Discounting of employment opportunities with urine drug testing requirements in opioid users enrolled in the Therapeutic Workplace

1 Introduction

A recent estimate of the cost of opioid use disorder (OUD) and fatal opioid use disorder in the United States is over $1 trillion per year (CDC, 2021). Of these costs, it is estimated that $35 billion is spent on healthcare costs, $14.8 billion on criminal justice costs, and $92 billion in lost productivity (Florence et al., 2021). Fortunately, interventions such as contingency management (CM) effectively treat opioid use disorder and minimize the financial burden of substance use disorders. CM is one of those most robustly effective approaches to treating substance use disorders, including OUD (Dutra et al., 2008). CM interventions for treating OUD vary, but all involve reinforcing treatment successes with financial incentives or other tangible rewards. For example, CM interventions have rewarded adherence to medications such as methadone (e.g., Metrebian et al., 2021), naltrexone (e.g., Caroll et al., 2002; DeFulio et al., 2012), or buprenorphine (e.g., Holtyn et al., 2021) and/or evidence of abstinence from opioids through urine drug tests (e.g., Ling et al., 2013; Jarvis et al., 2019). Among variations of CM for treating OUD is the Therapeutic Workplace (TWP) which addresses the intertwined issues of opiate use and joblessness together by offering paid employment to individuals with OUD contingent on verified treatment success (i.e., medication adherence and/or abstinence).

TWP is a highly effective intervention for treating substance use and promoting long-term abstinence. Through TWP, individuals with substance use disorders are employed to work administrative tasks. Participants’ access to work is contingent on objective evidence of abstinence from drugs and/or verified adherence to treatment medications. This intervention targets the costs to the healthcare system by treating substance use disorders and losses in productivity by employing unemployed drug users. TWP has been used to promote healthy behaviors in a variety of populations, including individuals who use cocaine (e.g., DeFulio et al., 2009; Silverman et al., 2007), opioids (DeFulio et al., 2012), and alcohol (Koffarnus et al., 2011), and among people who are experiencing homelessness (Koffarnus et al., 2011; Silverman et al., 2012). Long-term outcomes of TWP are promising. However, this intervention is sometimes criticized because it requires people who use drugs to voluntarily enroll in a program that requires urine drug testing to access employment.

To address concerns over the endorsement of TWP, the purpose of this experiment was to assess the degree to which urine drug testing decreases the value of employment opportunities among adults with OUD. Behavioral economic choice paradigms allow for the quantitative analysis of the value. For example, discounting procedures are used to evaluate how delay or probability interact with reinforcer magnitude to affect the value of a reward through titrating discrete choice tasks (Odum, 2011; McKercher & Renda, 2012; Rachlin & Jones, 2008). Participants are asked to indicate whether they prefer a small immediate or certain reward over a large delayed, or probabilistic reward. Preference for the smaller reward across a wider range of delays and/or probabilities is correlated with riskier behaviors (e.g., drug use) (Athamneh et al., 2017; Businelle et al., 2010; Kollins, 2003).

In the current experiment, the psychophysical titration approaches to assess relative value of the discounting paradigm were adopted to assess the value of employment opportunities that require urine drug testing. Participants were asked to indicate whether they preferred a non-urine-testing job offered at a wage slightly above the minimum wage they would accept to work any job, or a urine-testing job offered at a range of hourly wages. These procedures were conducted to characterize the value of employment based on pay rate and requirement of urine drug testing in individuals enrolled in the Therapeutic Workplace for treatment of OUD.

2 Methods

3 Results

Participants were assessed at intake, 6 months (end of treatment), and 12 months (6 month follow-up post-treatment). Occasionally, participants were also assessed at other nonsystematic (convenience) timepoints between intake and 6 months, as some participants were already enrolled in the TWP when the study began. Mean indifference points and percent change in indifference point for a job with urine testing were assessed. Results are summarized in Table 3. Participants were considered “indifferent” between urine and non-urine testing jobs if their required wage premium for accepting a job requiring urine testing was less than or equal to 1.25 times the wage for the job not requiring urine testing. About 67% of participants were indifferent or preferred a job with urine-testing. About 36% preferred a urine-testing job to a non-urine testing job, as indicated by willingness to accept a lower wage for a job requiring urine testing. In contrast, about 7% of participants exclusively preferred a job without urine-testing across all assessed wages. All participants who preferred a job without urine-testing across all wages also reported recent cocaine or heroin use.

Percent wage difference to accept a job that required urine testing was analyzed using GEE. Results revealed a significant main effect of recent drug use (χ²(1) = 10.07, p < 0.01), but not timepoint of assessment. Figure 1 displays average indifference points across timepoints with respect to recent drug use. Participants who reported cocaine or heroin use in the past 30 days required higher wages across all timepoints to switch preference to a urine-testing job. Self-reported substance use was compared against monthly urinalysis results. Of the monthly urine samples provided, 81.5% matched self-reported substance use in the past 30 days. However, this number may be conservative, as self-reported drug use was assessed for a wider span of time than was detectable via monthly urinalysis.

OPEN IN VIEWER

Fig. 1

Percent difference in acceptable wage for a job that requires urine testing compared to a job that does not require urine testing in participants with and without recent reported drug use across study timepoints. Note that all participants reported recent drug use at intake.

A vast majority of participants in this study reported willingness to accept a job that requires urine testing if it offered a similar (i.e., no more than 1.25 times greater) hourly wage. Indeed, only 7% of participants indicated that they would be unwilling to accept a job that required urine testing at wages up to 5 times greater than a job not requiring urine testing. For most participants, urine testing did not decrease the value of employment substantially. This suggests that therapeutic interventions, such as TWP, requiring urine drug testing are practical and acceptable for the majority of heroin and cocaine dependent individuals tested. Jobs that require urine drug tests may be an effective method of employing individuals in treatment for stimulant or opioid use disorders, which can address issues of lost productivity and costs to the healthcare system. Employing individuals in treatment for stimulant and opioid use disorders in positions that require urine testing can sustain the therapeutic benefits of CM long-term in the naturalistic environment. Longer durations of CM implementation have been shown to produce better long-term outcomes (Petry et al., 2018; Roll et al., 2013).

It is interesting that participants who reported no recent drug use across timepoints also reported a lower acceptable wage for a job with urine testing. It is possible that participants who found jobs with urine testing acceptable viewed this requirement as a motivating factor. The use of urinalysis in interventions such as TWP may be an important feature of treatment for people in recovery from stimulant or opioid use disorders entering the workforce. Participants who reported recent cocaine and/or heroin use required higher wages to accept a urine-testing job. It may be important to investigate other factors that influence acceptability of urine-testing jobs, such as participants’ motivation to quit using drugs, use severity, or other individual characteristics.

5 Conclusion

A majority of participants with OUD were willing to accept a job that required urine drug-testing across wages similar to jobs that did not require urine drug-testing. Participants who reported recent cocaine or heroin use were less likely to endorse employment requiring urine drug-testing. While recent drug use was related to indifference points, timepoint in the study was not. Further research is needed to investigate the factors that influence acceptability of urine-testing jobs.