Social Networks among Residents in Recovery Homes

Abstract

Although evidence exists that substance abuse abstinence is enhanced when individuals in recovery are embedded in social networks that are cohesive, few studies examined the network structures underlying recovery home support systems. In two studies, we investigated the mechanisms through which social environments affect health outcomes among two samples of adult residents of recovery homes. Findings from Study 1 (n = 150) indicated that network size and the presence of relationships with other Oxford House (OH) residents both predicted future abstinence. Study 2 (n = 490) included individuals who lived in an OH residence for up to 6 months, and their personal relationship with other house residents predicted future abstinence. Implications of these findings are discussed.

I. Introduction

Substance abuse creates significant and pervasive health and economic burdens in the United States. Approximately 22 million (9%) of US citizens meet diagnostic criteria for substance dependence [1], and the annual costs associated with illicit drug use are estimated to be $181 billion dollars [2]. Drug abuse and addiction research currently suggests that a complex set of interacting environmental and individual factors are predictive of addictive behaviors [3]. Some factors included parental and peer modeling, social pressure, socioeconomic status, social networks, and geographic location [3]. Vaillant [4] noted that environmental factors may be key contributors to whether or not individuals maintain abstinence, and these factors included the support one receives for abstinence among their support networks. Moos [5], [6] pointed to similar environmental factors that predicted abstinence maintenance. Moos [7] maintained that effective interventions for recovering individuals might be those that engage clients and promote naturally-occurring healing processes, such as self-help based treatments.

Social support for abstinence may be critical to facilitating abstinence among persons with substance use disorders. Such social support is often acquired and utilized through participation in mutual-help groups [8], where individuals develop peer networks consisting of abstainers in recovery. Investment in abstinence-specific social support was reported to be one of the best post-treatment prognostic indicators of recovery [9], [10]. Regrettably, few studies examined the composition of or change within these social network structures, which underlie many community-based recovery support systems.

Much of the prior research on social networks and substance abuse focused on adoption (e.g. social network influences of using) and maintenance of alcohol or drug usage [11]– [12]. Neaigus et al. [13], for instance, found that drug injectors with more frequent social contacts with non-injectors engaged in lower levels of injecting risk behavior. Buchanan and Latkin [14] examined the social networks for heroin and cocaine users. In that study, which investigated social networks before and after quitting, found that those who quit had a significant change in the composition of their social network.

One type of community, mutual-support setting is Oxford House, a network of recovery homes providing affordable and safe housing for individuals in recovery for substance abuse (see [15], [16] for review and contextual information). This self-help organization has grown enormously over that last two decades from 18 Oxford Houses to over 1,400 [17]. The residences are rented, single family homes with a capacity of 6 to 12 individuals and they are gender segregated. Houses are usually located in middle class neighborhoods with access to public transportation and employment opportunities [18]. The average rent paid by an Oxford House resident is generally $100 per week, and the Oxford House represents sustainable, affordable housing for those working full-time with even minimum wage jobs. Over 10,000 people live in these recovery homes, making them the largest self-help residential recovery program in the US [19]. These houses are self-supporting and democratically run; the houses are chartered and expected to follow guidelines and traditions as suggested by Oxford House, Inc., a 501(c)(3) non-profit corporation [20]. The guidelines include models of governance, reporting, and member practices (e.g. elections, officer roles, financial reporting, financial operations, interviewing, basic member behavior guidelines, problem resolution, etc.[21]. Unlike traditional recovery homes, Oxford Houses have no professional staff or limitations on length of stay. Residents may remain at an Oxford House as long as they are abstinent, pay rent, do their fair share of chores, and are not disruptive [22]. In addition, Oxford House guidelines usually strongly encourage some level of participation in a 12-step group. The uniform, documented approach to operating an Oxford House allowed us an unparalleled opportunity to investigate a substantial sample of homogeneous communal housing units whose primary mission involved supporting recovery from substance use.

One study of Oxford House members examined abstinence-specific social support and abstention from substance use in a national sample of Oxford House residents. Results from this study found that only 18.5% of the participants reported any substance use over a one year period of time [23]. Additionally, over the course of the study, the percentage of their social networks who were abstainers or in recovery increased. Finally, latent growth curve analyses indicated that less support for substance use by significant others and time in Oxford House predicted change in cumulative abstinence over the course of the study. It also was important to explore the 6-month length of stay in Oxford House criterion, given the fact that DiClemente, Fairhurst, and Piotrowski’s [24] claimed that efficacy expectations, related to addictive behavior change, stabilized after six months of abstinence [25]. Staying in an Oxford House at least 6 months increased self-efficacy and maintaining abstinence. This outcome suggested that maintaining residency for at least 6 months of time might be a critical factor in promoting positive outcomes.

In another study, 150 Oxford House individuals who completed treatment at alcohol and drug abuse facilities in the Chicago metropolitan area were randomly assigned to live in an Oxford House or community-based usual aftercare services [26]. Positive outcomes were evident in terms of substance use (31.3% of participants assigned to the Oxford House condition reported substance use at 24 months compared to 64.8% of Usual Care participants), employment (76.1% of Oxford House participants versus 48.6% of Usual Care participants reported being employed at the 24 month assessment), and days engaged in illegal activities during the 30 days prior to the final assessment (mean = 0.9 for Oxford House and 1.8 for Usual Care participants). This study reported that durations of stay of 6 months or more led to better outcomes for individuals in recovery from substance abuse [26].

From the studies above taken together, it appears that the first 6 months of living in an Oxford House appears to be critical. However, it is still unclear how network size changes over time and may predict substance use. It is also unclear whether network composition, or the ties and relationships with Oxford House residents, may predict usage behavior. Finally, while the presence of heavy users in an individual’s network was associated with the likelihood of relapse [27], it is unclear whether an individual changing this presence of “heavy users” might predict a change in future usage. The present research predicted, across two samples of participants, that individuals with larger networks and networks that included relationships with OH residents would do better than those adults without such support. In addition, it was hypothesized that individuals who reduced the number of heavy users in their network would decrease their odds of relapse.

II. STUDY 1

A. Sample

Participants were recruited from residential substance abuse treatment facilities located in northern Illinois. Clients were asked if they were interested in taking part in a research project assessing post-treatment recovery patterns by measuring the function and outcomes related to substance use across two years following discharge. Participants were recruited over a one and a half-year period to allow a gradual transition of individuals into both conditions. Data were collected from 2002–2005, including both recruitment and two year follow-up data (see [26] for details).

In order to participate in the study, inpatient clients agreed to be randomly assigned to an Oxford House or usual after-care condition. Of those persons approached to be in the study, only four individuals indicated that they were not interested in being involved in the project. A total of 150 adults approached at treatment centers agreed to participate, and these individuals were randomly assigned to either one of the two conditions. Thus, there were 75 adults (46 women, 29 men) in the Oxford House and 75 adults (47 women, 28 men) in the usual after-care conditions.

Over the two-year follow-up, participants in the Oxford House condition spent an average of 256.2 days (range 8–730) in this setting. Of the 75 Oxford House participants, 5% stayed in Oxford House for the entire 24 months of the study, 35% moved into their own home or apartment after leaving the Oxford House, 20% went to relatives’ homes, 15% moved into a partner’s or spouse’s home, 9% went to a friend’s home, 5% went to a treatment program, 4% went to jail, 4% went to another staffed recovery home, and 3% went to a homeless shelter. Over the course of the study, two individuals assigned to the usual care condition had applied for and gained admission to an Oxford House (both decided to apply for entry into an Oxford House after spending time at other sites following discharge from the treatment facility). Using intent-to-treat rules, both individuals continued to be assigned to the usual care condition until the end of the study.

B. Procedure

All participants underwent a baseline questionnaire assessment two to three days before discharge from inpatient substance abuse treatment programs. Clients assigned randomly to the Oxford House condition, however, were scheduled to visit one of 20 Illinois Oxford Houses with one of our research staff members. During that initial visit, the participant filled out a one-page application form for entry into the Oxford House and was interviewed by the House residents. Residents then voted within 24 hours of the interview on whether or not to accept the applicant into the House. If the applicant was voted into the Oxford House, that participant moved into the house at their planned release date from the treatment facility. All Oxford House participants except one person were successfully voted into a house at this initial attempt. The participant not voted into the first Oxford House visited was brought to a second Oxford House and was then accepted as a resident.

Participants randomly assigned to the control or usual care condition were referred following discharge from the inpatient treatment facility by their case managers to different forms of outpatient treatment, self-help groups, or other resources in the community. Participants assigned randomly to the usual care condition, after leaving the treatment setting, went to the following sites: a relative’s home (32%), a staffed recovery home (18%), a partner’s or spouse’s home (16%), their own home or apartment (16%), a homeless shelter (10%), a substance abuse treatment program (4%), or a friend’s home (3%).

After participants entered the study, they were interviewed every six months over a two-year period, yielding a total of five assessments (i.e., baseline and 6, 12, 18, and 24 month follow-ups). In order to reach the participants during these four subsequent assessment waves, the interviewers used data from a detailed tracking information sheet developed for this study. This sheet contained, for instance, telephone numbers and addresses of family, friends, neighbors, employers, post offices, credit unions, and tax offices. Name and contact information for the person who always knew how to reach the participant also had been obtained at the beginning of the study and in each subsequent wave. Participants were paid $40 for filling out the pretest questionnaire at baseline, and equivalent incentives were used for the subsequent four interview waves. Study completion rate across the two years was comparable for Oxford House (89%) and usual after-care (86%) participants.

The validity of abstinence self-report data was enhanced by having a person in each participant’s support network listed on the 24 month follow-up assessment confirmed the participant’s level of abstinence at the 2 year assessment [28]. This collateral information, based on a procedure developed by Clifford and Longabaugh [29], was obtained from the person who was rated by the respondent as a most important person in his or her life. If the collateral report indicated alcohol or drug use, and the individual reported no use, we counted this person as using. This is a standard method in substance abuse research, and is generally seen as a more conservative approach to assessing abstinence.

C. Measures

All participants completed the Form 90 Timeline Follow-back [30] administered at baseline and subsequent interviews to measure alcohol and drug use in the previous 180 days. Reliability research, including test/retest interviews, on the Form 90 found the instrument to be sufficiently reliable for alcohol and drug treatment research and individual assessment using self-reported usage [31].

Participants also completed at each wave the Important People Inventory (IP) [29], [32]. The IP solicits information regarding an individual’s social support network including the enumeration of people, their relationship to the participant, duration of relationship, frequency of contact, general support and their own drinking and drug usage behavior. The IP has been utilized extensively in addiction research since its development for Project Match [33]. This instrument provided information on the participant’s network characteristics.

D. Results and Discussion

Table I shows the trends of descriptive measures for self-efficacy, unemployment, and the usage behavior of an individual’s most important person list. For these measures of medians and proportions, the dramatic changes occur in the first 6 months of the study; the slopes of change post-6 months are generally flat. This suggests, at least for these measures, an individual’s progress in recovery occurs in the initial 6 months post treatment. These findings suggest that significant changes occur over those first six months with respect to likelihood of employment, change in median abstinence specific self-efficacy, and percentage of sober members in the most important person network. For example, the median abstinence specific self-efficacy for the Oxford House sample increased by nearly 10 points in the initial 6 month measurement period, the unemployment rate dropped by over 52 percentage points, and the most important four person social network of the last 90 days became a 100% sober network.

TABLE I.

Indicators of Timing of Individual Improvement—Oxford House Condition

Indicator	Baseline	6 Months	12 Months	18 Months	24 Months
Self-Efficacy-Median	81.5	91.0	94.5	97.5	98.5
%Unemployment Rate	80%	28%	18%	23%	21%
% Sober Network-Median	75%	100%	100%	100%	100%

In Study 1, a mediation logistical path model was used to test whether network size, the presence of relationships with Oxford House residents, and changes in the numbers of heavy users within an individual’s important person network were predictive of short term (usage at 6 months) and longer term (usage at 24 months) abstinence. Since individuals in both conditions could have relationships with OH residents, condition was used as a control variable thereby eliminating the possibility of relationships with OH residents serving as a proxy for treatment condition. At 6 months, 37.3% of the participants had not remained abstinent in the previous 6 months and at 24 months, 45.2% had not been abstinent in the preceding 6 months.

While treatment condition was used as a control method, it was not significantly predictive of usage behavior in the first 6 months of the study (see Table II). Network size at 6 months and the number of Oxford House relationships were both significantly related to usage at 6 months. Changing the number of heavy users in the network was not predictive of usage behavior; that is on average, reducing the number of heavy users in the network did not increase the odds of remaining abstinent.

TABLE II.

Mediation Path Model Logistic Regression Coefficients and Significance Tests for the Relationship Between Network Characteristics and Usage at 6 Months

Variable	Estimate	SE	t	p(2-tailed)	Odds Ratio
Condition	−0.589	0.443	−1.330	0.183	0.555
Network Size	−0.402	0.164	−2.449	0.014	0.669
OH Relationships	−2.129	0.667	−3.194	0.001	0.119
Change in Heavy Users	0.059	0.317	0.186	0.852	1.061

For the two significant relationships, network size and number of OH relationships, the changes in odds for not being abstinent were material. The overall average odds ratio was approximately .60 and an increase of one person in the social network size was associated with a reduction in predicted odds to about .40 or a 28.6% likelihood of not being abstinent compared with the overall average of 37.3%. If an individual reported a relationship with an Oxford House resident, the effects were even stronger. A one person addition of an Oxford House resident to the individual’s important person inventory predicted a reduction in the odds ratio of approximately 88%. This relationship would suggest individuals with OH personal relationships were unlikely to not remain abstinent at 6 months.

Simultaneously, the path model measured the relationship between usage at 6 months (for the prior 6 months) and usage at 24 months (for the prior 6 months). In addition, the mediation paths of network size and OH relationships were tested (see Table III). Usage behavior at 6 months was strongly predictive of behavior at 24 months (t = 5.29, p = .000). The odds ratio of 9.75 strongly suggests that those using at 6 months were also using at 24 months and those that were abstaining at 6 months had significantly lower likelihoods (≈ .20) of not abstaining than the overall average of 45.2%.

TABLE III.

Mediation Path Model Regression Coefficients and Significance Tests for the Relationship Between Usage at 24 Months and Usage at 6 Months and the Mediated Path Coefficients

Variable	Estimate	SE	t	p(2-tailed)	Odds Ratio
Usage @ 6 Months	2.278	0.430	5.293	0.000	9.755
Mediation Paths:
Network Size	−0.916	0.412	−2.223	0.026
OH Relationships	−4.850	1.773	−2.735	0.006

Overall, these results suggest a powerful association between abstinence and personal relationships with Oxford House residents and that overall, social network sizes are predictive of abstinent behavior and that these phenomenon carried through to 24 month behavior.

III. STUDY 2

A. Background

Several studies focused explicitly on general and specific social support within Oxford House. Regarding general support, residents rated “fellowship with similar peers” the most important aspect of living in an Oxford House [34]. In terms of perceived specific support, cross-sectional research suggested that the Oxford House experience may provide residents with abstinent-specific social support networks consisting of other residents in recovery. For instance, among African Americans living in Oxford House, other house residents contributed to abstinent support networks [35]. Davis and Jason [36] found that longer lengths of stay in Oxford House were related to less specific social support for alcohol and drug use, which was related to abstinence self-efficacy. Likewise, Majer, Jason, Ferrari, Venable, and Olson [37] found that time spent in Oxford House combined with twelve-step participation was related to increased support for abstinence.

Using a national data base of residents of Oxford House members, Groh, Jason, Davis, Olson, and Ferrari [27] found that general support from friends was predictive of drinking rates. Groh et al. [33] performed factor analyses to develop a more structurally consistent model of the IP as compared to the original model. Results indicated a three-factor model, which explained about two thirds of the total variance. These three factors included: Support for Drinking from Network Members, Drinking Behaviors of Network Members, and General Social Support. Drinking Behaviors of Network Members consistently had a stronger relationship with alcohol use variables as compared to Support for Drinking from Network Members. This finding implied that whether one’s friends and family were drinkers may have a greater impact on one’s alcohol use than whether friends and family actually provide support for drinking. Nonetheless, these findings lead to the prediction that out of the three factors (which measure general support, network drinking behaviors, and support for drinking), Drinking Behaviors of Network Members may be the best predictor of alcohol use over a four-month period. In addition, it was expected that Drinking Behaviors of Network Members would be a better predictor of alcohol use over a four-month period than the Network Support for Drinking summary score from the original model of the IP. Subsequently, Groh, Jason, Ferrari, and Halpert [38] found of the three factors measuring general support, network drinking behaviors, and support for drinking, drinking behaviors of network members was the only significant predictor of alcohol use over a four month period. Study 2, then, explored issues of network size composition and Oxford House relationships with this data set.

B. Sample

Participants at the start of Study 2 (n = 490) consisted of 187 women and 303 men, a subset of a larger sample of 897 Oxford House residents who participated in a national study of recovery for Oxford House residents (see [26]). This sample was limited to those individuals who had spent 6 months or less in OH residency to minimize the effects of natural relocation effects post 6 months. The overall sample was ethnically diverse, with 58.4% European American, 34.0% African American, 3.5% Hispanic/Latino, and 4.0% others. The average age of the sample was 36.2 (SD = 8.9) and the average education level was 12.6 years (SD = 2.1). Regarding marital status, 49.0% were single or never married, 46.2% were divorced, widowed, or separated, and only 4.8% were currently married. The average participant had undergone alcohol treatment 2.8 times (SD = 4.2) and drug treatment 2.9 times (SD =.5) in their lives.

C. Procedure

Study 2 performed secondary analysis on data from a large national investigation funded by the National Institute on Drug Abuse of current residents of Oxford House who were at various stages in their alcohol and drug recovery. Data was collected from 2001 to 2004 at four time points starting at Time 0 and continuing every 4th month for 12 months. The nature, purpose, and goals of the study were explained to the potential participants. Participation was entirely voluntary, and payments of $15 were made to participants following each survey. These data were gathered by research staff who primarily administered questionnaires in person to the participants. Additionally, some data were collected by telephone. No significant socio-demographic differences were found based on methods of data collection.

D. Measures

Baseline Time 0 demographic information was obtained from items on the 5th Edition of the Addiction Severity Index-lite (ASI) [39]. The ASI assesses common problems related to substance abuse: medical status, drug use, alcohol use, illegal activity, family relations, and psychiatric condition. In addition, questions in the ASI-lite measure the number, extent, and duration of problem symptoms in the person’s lifetime and in the past 30 days. The ASI was used extensively in substance abuse studies over the past 15 years, and has been shown to have excellent test-retest reliability (≥.83) [39]. For the present study, demographic and background information from the ASI included age, sex, ethnicity, years of drug use, and whether participants abused alcohol, drugs, or both alcohol and drugs.

At Time 1,2, and 3, participants also completed a version of Miller and Del Boca’s [30] Form 90 Timeline Follow-back, which measures general health care utilization and residential history, in addition to past 90 day alcohol and drug use. The psychometric properties of the Form 90 were reported above in Study 1 and discussed in detail in Tonigan et al. [31]. In addition, participants completed the Important People Inventory [29], [32]. As in Study 1, this instrument collected information on an individual’s list of important people including relationship, length of relationship, frequency of contact, support, and their own usage behaviors.

E. Results and Discussion

In Study 2, logistic regression was used to model future abstinence as predicted by network size, changes in network size, and the proportion of the network that were Oxford House residents. Since this sample consisted solely of Oxford House residents and was a convenience sample with differing lengths of stay, length of stay was used as a control variable. In addition, this sample was constrained to those residents with 6 months or less of Oxford House experience to minimize the possible erosion of Oxford House relationships over time due to the natural transitioning of residents to more permanent housing.

Overall, the logistic regression model was significant (r² = .085, p = .04) with the control variable length of stay significantly predictive of future abstinence for the final 8 months of the study (see Table IV). The odds of relapse or not being abstinent decreases by 16.7% for every month increase in length of stay, Overall, the percentage of participants who did not remain abstinent in the future 8 months of the study was 29.4% with a mean length of stay of 2.6 months (SD = 1.8) at study initiation. This odds ratio would suggest that for a one month increase the likelihood of being abstinent would increase from 70.6% to 74.3%.

TABLE IV.

Logistic Regression Model of Future Abstinence as Predicted by Network Size, Change in Network Size, and Proportion of Network Who are Oxford House Residents

Variable	Estimate	SE	t	p(2-tailed)	Odds Ratio
Length of Stay	−0.182	0.076	−2.398	0.016	0.833
Proportion OH Residents	−1.628	0.668	−2.436	0.015	0.196
Network Size	0.070	0.047	1.474	0.141	1.072
Change in Network Size	−0.004	0.042	−0.093	0.926	0.996

The predictors, network size and change in network size, were not significantly related to future usage behavior. Descriptively, the mean network size was 6.00 (SD = 3.43) and the mean change in network size was 0.07 (SD = 3.62). The effect of having OH relationships was statistically and materially significant. Individuals with a larger proportion of their important people being OH residents were significantly more likely to remain abstinent in the future 8 months. The mean proportion of OH relationships was .215 (Md = .167, SD = .24). This relationship would predict that if an individual were to increase their OH network representation to .333 (or to 2 out of 6 individuals, for example) their predicted likelihood of substance usage would drop from 29.4% to 25.6%. In testing network composition for this sample of Oxford House residents, length of stay, as expected, was an important control variable, and the proportion or mix of OH residents within an individual’s important person inventory was predictive of likelihood of abstinence in the future 8 months. Network size and change in size were not significantly related to usage behavior.

IV. Conclusion

Significant changes occur over those first six months with respect to likelihood of employment, change in median abstinence specific self-efficacy, and percentage of sober members in the most important person network. For example, the median abstinence specific self-efficacy for the Oxford House sample increased by nearly 10 points in the initial 6 month measurement period, the unemployment rate dropped by over 52 percentage points, and the most important four person social network of the last 90 days became a 100% sober network.

Taken together, both studies suggest a strong relationship between an individual’s social connection with Oxford House individuals and their own likelihood of remaining abstinent. In Study 1, where individuals were generally recruited directly from 30 day treatment centers, the formation of a single Oxford House relation reduced the probability of relapse in the first 6 months by nearly a factor of 5 and the overall size of the important person network was materially significant as well. In the convenience sample that included 100% OH residents and where the average length of abstinence was 9 months (M = .77 years, SD= 1.11 years), Study 2 found that network size was no longer predictive of abstinence after this length of stay, and while OH relationships were still significant, the marginal positive effect of adding an OH relationship was substantially smaller. Given the characteristics of the samples, these results are supportive of the criticality of an individual’s development in early recovery and the dynamic nature of recovery as a continuous and evolving process.

Dynamic social network characteristics allow researchers to investigate the co-evolution of networks and behavior. These approaches allow for new insights into the initiation and maintenance of substance abuse. For example, Pearson, Steglich, and Snijders [40] illustrate the co-evolution of friendship networks and substance abuse among teenagers. They found strong network selection effects occurring with a preference for same sex reciprocated relationships in closed networks. Steglich, Snijders, and West [41] explored the co-evolution of social networks regarding the role of peer groups in friendship networks and alcohol consumption. Mercken, Snijders, Steglich, Vartiainen, and de Vries [42] tried to tease apart is the relative importance of selection versus influence in a study of Finnish adolescents, and they found that both played a role but that selection had a larger effect.

For many individuals with substance abuse problems, entry into the existing continuum of services begins in a detoxification program. In the optimal case, an individual completes the detoxification process and then moves through a time limited therapeutic program. However, these programs are becoming briefer as federal, state and local sources of funding for these services has decreased [15]. For a substantial portion of addicted persons, detoxification does not lead to sustained recovery. Instead, these individuals cycle repetitively through service delivery systems [43], [44]. Recidivism rates within one year following treatment are high for men and women, and 52–75% of all substance abusers drop out during treatment [45]. These kinds of programs are also expensive [46]. The missing element for many patients is supportive, cohesive settings following treatment for substance abuse.

Oxford Houses differ from traditional recovery homes in terms of who is responsible for enforcing the rules and policies. In traditional recovery homes, although the delegation of some tasks might be shared with residents, it is the owners or people designated with management authority who often have the most substantive responsibilities for implementing rules. In traditional recovery homes, owners or designated staff make key decisions regarding house governance such as entrance into the house as well as eviction. Therefore, traditional recovery homes might have resident social climate perceptions that are different compared to peer-run settings such as Oxford Houses [47], [48]. Self-run settings might require residents to perform duties normally completed by staff, therefore, self-governance in Oxford House peer-run settings might create social climates that are more supportive, involving, and oriented towards solving residents’ problems [49]. Other aspects of the Oxford House experience that may contribute to recovery include social control and reinforcement by fellow house members, social learning, reduced stress, and improved coping strategies [50].

This study could lead to a clearer understanding of mechanisms through which social environments affect health outcomes, which could help policymakers and health care providers reduce unnecessary health care costs by improving the effectiveness of residential recovery home system in the US and to restructure and improve other community-based recovery settings. These improvements could lead to better outcomes such as clients receiving adequate treatment doses to effect meaningful changes and prepare them for a successful transition to independent living in the community.