A Model for Sober Housing during Outpatient Treatment

Abstract

Finding a living environment that supports recovery is a major challenge for many clients attending outpatient treatment. Yet it is important because family, friends, and roommates who encourage substance use or discourage recovery can undermine the progress made in treatment. Destructive living environments are most problematic for clients who have limited incomes and reside in urban areas where housing markets are tight. Individuals who are homeless face constant threats to their sobriety and often lack the stability necessary to attend treatment consistently. Options Recovery Services is an outpatient program in Berkeley, California that uses sober living houses (SLHs) to provide an alcohol and drug free living environment to clients while they attend the outpatient program. This paper describes the structure and processes of the houses along with six month outcome data on 46 residents. Improvements included the number of months using substances, maximum number of days of substance use per month, arrests, and employment. Seventy six percent of the residents remained in the house at least 5 months and 39% reported being employed at some point during the past 30 days. Outpatient programs should consider establishing SLHs for clients who lack a living environment supportive of sobriety.

Introduction

Containing the costs of alcohol and drug treatment has become a major goal for state and local governments that fund treatment (Institute of Medicine, 1997; McLellan, 2006). One repercussion has been an increase in less expensive outpatient services and a decrease in more costly residential and inpatient treatment. Despite increasing popularity, outpatient treatment programs have the serious limitation of not being able to control the social and living environment of clients. They are particularly limited in terms of addressing the needs of homeless clients who face constant obstacles to their health and safety as well as their sobriety. In urban areas, where low income housing is limited, large proportions of clients attending outpatient treatment programs reside in destructive living environments that do not support recovery or they are homeless (Howard, La Veist, & McCaughrin, 1996).

The lack of supportive living environments for clients is a significant concern because a variety of studies show that provision of a social and living environment supportive of sobriety in related to better outcome (Beattie et al., 1993; Hitchcock, Stainback & Roque, 1995; Howard, La Veist, & McCaughrin, 1996). For example, Milby et al. (2005) specifically studied whether the provision of abstinent contingent housing during outpatient treatment was more effective than no provision of housing or housing that did not require sobriety. Although all three groups made improvement on outcome measures, the abstinent contingent housing group improved the most.

This paper takes the position that more efforts are needed to help outpatient clients establish a living environment that supports sobriety. A housing model that differs from the Milby et al. (2005) model is offered: Sober Living Houses (SLHs). After describing the general principles of SLHs as a recovery modality, the structure and processes of SLHs operated by Options Recovery Services in Berkeley, California are presented. Unlike many freestanding SLHs, Options houses are associated with an outpatient treatment program. Descriptive data on 46 individuals is provided to show areas of improvement between entry into the houses and 6-month follow-up. We hypothesized that individuals who entered the houses with high problems severity (e.g. substance use, arrests, unemployment, and psychiatric symptoms) would show significant improvement. We also expected that individuals who entered the houses with low problem severity would maintain low severity at 6-month follow up.

Sober Living Houses

As described elsewhere (e.g., Polcin, Galloway, Taylor & Benowitz-Fredericks, 2004; Polcin & Henderson, in press), SLHs are alcohol and drug free living environments for individuals attempting to establish or maintain abstinence from drugs and alcohol. They typically do not offer any formal treatment services, but encourage or require attendance at self-help groups such as Alcoholics Anonymous. However, the Options Recovery SLHs described here are an exception because these houses are part of a structured outpatient treatment program which residents are required to attend.

One of the advantages of SLHs is that residents are free to stay as long as they like. For the most part, they are financed through resident fees. Some houses are sufficiently inexpensive to accommodate residents who are on General Assistance or Social Security Disability Income. However, others are more expensive and serve primarily individuals who work full time or have access to other financial resources, such as support from their families. In California, some criminal justice funds will pay for the first few months of residence in a SLH for offenders who are eligible. A final characteristic of SLHs is an emphasis on involving the individuals who live there in making decisions about operations of the houses. Thus, most SLHs have some type of residents council.

Most SLHs operate as freestanding programs and have no affiliation with specific treatment programs, although residents may be attending various substance abuse, mental health, and other services in the community (Polcin & Korcha, 2006 November). However, there is no inherent reason why SLHs cannot be affiliated with specific programs, similar to halfway or aftercare houses offered by some residential treatment programs. The difference between SLHs and these other types of residences include: 1) residents can stay as long as they wish in SLHs even after they have completed treatment, 2) the houses are financed through resident fees, although they may be supplemented to some degree by the treatment program, and 3) the houses are not licensed by the state as treatment facilities. The housing model used in the Milby et al (2005) study was more typical of halfway house models associated with treatment in that it provided a free residence for clients at no cost and the maximum length of stay was 12 months.

This paper focuses on SLHs operated by Options Recovery Services in Berkeley, California, an outpatient addiction treatment program. Divergences from most of the typical freestanding SLHs include: 1) a three phase structure to the houses, 2) requirements of attendance at 12-step meetings and the agency’s outpatient treatment program, and 3) overall management of house operations. Descriptive data documenting resident functioning in multiple areas (e.g., substance use, employment, arrests, psychiatric symptoms, and family and medical problems) was collected on 55 residents at entry into the houses, 46 (84%) of whom were interviewed at six-month follow up. After reviewing these findings, the paper ends with a discussion about the potential benefits of SLHs as part of outpatient treatment.

Options Recovery Services

Options Recovery Services is an outpatient substance abuse treatment program in Berkeley, California that treats about 800 clients per year. Founded in 1997, Options currently treats a variety of individuals with addictive disorders. About 25% of the clients are women, 70% are court mandated and 53% have a history of homelessness. The incidence of psychiatric disorders is high, with a majority suffering from a DSM-IV Axis I psychiatric disorders in addition to their substance abuse problems.

The outpatient program offers a variety of recovery services including intensive case management, recovery groups, and aftercare. After an early treatment period lasting two weeks, the program is structured into three phases, each with specific requirements.

Phase 1

Clients are required to attend the Options outpatient program 3 hours per day, 5 days a week, and attend daily 12-step meetings at the program. In addition, clients attend 2 outside 12-step meetings as well. Clients also receive 25 sessions of acupuncture treatment. Phase 1 lasts 11 weeks.

Phase 2

Clients are required to attend the Options Outpatient program 3 days a week and attend daily 12-step meetings at the program. In addition, clients attend 2 outside 12-step meetings as well. Phase 2 lasts 13 weeks.

Phase 3

Clients are required to attend the Options program 1.5 hours per day three times per week. In addition, clients are required to attend 2 outside 12-step meetings per week. Phase 3 lasts 13 weeks.

Exceptions to the requirements of each phase are allowed on a case by case basis. Staff members work to individualize treatment, so requirements for a given client may be more or less stringent depending on the clinical situation.

For clients requiring psychiatric services, referrals are made to a local outpatient facility. A local psychology training program offers on-site mental health therapy for clients who are interested.

Options Recovery Services is fortunate that its facilities are located near a homeless shelter. This provides those outpatient clients who have no place to stay convenient though temporary housing. In response to the large number of clients in the outpatient program who needed housing, the program developed SLHs. The SLHs began with a single dwelling 2001 and today has 4 houses with 58 beds. Because of their status as SLH residences, they are not licensed by the state.

Structure and Processes of Sober Living Houses

Options Recovery Services SLH’s have been modified from the traditional SLH model in order to be consistent with the structure and operations of the outpatient treatment program and the nature of the population treated. For example, to be eligible for admission to the program’s SLHs, clients must have a minimal amount of sobriety (typically 30 days or more) and be in good standing in the outpatient program. All requirements of the outpatient program apply to the SLH residents. Because the SLHs target those clients who are homeless, prospective SLH residents are recruited from clients who live in the homeless shelter near the program. Most residents are eligible for some type of government economic assistance and they use that assistance to meet expenses at the SLHs. The agency adjusts fees based on amount of income. For those on General Assistance (GA), the fees are $250 per month and for those on Social Security Insurance (SSI) or working the fees are $350 per month. The program assists residents in their applications for GA and SSI.

In terms of the operations of the SLHs at Options Recovery Services, there are both similarities and differences with traditional freestanding SLHs. Like most other houses, one experienced, senior resident is designated a house manager. He or she is responsible for ensuring house rules are followed and consequences for rule violations are carried out. In addition, they monitor the facility and report needed repairs to the agency’s executive director. Each house has a weekly house meeting that all residents are required to attend. These meetings are primarily focused on resident responsibilities, such as rotation of household chores, resident responsibilities, and enforcement of house rules. Although residents can have input into development and enforcement of house rules and policies during these meetings, policy is ultimately developed by the agency’s clinical and executive directors with substantive input from house managers.

Some of the standard house rules include a curfew of 10:00pm from Sunday through Thursday and midnight from Friday through Saturday. All houses became nonsmoking facilities in January 2006. Relapse is usually handled by referring the individual to a higher level of service, such as residential treatment. If a resident completes treatment, leaves the SLH residence, and then relapses, he/she is typically invited to return. A brief stay at the shelter may be necessary if a SLH bed is not available.

Unlike most freestanding SLHs, which in general admit higher functioning individuals than those at Options Recovery Services, employment is discouraged until residents establish a strong program of recovery over a number of months. The emphasis is on establishing sobriety. However, in Phase III residents are encouraged to find work, attend school or volunteer when it is clinically appropriate.

Although residents in good standing can stay in the SLHs as long as they wish, nearly all leave within two years. The living quarters are small for the number of people in the houses and those who are succeeding in recovery usually want more space and privacy.

An Evaluation of Sober Living Houses

An Evaluation of Sober Living Houses” is a 5-year study funded by the National Institute on Alcohol Abuse and Alcoholism (Polcin, Galloway, Taylor & Benowitz-Fredercks, 2004). It is currently in its fifth year and is tracking 300 individuals over 18 months who live in 20 different SLHs. Two different agencies administer the houses, and they each operate facilities with significant differences. In this report, we focus on the 4 houses affiliated with Options Recovery Services in Berkeley. Data collection is ongoing, so we only examine outcomes at 6 months. A separate report (Polcin & Henderson, in press) describes 6-month outcomes for the 16 houses affiliated with Clean and Sober Transitional Living (CSTL) in Sacramento, California.

Methods

Procedures

To maximize generalization of results we employed few exclusion criteria. To reach individuals for follow up interviews we required they provide contact information (e.g., phone number, address, e-mail, names of friends who might know there whereabouts, family members’ phone numbers, health service professions from whom they received services, shelters they frequented, and criminal justice personnel). Those who refused to provide contact information were excluded. However, this was a rare occurrence and the vast majority of residents who were invited to participate in the study were enrolled.

Study participants were recruited within their first week of entering the SLH between January 2004 and June 2006 and again at 6-month follow up. Interviews required about two hours to complete and participants were paid $30 for the baseline interview and $50 for the follow up interview. All participants signed an informed consent to take part in the study and all were informed that their responses were confidential. Study procedures were approved by the Public Health Institute Institutional Review Board and a federal certificate of confidentiality was obtained, adding further protection to confidentiality.

Measures

1) Six month measures of alcohol and drug use

These measures were taken from Gerstein et al. (1994) and included: a) Number of months used any substance during the past 6 months and b) Peak Density – number of days of any substance during the month of highest use over the past 6 months.

2) Six month measures of employment

This measure was a simple count of the number of days the individual worked over the past six months and was adapted from a measure used by Gerstein et al. (1994) in their study on outcomes of drug treatment programs in California.

3) Six month measure of arrests

This measure was a simple count of the number of arrests over the past 6 months and was adapted from a measure used by Gerstein et al. (1994) in their study on outcomes of drug treatment programs in California.

4) Severity of problems

The Addiction Severity Index (ASI) (McLellan et al., 1992) was used to assess a variety of problem areas including drug, alcohol, medical, legal, employment, and family/social relationships. The ASI measures a 30 day time period and provides composite scores between 0 and 1 for each problem area. Although the instrument includes a measure of psychiatric severity as well, we opted to use a more comprehensive measure for psychiatric symptoms which is described below.

5) Psychiatric symptoms

To assess current psychiatric severity we used the Brief Symptom Inventory (Derogatis & Melisaratos, 1983). This 53-item measure assesses severity of psychiatric symptoms on nine clinical scales as well as three global indices. Items are rated on a 5-point scale and ask about symptoms over the past 7 days. We used the Global Severity Index as an overall measure of psychiatric severity.

6) Six Month Service Utilization

This measure was designed by the study team and used in previous analyses of SLHs (Polcin & Henderson, in press). The instrument assesses the number of services received for specific problems over the past 6 months. In the analyses reported here we used this measure to assess the number psychiatric and medical services received over the past 6 months.

7) Psychiatric Diagnostic Screening Questionnaire

This instrument was used to assess 13 DSM-IV Axis I disorders in 5 areas: substance use, somatoform, eating, mood, and anxiety. This 90-item self-administered questionnaire has a mean Cronbach alpha coefficient of .82 and test-retest mean alpha of .84 (Zimmerman & Mattia, 1999).

8) Demographic Characteristic

Standard demographic questions such as age, gender, ethnicity, marital status, and education.

9) DSM-IV Checklist: DSM IV Checklist for Lifetime Drug Abuse or Dependence

This measure was used to assess past year alcohol and drug dependence. Items are based on DSM IV diagnostic criteria (American Psychiatric Association, 2000) and the checklist has been used in multisite studies of addiction treatment (e.g., Forman, Svikis, Montoya & Blaine, 2004).

Analysis Plan

Findings reported below begin with a cross sectional description of resident problems at baseline and 6 months. Comparisons of measures between the two time points are then presented to show areas of significant improvement. For areas where residents entered the SLHs with low problem severity, we identify which areas remained low at six months. Because the distributions of most measures were non-normal, nonparametric analyses were used to test differences between the two time points, primarily Wilcoxon signed ranks tests for paired comparisons. This analytic strategy has an advantage over parametric tests (i.e. paired t-test) in that it makes no assumptions about the distribution of the data. Thus, it is a good option when data distributions are highly skewed, as was the case in our study. One disadvantage is that it has less statistical power than parametric methods to detect differences that actually exist.

Results

Demographics

Study participants consisted of 55 individuals entering 4 different SLHs that were operated by Options Recovery Services. Table 1 shows a description of resident demographic and clinical characteristics at baseline. Nearly all the participants were male (94%) due to the closing of the only women’s house shortly after the study began. The racial distribution was largely African American (59%) followed by Caucasian (31%), and the mean age was 42.6 (9.3) years.

Table 1.

Baseline Demographic and Clinical Characteristics (N=55)

Age, mean years (SD)	42.6 (9.3)
Male, N (%)	52 (94%)
Ethnicity, N (%)
African American	32 (59%)
Caucasian	17 (31%)
Other	6 (3%)
DSM-IV Substance Dependence Past Year N (%)
Cocaine	33 (60%)
Alcohol	30 (55%)
Cannabis	10 (18%)
Heroin	8 (15%)
Methamphetamine	7 (12%)
Usual Housing Status Past 6 Months N (%)
Homeless or Shelter	18 (33%)
Family or Friends	14 (25%)
Criminal Justice Incarceration	9 (16%)
Residential Treatment or Sober Living House	7 (13%)
Apartment	6 (11%)
Other	1 (5%)
Positive Psychiatric Screen N (%)
Social Phobia	31 (56%)
Post Traumatic Stress Disorder	25 (46%)
Psychotic Disorders	23 (42%)
Major Depression	18 (33%)
Obsessive Compulsive Disorder	18 (33%)

Descriptive Findings at Baseline

The DSM-IV Checklist for alcohol and drug disorders was used to assess one year dependence before entering the SLH. Table 1 shows that during the year before entering the program, the most common substances residents were dependent on were cocaine (60%) and alcohol (55%). Other dependencies were substantially less common.

Many of the residents had histories of homelessness. When asked to indicate their usual housing situation the past six months, a third indicated homeless or in a shelter. Twenty-five percent indicated they stayed with family or friends and 16% indicated their primary residence was criminal justice incarceration. Only 10% indicated their primary residence was renting their own apartment.

In our screening for psychiatric disorders we found the incidence of some disorders to be high, particularly social phobia (56%), post traumatic stress disorder (46%), and psychotic disorders (42%). It is important to note that meeting the screening criteria does not mean those individuals met criteria for the disorder. It merely suggests an indication of some symptoms that require further assessment. These data also do not suggest whether these conditions predated substance use or may have been a consequence of substance use. Nevertheless, these percents were high and consistent with results from our other measure of psychiatric severity, the Brief Symptom Inventory (Derogatis & Melisaratos, 1983). Scores on a global severity index (GSI) of this measure indicated clients had a level of psychiatric problems consistent with that of individuals attending outpatient treatment for psychiatric disorders (mean = 0.67, sd = 0.60) (Derogatis, 1993).

Descriptive Findings at 6 Months

At six months, 46 of the 55 participants (84%) were located for follow up interviews. Thirty two (76%) had been residing in the house at least some of the past 30 days. In terms of substance use, 67% (N=31) indicated that they had not used any drugs or alcohol over the past 6 months. Among the 15 individual who relapsed, 9 used substances 4 days or less during their month of highest use. Thus it appears they were successful at reestablishing abstinence shortly after relapsing.

Alcohol breathalyzers and urine screens were used to corroborate self report from residents. In general, biological screens confirmed self reports about substance use; over 90% of the drug screens were consistent with self reports. We found no instances of a positive breathalyzer. However, there were a few cases where a positive drug screen occurred when the participant indicated no substance use over the past 6 months. This occurred 3 times at baseline and 4 times at 6-month follow up. Because the number of discrepancies was roughly equal at the 2 time points the comparisons depicting improvement in substance use were probably affected minimally, if at all. However, it does appear that there was some, albeit limited, under reporting of drug use and that it is likely a bit higher than reported.

Despite the very high rates of reported psychiatric symptoms at entry into the houses, a modest number of residents reported that they received psychiatric treatment. The proportion receiving any type psychiatric treatment in an outpatient mental health program was 37% (N=17) and the proportion receiving any type of medication for psychiatric problems was 41% (N=19). Those who did indicate receiving some services for psychiatric problems were often only minimally involved. Nearly half (N=8) of the17 individuals attending outpatient treatment indicated they attended 6 days or less over the 6 month assessment period. In contrast, about 63% indicated they received medical services for concerns about health.

Overall, involvement in work over the 6 month assessment period was limited, with 48% (N=22) indicating no work at all and 61% indicating they worked 10 days or less. However, when we looked at employment on the ASI we found that 5 months after entering the SLH 39% had begun some involvement in work.

Comparison between Baseline and 6-Months

To assess whether residents improved between entry into the houses and 6-month follow up, we conducted paired comparisons of study variables. Because the data for most instruments were not normally distributed we used a nonparametric test, the Wilcoxon signed ranks tests for paired comparisons.

Most measures that assessed a six month time period showed significant improvement (See Table 2). These included the number of months residents used substances and number of arrests over the last six months. When individuals did relapse, their patterns of use were less severe. Peak Density for those who relapsed (N=14) declined from an average of 19 days during the month of highest use before entering the SLH to 11 days during the month of highest use at 6-month follow up. During the 6 months prior to entering the SLHs 40% had been arrested at least once and that declined to 11% percent during the subsequent six months.

Table 2.

Baseline and 6-Month Comparison of Variables Assessing Outcome over a 6-Month Period of Time (N=46)

Variable	Baseline	Six Months	Z Score
Number of Months Used Substances	3.0	0.80	−4.5***
Peak Density (maximum days/month)	19.4	11.1	−2.3*
Days Worked	23.3	28.0	NS
Arrests	0.7	0.1	−2.9**

^***p<.001,

^**p<.01,

^*p<.05

Note: Findings reported are means for the six month period before entering the SLH versus the six month period after. Paired comparisons were conducted using Wilcoxon Signed Ranks Tests. Peak Density only assessed individuals who relapsed (N=14).

One measure that did not show improvement at the 6 month time point was the total number of days worked during the past 6 months. However, when we examined employment on the ASI Employment scale we found significant improvement between baseline and 6-monh follow up. At the 6-month time point, 39% percent of the participants indicated that they had engaged in some type of employment over the past 30 days. During the month before entering the SLH only 11% indicated they had engaged in work during the past 30 days. While the average number of days worked over the past 30 days at the 6-month time point was low (6.7 days), it was nonetheless higher than the month before entering the SLH (1.4).

Some measures assessed time periods of one month or less and revealed low problem severity during the month before residents entered the SLHs. These included Addiction Severity Index (ASI) scores for the alcohol and drug scales. The values for these scales are indicated in Table 3 and are low relative to our studies of individuals entering treatment in our geographical area (e.g., Polcin & Beattie, 2007; Polcin & Weisner, 1999). Thus, there was little room for improvement on alcohol and drug severity. Nevertheless, these scores were lower at 6 month follow up, even though they did not reach the level of statistical significance.

Table 3.

Baseline and 6-Month Comparison of Addiction Severity Index Scales and Psychiatric Symptoms on the Brief Symptom Inventory (N=46)

Variable	Baseline	Six Months	Z Score
Alcohol	0.07	0.06	NS
Drug	0.05	0.03	NS
Family/Social	0.25	0.28	NS
Legal	0.10	0.05	−1.9*
Employment	0.84	0.69	−3.9***
Medical	0.33	0.34	NS
Global Severity Index (Brief Symptom Inventory)	0.71	0.71	NS

^***p<.001,

^*p<.1

Note: Findings reported for the ASI are means that assess the month before entering the SLH versus means that assess the month prior to the 6-month interview. Findings for the BSI are means that assess psychiatric symptoms the seven days prior to entering the SLHs with the seven days prior to the 6-month interview. Paired comparisons were conducted using Wilcoxon Signed Ranks Tests.

There were two areas where residents entered with relatively high problem severity that did not improve at six months (see Table 3). These were medical severity on the ASI and psychiatric symptoms on Global Severity Index of the BSI. The lack of improvement on psychiatric symptoms was particularly concerning because higher severity at 6 months correlated with more months of substance use (Spearman’s rho = .31, p<.05).

The results of other ASI scale comparisons varied. Residents entered SLHs with moderately high family severity (mean = 0.25) and that did not improve at 6-month follow up. However, legal severity showed a trend toward lower severity at 6 months (Z=-1.9, p<.10).

Discussion

Although a few papers have described the history of SLHs (e.g., Polcin 2001, Wittman, 1993) and preliminary outcomes in freestanding SLHs not affiliated with treatment (e.g., Polcin & Henderson, in press), this report represents to our knowledge the first evaluation of SLHs associated with outpatient treatment. Overall, residents made significant improvements in multiple areas of functioning and they were able to maintain functioning in areas where they evidenced few problems (e.g. 30 day alcohol and drug severity). Relative to subsidized “halfway” houses, SLHs have the advantages of offering an open ended length of stay and being financed largely through resident fees. Thus, outpatient treatment programs should consider establishing SLHs for clients who reside in destructive living environments that are likely to undermine the gains made in treatment. Various organizations are available to consult with programs about the logistical issues involved in establishing SLHs, such as the California Association for Addiction and Recovery Resources (CAARR) in Sacramento, California.

Understanding our results and their implications requires consideration of a number of issues. One reason for the improvements noted may be due to the fact that 76% of the respondents were still residing in the SLH at the six month interview. This is a very strong finding given the National Institute on Drug Abuse (1999) recommendation that a minimum of 90 days is necessary to maximize the effect of treatment. In addition, a variety of studies have shown that longer stays in treatment are associated with better outcome (NIDA, 1999).

A number of areas did not show significant improvements that were not surprising. These were areas where residents entered the houses with low problem severity that remained low at six month follow up. Because entry into the SLHs generally requires some sustained abstinence (typically 30 days), ASI scales for alcohol and drug were low at entry and thus had little room for improvement. Although both scales showed lower severity at 6 months, particularly the drug scale, they did not reach statistical significance. These findings illustrate how SLHs might play an important role in helping residents maintain the previous gains made in treatment and avoid regression of target problem areas.

Although large proportions of residents enter the SLHs with some kind of legal involvement, their scores were not particularly elevated in relation to individuals entering treatment. They may have felt that their legal problems were being sufficiently addressed by their involvement in Options and therefore they were not overly concerned about them. Their optimism may have been warranted because there continued to be a trend toward lower severity at 6 months.

Although the program discourages employment until later in treatment (phase III) and relatively few individuals were working on a regular basis over the 6 month assessment period, we nonetheless found significant improvement on the ASI employment variable. There appears to be a shift toward increased involvement in work after the residents have been in the house for several months. This is consistent with the program’s view of work, which emphasizes a focus on abstinence first, and a focus on work only after several months of sustained sobriety.

Two areas not showing improvement were family and medical severity. One issue with family severity was the fact that many residents were estranged from their family of origin and few (15%) were married or in a committed relationship. Thus, family conflicts and recent concerns about them were somewhat limited. The reasons for the lack of improvement on medical severity may be different. Given the emphasis on providing housing to individuals who have histories of homelessness, it is not surprising that study participants entered with relatively high medical severity. Many of these problems may be chronic in nature and that may be one reason medical status did not improve. Relatively large proportions (63%) did appear to be involved in some type of medical treatment so lack of medical services may not be a key factor.

A final area that did not show improvement was psychiatric severity. This finding was especially concerning because psychiatric severity was high and it correlated with fewer months of sobriety. To some extent, clients may be suffering from persistent psychiatric symptoms that do not remit easily. However, there may also be underutilization of psychiatric services as well. Many residents met screening criteria for DSM IV psychiatric disorders. However, relatively modest proportions received psychiatric evaluations to diagnose and treat these disorders. Thus, strategies for helping clients engage in psychiatric services may be a useful focus of program planning. However, it is notable that many residents achieved sustained abstinence and were working a 12-step recovery program despite the existence of ongoing psychiatric and medical problems.

Additional studies are needed in a variety areas. First, we need studies that examine longer follow up time periods beyond the six months described here. Findings at subsequent time points might show evidence of regression not tapped here. Second, studies are need with larger samples in different types of types of outpatient programs. The effectiveness of SLHs with different outpatient models might vary. This would also allow for an examination of individual resident and program factors that predict good outcome. Finally, it might be fruitful to examine if modifications in the structure and operations of SLHs might need to adapt to different outpatient programs and contexts.

Limitations

Several limitations should be noted. First, the study was conducted at one site and evaluations of other SLHs could yield different results. Second, although the study had a sufficient statistical power for the analyses proposed, the diversity of sample characteristics was limited. Different client characteristics could result in different findings.

Another limitation of the study was the finding that there was some degree of underreporting of drug use as evidenced by discrepancies between self reports and urine screens. While drug use may be slightly higher than reported, over 90% of the urine screens were consistent with self reports. Because discrepancies occurred at a baseline and 6-month follow up in roughly the same proportions they likely had little effect on paired comparisons between the time points. In addition, the findings on improvements in substance use and other areas, such as legal severity, housing stability and arrests, appeared to be robust.

Conclusion

Outpatient substance abuse treatment programs face significant challenges treating clients who are homeless or who reside in destructive living environments. The stresses these clients face and the social encouragement that many receive to use substances often eclipse the gains made in treatment. In publicly funded urban programs, the problem of finding adequate housing that supports abstinence is especially problematic.

Few outpatient programs have made accommodations to address housing problems among their clients. This paper has described a model for sober housing that other outpatient treatment programs might want to adopt: the Options Recovery Services model of using SLHs for clients while they attend outpatient treatment. Modifications of traditional SLHs made by Options include implementing a 3-phase structure to the residence, mandating attendance at the treatment program and 12-step meetings, and requiring additional restrictions such as curfews.

Overall, the promising findings reported here suggest SLHs might be able to play much stronger roles in meeting the housing needs of clients attending outpatient treatment. SLHs are a useful way of helping clients establish as well as maintain recovery. To strengthen their impact, program planning efforts should explore additional ways to address psychiatric symptoms, which correlated with worse outcome.