The Cost of Recovery

Abstract

This paper describes the economic changes experienced by former Detroit sex workers as they progressed through a court-supervised treatment and recovery program. Semistructured interviews were conducted with women during three phases of the program: pretreatment (n = 31), treatment (n = 24), and transitional housing/independence (n = 31). Interviews were also conducted with women who had terminated from the program (n = 8). Sources and levels of income and expenses in each phase were recorded in detail. We found that women earned and spent substantially more money when they actively engaged in street sex work. Legal income remained low after the treatment phase, with only 39 percent of women in the final phase reporting regular work. We conclude that decreased drug expenditures are not enough to offset the decrease in income that accompanies termination of sex work. Policies supporting income, employment, and education for people in transition are recommended.

Keywords

addiction court drugs economics work/occupation

Introduction

This paper describes the economic changes experienced by former sex workers in Detroit as they progressed through a drug-court-supervised treatment and recovery program. The program required that participants discontinue both their drug use and sex work, thereby causing participants to lose both their largest source of income and their largest expense. The court, as well as treatment professionals, provided counseling to the offenders for both purposes, with the court assigning individual case managers for each client.

Drug courts are problem-solving courts dedicated to keeping drug-addicted offenders in treatment long enough for the treatment to be successful. They were introduced in the United States in the 1990s and are now well established across the country. They are often portrayed as a more compassionate mode of administering justice, focused on rehabilitating the individual rather than simply doling out punishment. A recent meta-analysis of 154 drug court evaluations published in the criminal justice literature found that drug courts reduce recidivism (Mitchel et al. 2012). However, they have been criticized for infusing therapeutic settings with penal-style surveillance, thereby extending the arm of the justice system into peoples’ lives in a way that may be more intrusive than traditional incarceration (Burns and Peyrot 2003; Fischer 2003), although others have argued that some coercive element is necessary for substance abuse treatment programs to be truly effective (Bean 2004). The social science literature has offered a mixed review of drug courts. The criticisms presented in this literature have offered evidence of gender discrimination (Spohn 1999), racial discrimination (Brennan and Spohn 2008), and coercion (Andersen 2001; Inciardi, McBride, and Rivers 1996; MacKenzie 2001; Young and Belenko 2002).

This study of a drug court represents a portion of the findings of a larger study that began as a conversation with administrators of the Detroit health department’s substance abuse and HIV division. The goal of the study, in its entirety, was to use social science research to investigate factors contributing to the success, or failure, of substance abuse treatment programs targeted at marginalized populations. In particular, officials were concerned about the sources of relapse for women who had successfully completed treatment and seemed well on their way to recovery as well as the economic and contextual factors that contributed to successful long-term recovery. We developed a mixed method research proposal designed to capture changes in social networks and economic resources across three phases of court-supervised treatment and recovery. Several previously published papers present further detail on the qualitative and quantitative findings of this larger study (Draus, Roddy, and Asabigi 2015; Draus et al. 2014; Roddy et al. 2013). The findings in this report offer quantitative evidence regarding income reduction due to discontinuation of sex work and expense reduction due to the discontinuation of drug use.

To present this evidence, we examine results from the economic surveys that we conducted with women formerly engaged in street sex work who have undergone substance abuse treatment through a semicoercive program administered by a Detroit drug court. Our focus is on the phases of recovery and their economic correlates. These phases, not the individual women passing through them, are treated as the units of analysis in the study. We analyze patterns of income, savings, and expenditure evident in our sample, within each phase and across the three phases, with the hypothetical expectation that each phase will have a specific structure that contributes to its output. At issue is the specific relationship between the context surrounding individuals and the intended goals of the program, which are to help individuals stop using drugs and to help them become healthy and productive individuals. These goals are not necessarily the same thing. However, both goals are contingent not only on the motivation of individuals but also on the array of resources, networks, and opportunities surrounding them.

The phases in this study are therefore viewed as assemblages, sets of relationships between heterogeneous elements (people, places, and things) that both enable and constrain women’s agency in particular ways (Duff 2011, 2013; Fraser, Moore, and Keane 2014). As explained by Duff (2014:126; emphasis in original),

At issue is how to adequately explain the role of social contexts in shaping AOD [alcohol and other drugs] use. Or, to pose the problem with recourse to a rival vocabulary, how to explain the social and structural determinants that produce consumption (or addiction)? That these forces are active in the expression of AOD use is rarely contested, even though compelling accounts of the terms of this mediation are as elusive as ever. Yet what if one were to start with the connection, rather than the subject (or its behaviours) as the basic unit of analysis? What if one were to prioritise the analysis of the drug assemblage rather than the drug user?

We seek to employ this “rival vocabulary” of drug use, emphasizing not the work (or the success or failure) of the individual subject or agent in achieving desistance or recovery, but rather the components of the assemblage or cluster and its contribution to the outcomes of drug use, drug use cessation, income, savings, and expenditure.

In the recovery world, it is often stated that drug users need to hit “rock bottom” before they can become clean. This statement implies that the consequences of continued drug use will eventually become so severe that an individual will voluntarily desist from drug use. However, the process of getting to that point is complex. For one thing, different people have different “bottoms” (see Daniels 2012; Matzger, Kaskutas, and Weisner 2005; Oksanen 2012). The accumulating negative consequences of addiction, by themselves, do not guarantee that recovery will be sustained. The individual must also have positive motivations for becoming clean. In economic terms, the marginal benefit of desistance and abstinence must be greater than the marginal costs of using for an individual drug user to be sufficiently incentivized to stay clean. Substance users may describe this as a process of “weighing the pros and cons” (Cunningham et al. 1994).

Yet desistance from active drug use is not an easy path, and the desire to be clean may not be enough. Something must support the individual on the other side. Cost-benefit analysis is a methodology that economists use to judge or recommend alternatives and choices for rational agents. Opportunities that yield higher benefits than costs are said to have a positive impact on welfare and represent a rational choice. The benefits of recovery, especially for those who have been immersed in a street-oriented lifestyle, may include access to increased legal income (Richardson et al. 2012). Legal income sources in turn allow people in recovery to avoid typical black-market street activities associated with drug use and criminal life (Blankertz et al. 2005). In addition, employment offers structure that often mediates vulnerability to relapse (Castellani et al. 1997).

Finally, a recovering user may experience reduced drug expenditures, further increasing the net benefits of recovery for the individual. However, most active users are not without some income, even when they hit “rock bottom.” Any income generated from activities related to active substance use will also decline when a person enters treatment. Sometimes, this loss can be substantial. To truly estimate the economic incentives for recovery and the real costs of recovery, the amount of income earned in association with drug use must be taken into consideration. This income must then be compared with the individual’s income and expenditures in the treatment and recovery phases. Public policy that seeks to address substance use in an effective way must consider the true costs of recovery as well as the costs of active addiction for individuals engaged in this difficult transition.

The tacit assumption is that lower levels of drug use, combined with higher levels of income and savings and more diverse networks and opportunities, will yield a net benefit that incentivizes abstinence. In this paper, we do not seek to assess quality of life directly, although in economics, a positive increase in resources would typically imply a higher quality of life. Instead, we examine the measurable correlates of each phase in terms of economic resources and expenditures. Following a description of our methodology and sample, we report our empirical findings and discuss their implications in terms of enhanced substance abuse treatment programs and broader policy.

Description of the Program

This study examines the economic profiles of street sex workers enrolled in a Detroit-area substance abuse treatment program for women engaged in street sex work. The treatment program was overseen by a drug court judge and used negative and positive incentives such as increased or decreased jail time to increase treatment adherence and reduce relapse. For this reason, we describe the program as semicoercive: although it was technically voluntary, it operated in direct contrast to traditional voluntary treatment programs.

As described above, the program under study offered treatment and support services in conjunction with intensive supervision through a local drug court. For women who have been incarcerated and particularly for those who have engaged in sex work, the provision of housing and services upon release from jail may provide a unique opportunity to influence outcomes in a positive way (McLean, Robarge, and Sherman 2006). The stated goals of the program considered here included removing women from active street sex work, providing treatment for substance dependency, and fostering engagement in the Detroit-area recovery community. Women (and some transgender individuals) were arrested in periodic sweeps of neighborhoods where street sex work was regularly reported. (We have discussed the geography of these street sex work zones in a previous publication, so we do not focus on them here.) The program not only provided an alternative to extended jail time for the participants but also imposed requirements on them as a result of their participation. These requirements could include volunteer work and family reconciliation, as well as extended residential treatment and regular drug testing. However, women in the program could also receive needed medical treatment, child care, employment training, and support for continuing education.

The program had a three-phase structure. In the first phase, women were arrested in planned sweeps conducted by the Detroit Police Department on Friday evenings. They were subsequently detained in a county jail in downtown Detroit and were given a weekend to detoxify before appearing in the drug court on Monday morning. Women were diverted from regular criminal court to the 36th District Drug Treatment Court, where the program was formally offered to them by the judge. They were then sent back to the jail for a 30-day stay, during which time they had the opportunity to consider the program. If they chose not to participate in the program, they were levied a $500 fine and could be sentenced by the judge to 90 days in jail for every outstanding ticket. It is therefore evident that although they did not have to participate in the program, choosing not to do so could carry a heavy price.

During this 30-day stay, the particulars of the program were explained to the women, and they began counseling for their substance use. (This jail time was later reduced to less than 30 days because of budget and space constraints within the Wayne County jails.) Phase 2 began when the women were assigned to a residential treatment program immediately following the jail time. After successful completion of residential treatment, the women were assigned one to two years in transitional housing; this was Phase 3. At some point in the process (at the discretion of the judge), the women transitioned to independent housing, depending on employment status, preference, and opportunity.

Method and Sample

As noted above, the program under study was designed for women having a history of both substance abuse and sex work. The purpose of this paper is to examine the economic factors that might contribute to relapse or recovery. The pool of participants for this research was limited to those who engaged with the program following the initial “sweep” stage. The overwhelming majority of the program enrollees were women; however, we encountered three male-to-female transgender individuals and obtained interviews with two. Ninety-four total interviews were obtained. For theoretical purposes (discussed above) and because of the time constraints of the funding mechanism (two years), we decided to focus on phases, rather than individual women, as our major unit of analysis. The data sample is therefore cross-sectional, not longitudinal. Most women were not interviewed multiple times as they transitioned through the phases of the program. However, six women were interviewed twice, in different phases, because their progress in the program coincided with the duration of the study. In such cases, we required at least six months between interviews of an individual.

All participants, in all phases, were interviewed in a private setting with no program or correctional employees present. The interviews were conducted according to a structured format and were recorded on paper. Full institutional review board approval was obtained in November 2009. Written informed consent was obtained, and the protections and limitations of the certificate of confidentiality were disclosed during that process. The first interview took place during December 2009, and the last interview took place in August 2011.

The researchers learned of new arrests and potential program enrollees by attending the 36th District Drug Treatment Court staff meetings and attending court sessions. For Phase 1 interviews, the researchers (first and second authors) would arrive at the jail in the morning hours, several days after a sweep had occurred, and request to speak with specific women, one by one. The researchers then explained what the study was about and offered them the opportunity to participate, explaining that it was completely voluntary and carried no positive or negative effects in terms of their status in the program. However, participants were offered $35 as compensation for their time, which would be placed on their commissary account following completion of the interview. The interviews, including the consent process, typically required 90 minutes. Although the participation rate was not tracked, very few women refused to participate. Several interviews were interrupted by the business of the jail and were completed on subsequent visits. However, the money was always deposited in the commissary on the first day.

Interviews in Phase 2 were conducted in various treatment facilities. Women were typically scheduled for 30 to 60 days of residential substance abuse treatment, though some women were granted more time, and others were released early. The names of women in the program and the treatment facilities where they resided were acquired from a list kept by the 36th District Drug Treatment Court that was distributed at the monthly staff meeting. Women in this phase were also compensated for their interviews in cash ($35).

Interviews in Phase 3 were conducted with women who had graduated from residential treatment and were housed independently, usually within the participant’s place of residence or at the Detroit Department of Health and Wellness Promotion offices, located in a large and recognizable public building that was seen as a safe and familiar place by many residents of Detroit (the building is now closed). Some participants were approaching graduation by the time of this interview and had spent 12 to 18 months in the program. Court oversight and accompanying support typically decreased in this period. Women were compensated $35 in cash for participation. The same qualitative and quantitative instruments used in previous phases were applied in this phase. Although the qualitative findings are not discussed in this paper, the methods used and findings from the qualitative interviews and the combined qualitative/quantitative results are offered in other publications (Draus et al. 2015; Draus et al. 2014; Roddy et al. 2013). Finally, additional interviews were conducted with women who had terminated from the program prior to graduation (n = 8). Most of these women were interviewed in private offices of the Detroit Department of Health and Wellness Promotion, and they were also compensated with cash ($35). We refer to these participants as being in Phase X.

Measures

This report offers descriptive results from the quantitative portion of the interviews, which focused on the economic aspects of each participant’s drug use and sex work. Answers to questions about the drug of choice, purchasing methods, and smoking and injection habits were recorded. The participant’s income and expenses over the last 30 days were also detailed, with an effort made for totals to be within $100 of each other unless savings was reported, which was rare. In Phase 1, participants reported on their levels of income and expenditure in the 30-day period prior to entering the program. This information therefore reflected their income from sex work and expenditures related to drug use, as well as other types of income and expenditure. Categories of both income and expenses allowed for the reporting of illegal activity. The interview instrument was used in previous studies to assess drug seeking, purchasing, and use habits of daily opiate users (Roddy and Greenwald 2009; Roddy, Steinmiller, and Greenwald 2011).

The survey was administered as a structured interview that relied on self-reported data from the participants. Studies examining the reliability of self-reported data for drug use and other risky behaviors cite threats to accurate reporting that include memory failure, social stigma, and inaccurate interpretations of the question. Many studies assess the reliability of self-reported data by postsurvey urinalysis (Calhoun et al. 2000; Napper et al. 2010), and substantial underreporting has been acknowledged (Chen et al. 2006; White, Day, and Maher 2007). However, most studies acknowledge that self-reporting is a practical and cost-effective alternative to direct observation and/or urine testing.

This report focuses on income in Phases 1 and 3. During the first phase, the interview takes place within the jail, when women have recently been removed from the streets. The timing of the interview was meant to enhance the accuracy of the self-reported measures of income, expenditures, and drug use. The survey was also constructed with several checkpoints to increase accuracy. Participants were asked to report both income and expenses. Reported drug use was reexamined if it was deemed unaffordable with the income level reported. In addition, the survey contained cross-checks on drug use and times that drugs are used. Typical daily use was detailed in six-hour increments, and the figures were compared with purchases per day and typical purchases per transaction. Effort was made to assist with accurate recall by reviewing previous responses and relating responses to key events, though no formal or structured timelines were regularly used. In the next section, we report the results of these surveys for each phase.

Results

Descriptive Statistics

Table 1 lists descriptive characteristics for the entire group and by phase in the program. Figure 1 breaks income into categories for each phase; percentages of mean income and participation rates are presented. In Figure 1, income earned from sex work is presented separately from other employment. Figure 2 displays expenditures as a percentage of mean income, along with participation rates.

Table 1.

Descriptive Statistics.

	Total sample mean	Median (first quartile, third quartile)	Phase 1	Phase 2	Phase 3	Participants who terminated
Sample size (n)	94	NA	31	24	31	8
Race (% black)	43.6	NA	36.4	36.4	42.9	100
Age (years)	38.9	40 (33, 45)	34.3^a,b	41.8^a	41.7^b	37
Education (years)	12	13 (11, 13)	11.8	11.8	12.4	11.9
Drug use (% using)	37.2	NA	95.3^a,b	0^a,b	3.2^b	62.5
Number of crack suppliers	4.6	4 (3, 5)	4.6^b	NA	0.1^b	3.3
Distance to supplier (miles)	1.1	0.1 (0.1, 0.5)	0.5	0.2	2.5	2.4
Walk or ride to primary supplier (%)	75	NA	80	NA	0	60
Unit purchase amount ($)	32.33	30 (20, 50)	48.33^b	NA	15.00^b	44.00
Unit cost ($)	11.37	10 (10, 10)	10.10	NA	5.00	20.00
Crack purchases per week (no.)	25	12.3 (0, 38)	37.5^a,b	0^a	0^b	18.1
Opiate purchases per week (no.)	10.8	1 (0, 19)	17.2^a,b	0^a	4^b	6.8
Nonopiate purchases per week (no.)	6.9	0 (0, 3)	11.5^a,b	0^a	4^b	0.6
Crack consumption (units per day)
12:00 a.m.–6:00 a.m.	3.3	1 (0, 5.5)	4.8^a,b	0^a,c	10^b,c	1.7
6:00 a.m.–12:00 p.m.	1.5	0 (0, 2)	2.1^a,b	0^a	0^b	2
12:00 p.m.–6:00 p.m.	2.9	2 (0, 5)	4.2^a,b	0^a	0^b	3.1
6:00 p.m.–12:00 a.m.	3.8	2 (0, 6.5)	5.3^a,b	0^a,c	10^b,c	3.3
Total	11.5	8 (0, 20)	16.4^a,b	0^a,c	20^b,c	10.1
Estimated hourly wage ($)	11.40	10 (7.50, 13.50)	12.22	10.03	12.13	9.66
Past-month income sources ($)
Employment	98.84	0 (0, 0)	9.68^b	20.83^c	273.90^b,c	0
Unemployment insurance	38.96	0 (0, 0)	0	41.67	85.87	0
Public assistance	161.02	200 (0, 200)	32.26^a,b	210.92^a	210.94^b	316.88
Pension/social security	57.15	0 (0, 0)	0^b	38.50	134.84^b	33.50
Family/friends	143.62	62.50 (0, 175)	172.42	136.67	114.55	165.50
Other	2,556.66	0 (0, 0)	6,536.25^a,b	148.81^a,c	178.06^b,c	3,576.37
Total	3,056.25	1,003 (606, 3,439)	6,750.61^a,b	597.40^a,c	998.16^b,c	4,092.25

Statistically significant between Phases 1 and 2, p < .05.

Statistically significant between Phases 1 and 3, p < .05.

Statistically significant between Phases 2 and 3, p < .05.

Figure 1.

Sources of income across three phases.

Figure 2.

Expenses as percentage of income across three phases.

The structured economic interview allowed for phase-based analysis of income, expenses, and time spent seeking, purchasing, and using drugs. The research resulted in 94 interviews, including 31 interviews in Phase 1, 24 interviews in Phase 2, 31 interviews in Phase 3, and eight interviews of participants who had terminated from the program (Phase X). Approximately, 44 percent of the population was black, and the mean age was 39 years. The mean education level was 12 years of schooling; most participants had either graduated from high school or obtained a general equivalency diploma (GED).

Although some demographic differences between the phases of the program were found, only age proved statistically significant between phases. Mean age increased from Phase 1 (34.3 years) to Phase 3 (41.7 years), perhaps reflecting attrition of younger women from the program. However, the time between Phases 1 and 2 was short (one to six months), while the time between Phases 2 and 3 was longer (one year or more), so this difference may also be related to the aging of participants. Although many women began the program with less than a high school education, one of the conditions of participation in the program was earning a GED, although some completed the program without one. Therefore, the higher mean level of education of Phase 3 participants is not surprising. Also unsurprising is the decrease in the percentage of participants using drugs from Phase 1 (95.3 percent) to Phase 3 (3.2 percent).

This analysis focuses primarily on changes in income and sources of income from Phase 1 to Phase 3. Mean income for the entire population, across all phases, was $3,056 per month. The largest consistent source of mean income for the entire sample, in all phases, was “other” income ($2,557/month), which came from sources such as lying, bartering, stealing, credit, gambling, and sex work as well as under-the-table payment for services such as cleaning, hairstyling, and day care. The second largest income category for the entire sample was public assistance ($161/month). Family and friends also contributed income at a mean of $144 per month. Table 1 also reveals the hourly wage that the women estimated they could earn if they were to seek employment, given their current skill set. The reported mean for this variable was $11.40.

Drug Use Patterns

Drug use clearly influenced both income and expenditures in this population. In other words, income to purchase drugs was necessary for drug use to persist. Furthermore, some women reported that sex work was difficult to perform without using drugs. For instance, a participant who left the program reported that she intended to resume only sex work as an income source after recovery but slipped back into drug use. The mean drug purchase amount for the total sample, including all phases, was $32 per transaction, with each unit amount averaging $11. The average number of crack purchases reported per week was 25. Opiate and nonopiate purchases per week were also reported (11 and seven, respectively). Most crack consumption took place between 6 p.m. and 6 a.m. Most women walked or rode in a car to purchase drugs. These figures are relevant to both employment and economic activity: consumption of crack took place during evening and late-night hours when people are less likely to be working, and transportation to procure drugs during this time often incurred additional costs.

Income by Phase

Sources of income changed by phase in the program. During Phase 1, the dominant category was income earned from sex work, which contributed 74 percent of mean monthly income. The rate of participation in sex work during the last 30 days for women in Phase 1 was 87 percent. The second largest contributor to mean monthly income was other income (13 percent), such as lying, bartering, gambling, stealing, credit, and other under-the-table activities. In Phase 1, 55 percent of women reported generating income in this manner.

During Phase 2, public assistance contributed the largest amount to mean monthly income (54 percent), and 96 percent of the women reported this type of income. The second largest category was contributions from family and friends, which represented 27 percent of income; the rate of participation in this category was 79 percent. Participation in sex work dropped to 8 percent and represented a mere 2 percent of income. Other income, reported by 33 percent of women in this phase, contributed 4 percent to mean monthly income. Employment represented 3 percent of income, and only 8 percent of Phase 2 interviewees reported employment income.

During Phase 3, public assistance remained the largest contributor to mean monthly income but contributed a smaller percentage (31 percent). The participation rate also declined to 84 percent. The second largest contributor to income was employment, which represented 24 percent of Phase 3 income, with 39 percent participation. Contributions from family and friends remained high at 17 percent of mean monthly income, with 58 percent of the women receiving this type of transfer. Participation in sex work dropped even further, to 3 percent, and represented 1 percent of mean monthly income. Some women still engaged in exchanges of sex for money, primarily with former clients who remained in contact with them while they were in treatment, though this practice was rare in this sample.

Expenditures by Phase

Self-reported expenditures also revealed changes across phases (Figure 2). During Phase 1, the largest expenditure reported was on illegal drugs, which consumed 67 percent of mean monthly income, with 90 percent of those interviewed reporting drug use during the last 30 days. The second largest expenditure category was alcohol, consuming 15 percent of monthly income; 35 percent of Phase 1 interviewees reported purchasing alcohol.

During Phase 2, the largest expenditure reported was food (48 percent of monthly income). Other expenditures (toiletries, cell phones, transportation) were the second largest expense category in Phase 2 and consumed 31 percent of income. Drug and alcohol expenditures consumed less than 1 percent of mean monthly income. During Phase 3, other expenditures rose to 43 percent of mean monthly income; 97 percent of interviewees had expenditures in this category. Food consumed 27 percent of income. Drug use accounted for 1 percent of income, and 3 percent of the participants reported spending on illegal drugs during this phase.

Transitions by Phase

As expected, the biggest change was seen in spending on drugs from Phase 1 to Phases 2 and 3. Drugs consumed 67 percent of income in Phase 1 but only 1 percent of income in Phase 3. Shelter expenses increased as a percentage of income and in participation from Phase 1 to Phase 3. Food expenses also increase as a percentage of income from Phase 1 to Phase 3. Participation in the food stamp program was facilitated during treatment. In this study, food stamps were recorded both as income and as food expenditure, if the participant reported that the stamps were spent on food. However, some participants reported selling food stamps, and if so, the amount secured was recorded as other income. Regardless, public assistance became a notable portion of income in Phase 2. Participation in spending on cigarettes decreased from Phase 1 to Phase 3.

Expenses changed most dramatically when women transitioned from street life to recovery. Most notably, expenditures on drugs decreased from $4,523 (67 percent × $6,751) to $10 (1 percent × $998) per month, a reduction of $4,513. The loss in income associated with terminating sex work from Phase 1 (74 percent × $6,751 = $4,996) to Phase 3 (1 percent × $998 = $10) was $4,986. The net loss in purchasing power associated with these categories was $473 per month ($4,986 − $4,513), or 47 percent of Phase 3 monthly income. In addition, other expenses increased as the women transitioned to increasing responsibility and independence. Shelter expenses increased from 10 percent to 16 percent, and food expenditures increased from 10 percent to 27 percent of monthly income. Finally, “other” expenses also increased from Phase 1 to Phase 3. These expenses included toiletries, utilities, entertainment, transportation, and clothing. Reported total mean expenses in Phase 1 were $7,898 per month (117 percent of income), whereas total mean expenses in Phase 3 were $968 per month (97 percent of income), or a reduction in mean expenses of $6,930. Reported total mean income in Phase 1 was $6,751 per month, whereas total mean income in Phase 3 was $998 per month, or a total reduction in mean monthly income of $5,753.

The decision to enter treatment therefore represented an 85 percent loss in income ($5,753/month). Although there were gradual gains in employment income, unemployment assistance, and public assistance, there was a substantial loss in income from family and friends, and a dramatic loss in income from sex work and other black-market activity. In summary, regardless of the shifts in other income and expense categories, when transitioning from street life to a life focused on abstinence from drug use and sex work, the women experienced a reduction in income from terminating participation in sex work of $4,986 per month and a reduction in drug expenditures of $4,513 per month. The loss of income exceeded the decrease in expenses directly related to drug use by $473, whereas the gain in income related to employment was only $274 per month.

Conclusion

In other papers (Draus et al. 2015; Draus et al. 2014; Roddy et al. 2013), we have examined the process of developing a collaborative research partnership to conduct this research, the street environments described by women exiting sex work, and the adjustment of women to treatment regimes. One conclusion shared by all of these papers is that resource constraints are a dominant factor shaping the behavior and choices of these women in each phase. This study therefore focuses specifically on the self-reported economic costs and benefits experienced by participants in this recovery program, with the goal of clarifying the real changes in resources that are experienced by women undergoing this transition.

For a person who meets the clinical criteria for addiction, the decision to abstain from drug use is not easy, even though the benefits of abstention may seem obvious from the outside. Recovery from persistent drug use and termination of sex work activities both come at a cost. In this paper, we have examined the limited net financial gain as well as the financial loss that is experienced by women in different phases of a recovery program targeted at street sex workers. If the program is examined simply from the standpoint of the behaviors targeted for termination (drug use and sex work), participants experience a mean financial loss of $473.

Although this analysis falls short of a full cost-benefit analysis, it establishes some measurable economic variables that could be considered in a more robust analysis. Estimates of the health benefits of recovery, the costs of establishing new social networks, and the costs or benefits of entering regular legal employment and the structure associated with this change should be made to better understand the motivating factors facing an individual contemplating recovery. Accurate probability estimates of each event, including the likelihood of employment, would support informed decision making for both program designers and participants.

When asked to rank statements about why they began to participate in sex work, most participants (n = 75) reported “limited employment opportunities” and “amount of money earned” among their top three reasons for initiating sex work. The quantitative survey did not request information on initiation of drug use; however, narrative accounts of trajectories into both sex work and drug use were recorded in qualitative interviews. Many participants’ drug use began with experimentation during adolescence and was accompanied by increasing social and geographic marginality and decreasing economic opportunity, which contributed to their eventual entry into street sex work. This collaboration of factors is reflective of findings in numerous studies of sex work across varying cultural contexts (Goldenberg et al. 2011; Sanchez 1997; Simic and Rhodes 2009). The geographic targeting of women engaged in street sex work, followed by their spatial and social separation from the street environment, may be seen as attempts to overcome this complex entanglement of contextual factors and individual behaviors. However, this separation from specific “risk environments” (Rhodes 2009) must also be accompanied by the strengthening of both social and economic supports, contributing to what Duff (2010:338) calls “enabling” environments, those which “facilitate the production and circulation of those resources necessary for the realisation of specific enabling practices and processes.”

In the program that we studied, progress through the three phases was meant to address some of these environmental factors while also providing individually focused treatment for substance abuse. Strengthening the economic autonomy of program participants through education and employment was therefore a major goal, and economic resources clearly shifted across the three phases. Participation in regular (legal) employment was recorded at 39 percent during Phase 3, compared with 3 percent in Phase 1 and 8 percent in Phase 2. At the same time, overall monthly income plunged from more than $6,000 in Phase 1 to less than $600 in Phase 2, before increasing again to almost $1,000 in Phase 3. Although some of this rebound in income was due to employment, much of it stemmed from increased access to public assistance as well as Social Security payments. Therefore, connecting participants to public programs for which they were eligible was also a notable accomplishment evident in Phase 2, when average income from public assistance jumped from around $32 to $210; income from this source continued to rise into Phase 3. Abstinence from illicit drug use also reduced overall expenditures substantially, though it is worth noting that this drop in spending was exceeded by the drop in income associated with curtailment of sex work and that other expenses, such as food and housing, increased.

Of course, these changes in income cannot be considered apart from the social changes that also occurred, and we must therefore consider whether these losses in income were accompanied by other offsetting gains. Laudet (2011) asserts that quality-of-life measures may enhance recovery studies by providing another index of success aside from substance use abstinence. In addition, Zschau et al. (2015) suggest that drug court participants are limited in their access to networks outside of their therapeutic communities and that this situation might restrict social capital accumulation. Although we did not directly assess quality of life or other human capital accumulation in our study, we might presume that legal employment and the termination of sex work would enhance both (Hsiao et al. 2015).

However, incomes in our sample remained very low; in Phase 3, average incomes were below the poverty threshold of $11,702 for a single person in 2011 (U.S. Census Bureau 2014). This figure assumes that these women were supporting only themselves, whereas many participants had children. To increase success rates for former street sex workers in programs such as this one, care should therefore be taken to address the income limitations that will likely remain years after participants have graduated. As with other social service programs linked to drug courts, the loss of autonomy suggested by the use of the state’s coercive power may be outweighed by the eventual enhanced autonomy provided by additional resources and a multidimensional approach (Holsapple and Jensen 2013).

Research also suggests that recovery programs must address social supports as well as drug use behaviors to be successful (Hiller et al. 2013). Although we do not analyze these results here, our study also included social network measures and qualitative questions concerning daily routines and associations (see discussion in Draus et al. 2015; Draus et al. 2014). If we view substance abuse recovery programs as “assemblages of health” (Duff 2014), we would expect to see passage through the different phases associated with changes in these objective social and economic correlates: not just reducing substance use and removing women from sex work but also replacing those activities with alternatives that meet the same material needs in terms of resources while also providing nonmaterial benefits such as social support, a sense of meaning, and enhanced agency.

Limitations and Future Research

Our goal in this paper was not to debate the merits of the drug court model or the goals of programs intended to reduce participation in sex work. As noted, this study is also not a full cost-benefit analysis. For participants, the benefits of abstinence from both sex work and drug use include decreased exposure to violence, a healthier lifestyle with associated health benefits, integration with family, and decreased anxiety associated with termination of criminal activity. This study report is limited in that it examines the two behavioral categories that the program intends to influence (sex work and drug use) and their direct economic impact. Although the differences in total income from sex work and total expenditures on crack were substantial between Phases 1 and 3, this result is a natural consequence of the coercive nature of the residential treatment and supervised housing that was a major component of the drug court program. In other words, it is hardly surprising that the data show substantial differences in sex work income and drug expenditure between phases. A more nuanced study examining a more extensive inventory of costs and benefits as well as quality-of-life measures (e.g., wellness, life satisfaction, safety, social health, and finances) is in order.

A thorough cost-benefit analysis from a societal perspective would quantify savings associated with decreased law enforcement and judicial costs that could offset increased job training expenditures and other costs of recovery programs. No matter how much government, law enforcement, and treatment advocates try to convince individuals that abstinence is in their own best interest, society relies on rational decisionmakers and allows for choice. Individuals must be sufficiently incentivized for the theorized societal benefits of a recovery program to be realized.

Footnotes

Acknowledgements

We thank NIDA for the support. In addition, we thank Erin Jackson, University of Michigan–Dearborn, Master of Public Policy student, for data analysis.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by National Institutes of Health National Institute on Drug Abuse (NIDA) R21 DA127045.

Author Biographies

Juliette Roddy is an economist at the University of Michigan-Dearborn. She earned her PhD from Wayne State University in 2002. Dr. Roddy investigates the economics of substance use/misue, recovery and re-entry within the city of Detroit. Her most recent research examines the impact of Michigan’s medical marijuana legislation on marijuana consumers. Dr. Roddy is an enrolled member of the Sault Ste. Marie Tribe of Chippewa Indians.

Paul Draus is sociologist at the University of Michigan-Dearborn and currently serves as Director of the Master of Public Administration and Public Policy programs. He earned his PhD from Loyola University Chicago in 2001. From 1992 until 2000 he was a public health field worker, specializing in tuberculosis control, first in New York City and then in Chicago. He is the author of Consumed in the City: Observing Tuberculosis at Century’s End (Temple University Press, 2004), which was based on his dissertation research. He has published numerous articles on health behaviors and social contexts related to substance abuse in rural areas and social networks and work patterns of daily heroin users, former street sex workers and ex-offenders in Detroit. His most recent research focuses on the dynamics of change in neighborhood environments.

Kanzoni Asabigi is the associate director of Public Health for the City of Detroit Health Department. He is Certified in Public Health (CPH) and a Diplomate of the American Board of Quality Assurance and Utilization Review Physicians. He received his PhD (Community Health), from Walden University in 2009, Masters of Business Administration (MBA) from the University of South Florida in 2000, specializing in Health Care Administration, Master of Public Health (MPH) in Epidemiology from the University of Michigan in 1989, and Doctor of Medicine (MD) in 1985 from the University of Bucharest Medical School, Romania. His recent focus deals with improving and eliminating health disparities in the city of Detroit, emphasizing the need to consider Behavioral Health disorders as a major public health crisis.

References

Andersen

2001. “What to Do about ‘Much Ado’ about Drug Courts?” International Journal of Drug Policy 12(5–6):469–75.

Bean

2004. “Linking Treatment Service to the Criminal Justice System.” Pp. 219–35 in Drug Treatment: What Works? edited by Bean

Nemitz

London, England: Routledge.

Blankertz

Spinelli

Magura

Bali

Madison

Staines

Horowitz

Guarino

Grandy

Fong

Gomez

Dimun

Friedman

2005. “Integrating Fieldwork into Employment Counseling for Methadone-treatment Patients.” Journal of Employment Counseling 42(3):113–24.

Brennan

Spohn

2008. “Empirical Research on the Impact of Sentencing Reforms: Recent Studies of State and Federal Sentencing Innovations.” Journal of Contemporary Criminal Justice 24(4):340–44.

Burns

S. L.

Peyrot

2003. “Tough Love: Nurturing and Coercing Responsibility and Recovery in California Drug Courts.” Social Problems 50(3):416–38.

Calhoun

P. S.

Sampson

W. S.

Bosworth

H. B.

Feldman

M. E.

Kirby

A. C.

Hertzberg

M. A.

Wampler

T. P.

Tate-Williams

Moore

S. D.

Beckham

J. C.

2000. “Drug Use and Validity of Substance Use Self-reports in Veterans Seeking Help for Posttraumatic Stress Disorder.” Journal of Consulting and Clinical Psychology 68(5):923–27.

Castellani

Wedgeworth

Wootton

Rugle

1997. “A Bi-directional Theory of Addiction: Examining Coping and the Factors Related to Substance Relapse.” Addictive Behaviors 2(1):139–44.

Chen

W. J.

Fang

C. C.

Shyu

R. S.

Lin

K. C.

2006. “Underreporting of Illicit Drug Use by Patients at Emergency Departments as Revealed by Two-tiered Urinalysis.” Addictive Behaviors 31(12):2304–2308.

Cunningham

J. A.

Sobell

L. C.

Sobell

M. B.

Gaskin

1994. “Alcohol and Drug Abusers’ Reasons for Seeking Treatment.” Addictive Behaviors 19(6):691–96.

10.

Daniels

2012. “Intervention: Reality TV, Whiteness and Narratives of Addiction.” Pp. 103–25 in Advances in Medical Sociology. Vol. 14, Critical Perspectives on Addiction, edited by Netherland

Bingley, England: Emerald.

11.

Draus

Roddy

Asabigi

2015. “Streets, Strolls and Spots: Sex Work, Drug Use and Social Space in Detroit.” International Journal of Drug Policy 26:453–60.

12.

Draus

Roddy

White

Asabigi

2014. “Making Sense of the Transition to Drug Treatment from the Detroit Streets.” Qualitative Health Research 25(2):228–40.

13.

Duff

2010. “Enabling Places and Enabling Resources: New Directions for Harm Reduction Research and Practice.” Drug and Alcohol Review 29:337–44.

14.

Duff

2011. “Reassembling (Social) Contexts: A New Direction for a Sociology of Drugs.” International Journal of Drug Policy 22:404–406.

15.

Duff

2013. “The Social Life of Drugs.” International Journal of Drug Policy 24:167–72.

16.

Duff

2014. Assemblages of Health: Deleuze’s Empiricism and the Ethology of Life. New York: Springer.

17.

Fischer

2003. “‘Doing Good with a Vengeance’: A Critical Assessment of the Practices, Effects and Implications of Drug Treatment Courts in North America.” Criminal Justice 3(3):227–48.

18.

Fraser

Moore

Keane

2014. Habits: Remaking Addiction. Basingstoke, England: Palgrave MacMillan.

19.

Goldenberg

S. M.

Strathdee

S. A.

Gallardo

Rhodes

Wagner

K. D.

Patterson

T. L.

2011. “‘Over Here, It’s Just Drugs, Women and All the Madness’: The HIV Risk Environment of Clients of Female Sex Workers in Tijuana, Mexico.” Social Science & Medicine 72:1185–92.

20.

Hiller

S. P.

Syvertson

J. L.

Lozada

Ojeda

V. D.

2013. “Social Support and Recovery among Mexican Female Sex Workers Who Inject Drugs.” Journal of Substance Abuse Treatment 45:44–54.

21.

Holsapple

Jensen

2013. “A Model Program for Patient Navigation: Using the Justice System to Offer a Health Care Intervention to Improve Birth Outcomes—Understanding Constrained Choice and Drug Treatment Court.” Journal of Applied Social Science 8(1):41–60.

22.

Hsiao

C. Y.

Chen

K. C.

Lee

L. T.

Tsai

H. C.

Chang

W. H.

Lee

I. H.

Chen

P. S.

R. B.

Yang

Y. K.

2015. “The Reductions in Monetary Cost and Gains in Productivity with Methadone Maintenance Treatment: One Year Follow-Up.” Psychiatry Research 225(3):673–79.

23.

Inciardi

McBride

Rivers

1996. Drug Control and the Courts. Thousand Oaks, CA: Sage Publications.

24.

Laudet

A. B.

2011. “The Case of Considering Quality of Life in Addiction Research and Clinical Practice.” Addiction Science Clinical Practice 8(1):44–55.

25.

MacKenzie

2001. “Corrections and Sentencing in the 21st Century: Evidence-based Corrections and Sentencing.” The Prison Journal 81(3):299–312.

26.

Matzger

Kaskutas

Weisner

2005. “Reasons for Drinking Less and Their Relationship to Sustained Remission from Problem Drinking.” Addiction 100(11):1637–46.

27.

McLean

R. L.

Robarge

Sherman

S. G.

2006. “Release from Jail: Moment of Crisis or Window of Opportunity for Female Detainees?” Journal of Urban Health 83(3):382–93.

28.

Mitchel

Wilson

Eggers

MacKenzie

2012. “Assessing the Effectiveness of Drug Courts on Recidivism: A Meta-analytic Review of Traditional and Non-traditional Drug Courts.” Journal of Criminal Justice 40(1):60–71.

29.

Napper

L. E.

Fisher

D. G.

Johnson

M. E.

Wood

M. M.

2010. “The Reliability and Validity of Drug Users’ Self Reports of Amphetamine Use among Primarily Heroin and Cocaine Users.” Addictive Behaviors 35(4):350–54.

30.

Oksanen

2012. “To Hell and Back: Excessive Drug Use, Addiction and the Process of Recovery in Mainstream Rock Autobiographies.” Substance Use & Misuse 47(2):143–54.

31.

Rhodes

2009. “Risk Environments and Drug Harms: A Social Science for Harm Reduction.” International Journal of Drug Policy 20:193–201.

32.

Richardson

Wood

Montaner

Kerr

2012. “Addiction Treatment-related Employment Barriers: The Impact of Methadone Maintenance.” Journal of Substance Abuse Treatment 43(3):276–84.

33.

Roddy

Draus

White

Asabigi

2013. “Moving Parts and Balancing Acts: Building and Maintaining a Collaborative Partnership in Detroit.” Advances in Applied Sociology 3(1):26–36.

34.

Roddy

Greenwald

M. K.

2009. “An Economic Analysis of Income and Expenditures by Heroin-using Research Volunteers.” Substance Use & Misuse 44(11):1503–18.

35.

Roddy

Steinmiller

Greenwald

M. K.

2011. “Heroin Purchasing Is Income and Price Sensitive.” Psychology of Addictive Behaviors 25(2):358–64.

36.

Sanchez

L. E.

1997. “Boundaries of Legitimacy: Sex, Violence, Citizenship and Community in a Local Sex Economy.” Law & Social Inquiry 22(3):543–80.

37.

Simic

Rhodes

2009. “Violence, Dignity and HIV Vulnerability: Street Sex Work in Serbia.” Sociology of Health & Illness 31(1):1–16.

38.

Spohn

1999. “Gender and Sentencing of Drug Offenders: Is Chivalry Lost?” Criminal Justice Policy Review 9(3–4):365–99.

39.

U.S. Census Bureau. 2014. “Poverty Thresholds.” Retrieved July 4, 2015 (http://www.census.gov/hhes/www/poverty/data/threshld/index.html).

40.

White

Day

Maher

2007. “Self Reported Risk Behaviour among Injecting Drug Users: Self versus Assisted Questionnaire Completion.” AIDS Care 19(3):441–47.

41.

Young

Belenko

2002. “Program Retention and Perceived Coercion in Three Models of Mandatory Drug Treatment.” Journal of Drug Issues 32(1):297–328.

42.

Zschau

Collins

Lee

Hatch

2015. “The Hidden Challenge: Limited Recovery Capital of Drug Court Participants’ Support Networks.” Journal of Applied Social Science. Published electronically July 6. doi:10.1177/1936724415589633.