Journal of Substance Abuse Treatment 21 (2001) 19 – 26

Regular article

A community-based trial of vocational problem-solving to increase employment among methadone patients David A. Zanis, Ph.D.a,b,*, Donna Coviello, Ph.D.a, Arthur I. Alterman, Ph.D.a, Sharon E. Appling, MSW c b

a Department of Psychiatry, University of Pennsylvania, Philadelphia, PA, USA Graduate School of Social Work, Marywood University, Scranton, PA 18509, USA c Crouse Hospital, Syracuse, NY, USA

Received 15 September 2000; received in revised form 7 March 2001; accepted 23 March 2001

Abstract Unemployment remains a common problem among methadone patients. This study examined the effectiveness of the Vocational Problem-Solving Skills (VPSS) intervention to help unemployed methadone patients obtain employment. Methods: 109 patients were randomly assigned to receive 10 sessions of either VPSS (n = 62) or Interpersonal Problem-Solving (IPS) (n = 47). Fourteen counselors from two methadone programs were trained to deliver both the VPSS and IPS counseling sessions. Results: Overall, 93% (101/109) of the patients completed a 6-month follow-up assessment which revealed that 58.6% (34/58) of patients in the VPSS condition were employed, whereas 37.2% (16/43) in the comparison condition worked (chi-square = 4.53, 1 df, p < 0.05). However, a multinomial regression equation found that the VPSS intervention did not significantly contribute to the prediction of employment. Other factors such as longer length of previous work experience, 40 years of age or older, treatment site, and baseline patient motivation to work emerged as significant predictors. Conclusion: Although VPSS alone was not a strong independent predictor of improved employment functioning, improvements in employment functioning were detected at one of the two sites as a function of VPSS. D 2001 Elsevier Science Inc. All rights reserved. Keywords: Methadone; Employment; Drug use

1. Introduction Following drug use stabilization, employment has long been considered an important secondary goal for patients enrolled in methadone maintenance treatment programs (MMTPs) (Dole, Nyswander, & Warner, 1968). Several studies have found moderate positive correlations between increased employment, decreased substance use, and positive social functioning (McLellan, Ball, Rosen, & O’Brien, 1981; Siegal et al., 1996; Zanis, McLellan, & Randall, 1994). Overall, employment rates in MMTPs are low and range from 15% to 44% (French, Dennis, McDougal, Karuntzos, & Hubbard, 1992; Hubbard, Rachal, Craddock, & Cavanaugh, 1984; Platt, 1995). Unfortunately, less than 4% of MMTPs provide specific employment services such as counseling job clubs (Etheridge, Craddock, Dunteman, & Hubbard, 1995). * Corresponding author. Tel.: +1-800-548-4898; fax: +1-570-9614742. E-mail address: [email protected] (D.A. Zanis).

Given the low rates of employment, patients’ desire for employment services, and the association of employment with improved outcomes, employment interventions appear to be important approaches to explore as potential adjunctive therapies to improve patient functioning. French et al. (1992) identified programmatic, structural, and client-level factors that contribute to the low rates of employment among clients enrolled in MMTPs. Program factors explain why employment services are underutilized in MMTPs. Two predominant issues include the lack of available funding to reimburse clinics for providing employment services and the lack of available, standardized, employment interventions to help counselors deliver employment services (Arella, Deren, Randell, & Brewington, 1990; Hall, 1981; Platt & Metzger, 1987). Structural factors include issues such as bias against hiring methadone clients, unavailability of jobs for a particular skill level, and lack of appropriate job-training programs. Patient factors that contribute to low rates of employment include inadequate job skills (e.g., poor reading or mathematics capa-

0740-5472/01/$ – see front matter D 2001 Elsevier Science Inc. All rights reserved. PII: S 0 7 4 0 - 5 4 7 2 ( 0 1 ) 0 0 1 7 7 - 5

20

D.A. Zanis et al. / Journal of Substance Abuse Treatment 21 (2001) 19–26

bilities), competing psychosocial problems (e.g., depression), job-seeking barriers (e.g., transportation), and low motivation to want to obtain employment (Brewington, Arella, Deren, & Randell, 1987; Kidorf, Hollander, King, & Brooner, 1998; Silverman, Chutuape, Bigelow, & Stitzer, 1996; Zanis, Metzger, & McLellan, 1994). One intent of the present study was to train methadone counselors to deliver a ten-session intervention, Vocational Problem-Solving Skills (VPSS), to unemployed clients (Metzger, Platt, Zanis, & Fureman, 1992). Because the methadone counselor participates as an active and important member in the patient’s recovery process, we reasoned that a counselor, if appropriately trained, could be an effective service delivery agent through which unemployed patients could receive individualized vocational counseling services. Thus, a counselor trained in the delivery of vocational services could potentially diminish some structural, programmatic, and client-level barriers associated with unemployment. To this end, our main hypothesis was to evaluate whether unemployed patients randomized to VPSS counseling would improve their employment functioning compared to patients randomized to a time- and attention-controlled comparison activity.

2. Methods 2.1. Subjects A total of 109 patients voluntarily recruited from two MMTPs consented to participate in the randomized controlled clinical trial. To be considered for the project, patients had to meet the following criteria: (1) unemployed or underemployed, defined as working ‘‘under the table’’ less than 10 hours per week; (2) stabilized on methadone and enrolled in the MMTP for a minimum of 3 months; (3) expressed interest and capacity to work at least 20 hours per week; and (4) actively seeking employment as defined by the Bureau of Labor Statistics (BLS, 1994). 2.2. Sites A convenience sample of two community-based methadone programs was selected in an effort to test the intervention in typical community-based settings. Site A had an enrollment of 383 patients, of which 77 (20%) reported current employment. This site employed 10 counselors, of which five voluntarily agreed to participate in the study. All counselors had a minimum of an associate’s degree and at least 1 year of experience as a methadone counselor. Site B enrolled 288 patients, of which 82 (28.5%) were employed. Site B employed nine counselors, of which two had a bachelor’s degree and seven had a master’s degree. All counselors had at least 1 year of experience as a methadone counselor.

2.3. Recruitment and randomization Subjects were randomly assigned to receive either 10 individual counseling sessions of VPSS or to a controlled comparison condition consisting of 10 individual sessions of interpersonal problem-solving (IPS). Since the study emphasis was on evaluating the effectiveness of the VPSS condition, the randomization was structured to produce a 3:2 ratio of experimental to control condition. Sixty-two patients were assigned to the VPSS condition and 47 patients were assigned to the IPS condition. Overall, 71 patients were recruited from Site A, with 37 randomized to the VPSS condition and 34 to the IPS condition. A total of 38 patients were recruited from Site B, with 25 randomized to VPSS and 13 to IPS. 2.4. Intervention conditions All patients, regardless of randomization condition, were required to receive a 30-minute counseling session each week as part of standard treatment services. Each session was designed to focus on generic drug counseling issues (e.g., methadone dose, drug use, attendance in self-help, etc.). This standard counseling session was provided by the same counselor who was trained to deliver the intervention counseling sessions. Vocational Problem-Solving Skills (VPSS) is a cognitive-based intervention designed to assist chronically unemployed patients transition to work (Metzger et al., 1992). All VPSS sessions were designed to be approximately 30 to 60 minutes in length and delivered within a maximum of a 12-week period. The variation in the length of the sessions and 12-week intervention duration was based on past administrations of the intervention and determined to be a reasonable expectation of service delivery. Both patients and counselors were provided with a manual and workbook, outlining each of the sessions. There are five objectives of the VPSS intervention: (1) help patients understand why they want to work; (2) help patients understand how to overcome current barriers to work; (3) set realistic vocational goals; (4) identify realistic resources to help locate job opportunities; and (5) take appropriate actions to obtain work. Interpersonal Problem-Solving (IPS) counseling was selected as the comparison condition because it was theoretically similar to the VPSS, was based on a manualized intervention developed by Platt (1980), and provided an opportunity to direct the nature of counseling to focus drug use. The IPS intervention was designed to be delivered in the same frequency and duration of counseling as the subjects in the experimental condition. The goal of IPS counseling was to help patients develop improved problem-solving skills to either reduce drug use or continue abstinence from drug use. The five objectives of the IPS counseling were to: (1) reduce/eliminate illicit drug use or maintain an abstinence plan; (2) understand the

D.A. Zanis et al. / Journal of Substance Abuse Treatment 21 (2001) 19–26

utility of social supports in recovery; (3) examine successful and unsuccessful efforts at recovery; (4) formulate realistic recovery plans; and (5) engage in planned activities. Please note: All IPS patients who expressed an interest in receiving employment services were referred to one of two city-based employment programs that offer free employment services. 2.5. Counselor participation A total of 14 methadone counselors from the two MMTPs were trained to deliver both the VPSS and IPS sessions. Counselor participation was voluntary, and each was compensated by either a US$100 stipend or with free lunches for their participation in the project. All counselors signed a consent form to participate in the study, giving permission for our research team to assess counselor effectiveness. 2.6. Counselor training Each counselor completed an initial 8-hour training program and received an implementation manual designed to help the counselor deliver the intervention. Prior to implementation of the intervention, all 14 counselors were required to demonstrate an acceptable set of competencies. Counselors were evaluated on a series of hypothetical roleplay situations simulated by two independent facilitators. Upon demonstrated competency of VPSS and IPS skills, counselors were permitted to participate in the study. Additionally, counselors participated in 12 one-hour weekly booster sessions facilitated by the first author as a means to monitor intervention quality and improve service delivery. 2.7. Data collection A trained independent research assistant assessed patients by interviewing them at baseline, biweekly for 12 weeks, and at 6 months postbaseline. Also, independent urine samples were collected during these assessment periods. The following instruments were administered at baseline and follow-up points of the study. 2.7.1. Addiction Severity Index (ASI) The ASI (McLellan et al., 1980) is a semistructured interview designed to collect historical and recent behavioral information regarding seven areas of functioning: medical, employment, alcohol, drug, legal, family, and psychological. The ASI has demonstrated good validity and reliability in samples of methadone patients (McLellan et al., 1985; Kosten, Rounsaville, & Kleber, 1983). The ASI yields seven composite scores ranging from 0 to 1, with higher scores representative of poorer functioning in the specific domain.

21

2.7.2. Vocational/educational assessment (VEA) The VEA is a 48-item interview created by the authors to measure four domains associated with employment functioning: (1) knowledge of employment resources; (2) attitudes and motivation toward employment; (3) employment barriers; and (4) employment-related behavioral outcomes (income, participation in vocational services, number of job interviews, etc.). This nonstandardized interview was created to measure other possible outcomes associated with employment interventions. 2.7.3. Treatment services review (TSR) The TSR, a brief, 5-minute structured interview designed to record the type and frequency of services received during treatment, was administered every 2 weeks for a period of 12 weeks. The TSR has satisfactory validity and reliability within a methadone population and has been used to show relationships between treatment services and patient functioning (McLellan, Arndt, Metzger, Woody, & O’Brien, 1993). The purpose of the TSR was to measure the number of units of service received by the type of service (e.g., employment, drug counseling, etc.). 2.8. Chart review/employment verification Chart reviews were conducted by the project research technician to obtain data on several programmatic variables such as clinic attendance, methadone dose level, attendance, and participation in the intervention conditions. At the 6-month follow-up interview, the research technician verified employment income either through a pay stub or by employer confirmation. Only verifiable employment was included in the analyses. A total of five subjects (5%) reported income that was nonverifiable. None of these persons reported working more than 5 days in the past 30.

3. Results 3.1. Patient characteristics Baseline demographics and level of patient functioning were similar between the two intervention conditions except that patients in the VPSS condition (21%) were more likely than control patients (11%) to have a valid driver’s license (Table 1). Age ranged from 24 to 67, with a mean age of 43.5. The majority of patients were black (61.5%) and most were men (60.5%). Prior to study enrollment, mean unemployment was 16 months (SD = 13). Over 90% of patients received some form of public welfare (Medicaid, food stamps, cash assistance, etc.). The mean amount of monthly cash assistance (US$285), food stamps (US$150), and work income received (US$153) were not different between conditions.

22

D.A. Zanis et al. / Journal of Substance Abuse Treatment 21 (2001) 19–26

Table 1 Baseline demographics Condition Background characteristic Age (mean) Gender Male Race White Black Hispanic High School Diploma Employed (at least 1 day past month) Marketable Skill or Trade Current Driver’s license Receiving Public Welfare Receiving a Pension (VA or SSDI) Marital Status Married Divorced Single Widowed Psychiatric Functioning Previous hospitalization Legal History Incarcerated > 30 days lifetime Currently on probation/parole Engaged in illegal activities for profit within the past 30 days

Experimental Control n = 62 n = 47 43.5 (SD = 7.8)

43.5 (SD = 4.7)

61%

60%

37% 61% 2% 66% 25%

38% 62% 53% 19%

66% 26% 94% 15%

66% 11% 89% 19%

20% 36% 34% 10%

13% 38% 43% 6%

37%

47%

50% 10%

49% 9%

23%

21%

Difference

Within the experimental condition, the total number of employment sessions differed significantly by site (t = 4.7, df = 56, p < 0.01). The mean number of employment sessions in Site B was 8.2 (SD = 2.8) compared to Site A’s 3.4 (SD = 3.8). The primary reasons for not completing the VPSS sessions as identified by patients at the follow-up assessment were continued drug use; psychiatric problems; no longer interested in employment; and the patient became employed. Staff also cited that lack of time to provide services, lack of funding to reimburse employment services, and high caseloads as reasons for not delivering services. 3.4. Six-month findings

p < 0.05

N = 109.

3.2. Pre-enrollment employment barriers The extent of self-identified barriers to obtain work is noted in Table 2. Overall, 11% of clients in the control group reported no employment barriers, compared to 5% in the VPSS group. There were no statistical differences between conditions. Importantly, baseline employment functioning was not statistically different as a function of condition (Table 3). Patients randomized to the experimental condition worked an average of 3.7 days, had an income of US$178 per month, and had a high ASI employment composite score (0.82). Similarly, patients randomized to the control condition worked an average of 4 days, had a monthly income of US$125, and had an ASI employment composite score of 0.86. 3.3. Program attendance and services received Overall, the mean number of employment sessions received by patients was 5.3 (SD = 4.2) in the experimental condition and 0.85 (SD = 1.9) in the control condition. Overall, the methadone counselors provided 96% of the employment services, with only 4% of services reported to be received by a referral site. Patients in the control condition received an average of 5.1 IPS counseling sessions.

Study results are based on the 6-month outcomes of all 101 located patients (93%). Overall, 58/62 (93.5%) VPSS and 43/47 (91.5%) IPS patients completed the assessment. Attrition differences were not detected between groups. At the 6-month follow-up assessment 58.6% (34/58) of the located patients in the experimental condition had been employed one or more days in the past 30 days, whereas only 37.2% (16/43) of the comparison condition had been employed to this extent. For this analysis, employment was defined as working one or more days in the past 30 days. Based on the Yates adjusted chi-square statistic, the employment rate difference between conditions was statistically significant (c2 = 4.53, 1 df, p < 0.05). Because employment functioning can be operationalized in several ways, we examined a series of continuous employment measures (net income, number of days paid for working, and the ASI employment composite score) and found no differences in employment functioning between the two intervention groups at the 6-month follow-up point when controlling for baseline differences. These findings are presented in Table 3. As can be seen, a series of paired t-tests found that there were several significant within-group differences in the experimental condition from baseline to the 6-month follow-up point, but the differences were not Table 2 Barriers that contribute to unemployment by intervention condition Condition Barriers 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

No jobs available Drug abuse too severe Poor work history Unwilling to leave public welfare Police record Inadequate education Poor job skills Transportation Poor motivation Unsure how to look for work No current work barriers

Experimental N = 62

Control N = 47

Significance

52% 47% 35% 27%

57% 36% 23% 26%

ns ns ns ns

26% 29% 24% 23% 21% 19%

23% 19% 15% 21% 21% 17%

ns ns ns ns ns ns

5%

11%

ns

D.A. Zanis et al. / Journal of Substance Abuse Treatment 21 (2001) 19–26

23

Table 3 Analysis of variance of patient functioning by condition and time Experimental

Control

ASI composite scores

Baseline n = 58 Mean

t-Test

6-Month n = 58 Mean

Baseline n = 43 Mean

t = Test

6-Month n = 43 Mean

ANCOVA condition by time

Medical Alcohol Drug Family/Social Legal Psychiatric Employment Days paid for working Employment income Public Assistance Income from family Illegal income Days of illegal activities

0.17 0.10 0.25 0.13 0.08 0.29 0.82 3.70 US$178 US$292 US$73 US$145 3.20

ns * *** ns ns ns * ** ** ** * ns *

0.26 0.06 0.17 0.17 0.07 0.31 0.75 7.60 US$322 US$249 US$20 US$56 1.00

0.20 0.08 0.22 0.18 0.09 0.28 0.86 4.10 US$125 US$289 US$79 US$73 2.80

ns ns ns ns ns ns ns ns ns ns ns ns ns

0.22 0.05 0.19 0.21 0.10 0.31 0.82 5.80 US$264 US$265 US$71 US$30 2.30

ns ns ns ns ns ns ns ns ns ns ns ns ns

* p < 0.05. ** p < 0.01. *** p < 0.001.

statistically significant across condition and time. Although more patients were employed as a function of condition, neither net income levels nor the mean number of days worked in the past month were statistically significant between conditions when controlling for baseline factors. Also, no other differences in the six areas of change measured by the ASI or other factors associated with obtaining employment (e.g., decrease in illegal income) were detected as a function of intervention condition.

for working in the past 30 days), 19 (19.6%) were termed part-time employees (paid for working between 1 and 14 days), and 25 (25.8%) were considered full-time employees (paid for working 15 or more days in the past 30). The sample

Table 4 Multinominal regression of factors that predict employment Number of days employed

3.5. Multinomial regression analysis A series of bivariate analyses were conducted to examine the correlation between potential predictor variables (e.g., intervention condition, gender, race, etc.) and the criterion variable (employed/not employed at least one day in the past 30 days at the 6-month follow-up point). Intercorrelation tables were calculated to identify and then control for possible multicollinearity. Eight variables were then selected that were either demographically important or independently correlated with employment functioning at the 6month follow-up; these were simultaneously entered into a multinomial regression analysis. The variables included: gender (male vs. female); age (>40 vs. 40 or less); patient work history (more than 12 vs. 12 or fewer months employed in the past 3 years); patient motivation (continuous); intervention condition (VPSS vs. IPS); ASI medical composite (continuous); transportation barriers (yes or no); and treatment site (Site A vs. Site B). Because employment functioning can be operationalized in many ways, it was important to examine the factors that predicted different employment patterns. Table 4 details the parameters and parameter estimates of a model predicting three types of employment patterns. At the 6-month followup point, 53 (54.6%) were considered unemployed (not paid

1 to 14 days

15 to 30 days

Exp(B)

Exp(B)

Gender Male Female

2.84 1.00

Age 40 or less >40 Site A B

2.24

1.00 6.79

1.11 – 41.25*

61.94

3.47 – 1107.51*

1.00 5.81

1.34 – 25.12*

14.70

2.46 – 87.87*

4.01 – 317.31*

26.27

1.85 – 373.15*

1.11 – 29.28*

34.66

5.09 – 236.21**

Length of work history >12 months 5.70 12 months or less 1.00 Medical ASI score 1.28 Motivation 1.58

n = 97. * p < 0.05. ** p < 0.01.

CI

0.67 – 11.99

Transportation barrier Yes 35.66 No 1.00

Condition Experimental Control

CI

3.14 1.00

0.38 – 13.23

0.18 – 9.28 1.08 – 2.33*

0.002 2.54

0.00 – 0.19 1.55 – 4.26*

0.79 – 12.48

1.08

0.23 – 5.24

24

D.A. Zanis et al. / Journal of Substance Abuse Treatment 21 (2001) 19–26

size for this analysis was 97; four cases contained missing data at the 6-month follow-up point and were excluded from the analysis. Overall, the model was able to correctly classify 83.7% of cases that did not work; 48% of cases who worked 1 to 14 days; and 73.9% of cases who worked more than 15 days. Five variables (age greater than 40, increased motivation, work history of 12 months or longer, transportation problems, and intervention site) were predictive of both full (15 or more days worked) and part-time (between 1 and 14 days worked) employment patterns. Additionally, lower severity of medical problems predicted full-time employment. Enrollment in the VPSS condition did not predict employment when entered in with the other variables.

4. Discussion The 6-month follow-up evaluation found that significantly more VPSS patients obtained employment than patients in the control condition, initially suggesting that VPSS counseling may have increased patient motivation to seek and obtain work. Although a greater percentage of patients who received the VPSS counseling program actually worked, VPSS was not predictive of employment after controlling for other factors. The finding that other variables (e.g., older persons, those more motivated to work, and those with more extensive work histories) were more likely to be employed has been demonstrated in other studies (Platt, 1995; Zanis, McLellan, & Randell, 1994). Individuals who had reported transportation barriers at baseline were more likely to obtain either full-time or part-time employment. However, at the 6-month follow-up point, these same persons no longer reported transportation as a barrier; thus, the intervention may have contributed to helping patients resolve transportation issue. Participation at Site B was also associated with increased employment for both intervention conditions. Although the current study was unable to detect a statistically significant difference between intervention conditions within Site B due to the small sample size, we calculated a power analysis to determine that a sample of 130 persons would have been necessary to yield a difference by condition. Importantly, 75% of the clients randomized to VPSS in Site B obtained employment, compared to 62% randomized to IPS counseling resulting in an effect of over 20%. This finding is very encouraging on two counts. First, when compared against a no-treatment intervention, the overall effect could have been somewhat greater, and, secondly, greater observed compliance in the administration of the intervention protocols at Site B was noted during counseling supervision. For example, there was a significant difference in session attendance, with patients from Site B attending an average of 8.2 VPSS sessions compared to 3.4 VPSS sessions among Site A participants. This was concerning, because we had hypothesized that a minimum of eight sessions would serve as an appropriate therapeutic dose of service. Overall, only 40%

of the patients assigned to the VPSS condition across both sites completed 8 of the 10 VPSS counseling sessions. Although participants received more vocational counseling sessions from Site B vs. Site A, subsequent analyses found that the number of sessions did not predict employment. In fact, at Site B those patients who attended fewer VPSS sessions were more likely to be employed at the 6-month follow-up period. One explanation for this finding was that a few VPSS sessions appropriately administered were sufficient to help some patients take action and find employment. Finally, lower scores on the ASI medical composite score were predictive of full-time employment. This finding has several implications, because many of the patients enrolled in methadone programs have medically related issues that may limit their participation in full-time employment. It remains unknown if these medical related issues are perceived or real barriers toward obtaining employment. 4.1. Why didn’t patients attend sessions? The methadone patients participating in this study reported many barriers to employment, including lack of job skills, lower levels of education, lack of motivation to work, substance abuse, etc. (See Table 2.) Interestingly, the most common identified barrier to employment was lack of available jobs. This perceived barrier may have been addressed in the initial counseling sessions, following which patients actively searched for and obtained employment, thus dropping out of sessions. The unemployment rate at the time of the study ranged from 4% to 8% in the immediate five-county area, resulting in the availability of unskilled jobs. Another factor that patients identified as influencing attendance was the duration of sessions (one session each week for a period of 10 weeks). Although the VPSS intervention was designed to help patients develop vocational problem-solving skills, many patients who were interested in securing employment immediately began actively pursuing employment options. To this end, it appears that patients wanted a job, and wanted the counselors to help them find a job, not information on how to problem-solve vocational barriers. Similarly, although some patients were successful in obtaining work on their own, others required considerable support to obtain and maintain a job. A study by Silverman et al. (1996) suggested that offering methadone patients vouchers to attend vocational sessions could be one avenue by which to reinforce patient participation in vocational training programs. Importantly, incentives may assist in helping patients attend vocational sessions that in turn can help internally motivate patients to take action. Similarly, Kidorf et al. (1998) found that behavioral contingencies can motivate many MM patients to obtain verifiable employment within the community. To this end, it may be that contingency management coupled with skill development training may an issue to further investigate.

D.A. Zanis et al. / Journal of Substance Abuse Treatment 21 (2001) 19–26

It is also important to question the skills, interest, and capacity of methadone counselors to deliver the VPSS intervention. Although counselors demonstrated competency in delivering the intervention and participated in regularly scheduled training sessions, there is little information about the impact of service delivered by the counselor. Because of the large number of counselors in the study it was not possible to statistically test for counselor effectiveness differences, although, based on observation, some counselors were clearly more motivated and effective than others. 4.2. Type of employment Although there were no differences in employment rates as a function of intervention condition, it was quite surprising that 46% of the participants did obtain employment in the 30 days preceding the 6-month follow-up evaluation. This represents nearly a 100% increase in documented employment from the baseline rate (23% had worked one or more days in the past 30 days). However, the majority (55%) of these new jobs were classified as ‘‘under-the-table,’’ but verified, legal jobs. Interestingly, most clients reported that they were not interested in legal, full-time work because of either institutional barriers (e.g., potential loss of Medicaid, inability to obtain methadone) or personal barriers (did not want a regular work schedule, lack of reliable transportation, etc.) that they were not willing to change. The types of jobs varied but consisted primarily of unskilled ones. Under-the-table jobs consisted mainly of those such as house repair, landscaping, cleaning, and warehouse work. Full-time jobs consisted of grocery store employees, nursing assistant, and custodial and restaurant work. Given some of the personal and systematic barriers faced by these clients (see Table 2) and their need to attend a methadone clinic each day, employment in conventional job settings may not be highly desired. 4.3. Study limitations Several methodological issues should be noted in the interpretation of the study results. First, although each counselor initially demonstrated interest in the project and a minimum set of competencies to deliver the intervention, several of the counselors reported complications in the delivery of the intervention. Issues identified included poor patient attendance of sessions, and inconsistent counselor adherence and compliance to the manuals. Counselors at Site A cited a lack of time to deliver the intervention. Clearly, the lack of adherence measures to monitor counselor delivery of the VPSS and IPS interventions as defined within the manuals may have led to counselor drift in the integrity of the theoretical model. Additionally, while all counselors demonstrated satisfactory competence prior to the start of the study, we were unable to monitor their competency during the delivery of the intervention. Second,

25

the use of the same counselor to deliver both the VPSS and IPS intervention conditions may have resulted in a bias of the services delivered. It is important to note that few control participants reported receiving employment counseling sessions, and both VPSS and IPS patients received equivalent doses of sessions. 4.4. Conclusion Overall, this community-based study found that employment rates between patients randomly assigned to receive up to 10 VPSS sessions had similar 6-month employment rates compared to patients assigned to a time- and attention-controlled condition. However, almost 50% of the patients had obtained some part-time employment and nearly 25% of these patients were employed on a full-time basis at the 6-month follow-up point, representing a significant increase in the proportion of patients who obtained employment. To this end, we conclude that structured employment interventions may assist unemployed methadone patients in obtaining employment; however, the type of employment services provided must reflect a variety of patient employment needs. Methadone programs should develop opportunities for patients to receive employment counseling services and advocate for employment opportunities within the community.

Acknowledgments This project was funded through grants provided by the National Institute on Drug Abuse.

References Arella, L. R., Deren, S., Randell, J., & Brewington, V. (1990). Structural factors that affect provision of vocational/educational services in methadone maintenance treatment programs. Journal of Applied Rehabilitation Counseling, 21, 19 – 26. Brewington, V., Arella, L., Deren, S., & Randell, J. (1987). Obstacles to the utilization of vocational services: an analysis of the literature. International Journal of the Addictions, 22, 1091 – 1118. Bureau of Labor Statistics. (1994). How the government measures employment. Washington, DC: United States Department of Labor. Dole, V. P., Nyswander, M. E., & Warner, A. (1968). Successful treatment of 750 criminal addicts. Journal of the American Medical Association, 206, 2708 – 2711. Etheridge, R. M., Craddock, S. G., Dunteman, G. H., & Hubbard, R. L. (1995). Treatment services in two national studies of community-based drug abuse treatment programs. Journal of Substance Abuse Treatment, 7, 9 – 26. French, M. T., Dennis, M. L., McDougal, G. L., Karuntzos, G. T., & Hubbard, R. L. (1992). Training and employment programs in methadone treatment: client needs and desires. Journal of Substance Abuse Treatment, 9, 293 – 303. Hall, S. M., Loeb, P., LeVois, M., & Cooper, J. (1981). Increasing employment in ex-heroin addicts; II. Methadone maintenance sample. Behavior Therapy, 12, 453 – 460. Hubbard, R. L., Rachal, J. V., Craddock, S. G., & Cavanaugh, E. R. (1984).

26

D.A. Zanis et al. / Journal of Substance Abuse Treatment 21 (2001) 19–26

Treatment outcome prospective study (TOPS): client characteristics and behaviors before, during, and after treatment. In F. M. Tims, & J. P. Ludford (Eds.), Drug abuse treatment evaluation: strategies, progress, and prospects, (vol. 51, pp. 42 – 68) (NIDA research monograph, Rockville, MD). Kidorf, M., Hollander, J. R., King, V. L., & Brooner, R. K. (1998). Increasing employment of opioid dependent outpatients: an intensive behavioral intervention. Drug and Alcohol Dependence, 50, 73 – 80. Kosten, T. R., Rounsaville, B. J., & Kleber, H. D. (1983). Concurrent validity of the Addiction Severity Index. Journal of Nervous and Mental Disorders, 171, 606 – 610. McLellan, A. T., Arndt, I. O., Metzger, D. S., Woody, G. E., & O’Brien, C. P. (1993). The effects of psychosocial services in substance abuse treatment. Journal of the American Medical Association, 269, 1953 – 1959. McLellan, A. T., Ball, J. C., Rosen, L., & O’Brien, C. P. (1981). Pretreatment source of income and response to Methadone maintenance: a follow-up study. American Journal of Psychiatry, 138, 785 – 789. McLellan, A. T., Luborsky, L., Cacciola, J., Griffith, J., Evans, F., Barr, H. L., & O’Brien, C. P. (1985). New data from the Addiction Severity Index: reliability and validity in three centers. Journal of Nervous and Mental Disorders, 173, 412 – 422. McLellan, A. T., Luborsky, L., O’Brien, C. P., & Woody, G. E. (1980). An improved evaluation instrument for substance abuse patients: the Addiction Severity Index. Journal of Nervous and Mental Disease, 168, 26 – 33. Metzger, D. S., Platt, J. J., Zanis, D., & Fureman, I. (1992). Voca-

tional problem-solving skills: a structured intervention for unemployed substance abuse treatment clients. Philadelphia: University of Pennsylvania. Platt, J. J. (1980). Cognitive interpersonal problem solving skills and the maintenance of treatment success in heroin addicts. Psychology of Addictive Behaviors, 1, 5 – 13. Platt, J. J. (1995). Vocational rehabilitation of drug abusers. Psychological Bulletin, 117, 416 – 433. Platt, J. J., & Metzger, D. S. (1987). Final report, role of work in the rehabilitation of methadone clients. Rockville, MD: National Institute on Drug Abuse. Siegal, H. A., Fisher, J. A., Rapp, R. C., Kelleher, C. W., Wagner, J. H., O’Brien, W. F., & Cole, P. A. (1996). Enhancing substance abuse treatment with case management. Its impact on employment. Journal of Substance Abuse Treatment, 13, 93 – 98. Silverman, K., Chutuape, M. A. D., Bigelow, G. E., & Stitzer, M. L. (1996). Voucher-based reinforcement of attendance by unemployed methadone patients in a job skills training program. Drug and Alcohol Dependence, 41, 197 – 207. Zanis, D. A., Metzger, D. S., & McLellan, A. T. (1994). Factors associated with employment among methadone patients. Journal of Substance Abuse Treatment, 11, 443 – 447. Zanis, D. A., McLellan, A. T., & Randell, M. (1994). Can you trust selfreports of substance users during treatment? Drug and Alcohol Dependence, 35, 127 – 132.