF, & Subramaniam S (2018). Behavior analysts in the war on poverty: Developing an operant antipoverty program. Experimental and Clinical Psychopharmacology, 26(6), 515. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Silverman K, Holtyn AF, & Toegel F (2019). The utility of operant conditioning to address poverty and drug addiction. Perspectives on Behavior Science, 1–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Stringhini S, Carmeli C, Jokela M, Avendaño M, Muennig P, Guida F, Ricceri F, d’Errico A, Barros H, & Bochud M (2017). Socioeconomic status and the 25× 25 risk factors as determinants of premature mortality: a multicohort study and meta-analysis of 1· 7 million men and women. The Lancet, 389(10075), 1229–1237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Toegel F, Holtyn AF, Subramaniam S, & Silverman K (2020). Effects of time-based administration of abstinence reinforcement targeting opiate and cocaine use. Journal of Applied Behavior Analysis. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Williams CT, & Latkin CA (2007). Neighborhood socioeconomic status, personal network attributes, and use of heroin and cocaine. American Journal of Preventive Medicine, 32(6), S203–S210. [DOI] [PMC free article] [PubMed] [Google Scholar]