Recovery residences: Which housing characteristics predict positive resident outcomes?

Recovery residences, sometimes known as sober living houses, have been shown to enhance a variety of recovery outcomes and are highly cost-effective, but little is known about which residence characteristics and practices are most important. The current study addressed this research gap by examining relationships between recovery residence characteristics (organizational, operational, programming) and residents’ recovery outcomes (substance use, criminal justice involvement, employment).


Secure housing is an important component of recovery, yet about 1/3rd of individuals entering addiction treatment report unstable living situations. Recovery residences offer a safe and supportive living environment and continuing care that can promote successful recovery outcomes. Although recovery residences do not typically offer structured activities or formal counseling, some operate according to established standards and guidelines that encourage resident participation in mutual help groups, enforce house rules, and promote resident accountability and communal learning. By providing affordable housing, peer support, and a resource rich neighborhood, recovery residences offer residents the opportunity to build financial, social, and community-based recovery capital. These continuing care facilities are shown to facilitate positive recovery outcomes (e.g., higher rates of employment, reduced substance use, lower criminal activity) that are maintained as far out as 18 months. However, few studies have looked at the individual contributions of specific recovery residence characteristics to these recovery outcomes. Given that recovery residence models vary greatly, a more detailed understanding of facility characteristics and their relationship to recovery outcomes can ultimately help to inform best practices and streamline recommendations and referrals. Guided by studies examining the organizational and management factors of treatment facilities and focusing on a high-risk population involved in the criminal justice system, this study is one of the first to assess the relationships between recovery residence characteristics and residents’ recovery outcomes.


This study examined the relationship between recovery residence characteristics (organizational, operational, and programming characteristics) and recovery outcomes (substance use, criminal justice, and employment outcomes) in 330 recovery residents. Participants were assessed upon entry to 1 of 49 recovery residences, specifically sober living houses located in California, and completed follow-up interviews 6 and 12 months later. All sober living houses were part of the Sober Living Network, a non-profit organization that implements housing standards for recovery residences across the United States. A quarter of the houses were affiliated with a treatment program and 69% of the houses were single-gender facilities.

Data included (1) Sober Living Network reports (e.g., administrative data on member houses, including resident gender, resident capacity, house fees, geographic region), (2) interviews with house managers and owners (e.g., % of clients on parole/probation, intake procedures, whether the house had an onsite manager, parole/probation referral program, or required 12-step meeting attendance), as well as (3) in-depth resident assessments (e.g., demographics, recent substance use, criminal activity, employment). These data were collected as part of a larger randomized controlled trial looking at the effectiveness of brief motivational interviewing and case management on service use and recovery progress. Substance use outcomes were measured via self-reported abstinence in the past 6 months (i.e., abstinence from alcohol, abstinence from other drugs, and overall abstinence from all substances). Criminal justice involvement was defined as any self-reported arrest over the past 6 months and employment was derived from self-reported number of days worked over the same period of time.

All participants were currently involved in the criminal justice system (e.g., probation, drug court) and were either HIV positive or had a lifetime history of at least one HIV risk behavior (i.e., men who had sex with men, commercial sex work, injection drug use, or unprotected sex with two or more partners during the past 6 months). Participant follow-up rates were 77% for the 6-month interview and 81% for the 12-month interview. Various racial and ethnic backgrounds were represented (47% white, 24% black, 10% other/mixed race,19% Latino/Hispanic). The majority of participants were male (74%). On average, participants were 39 years old and about one-third had at least some college education. The authors statistically controlled for participant demographics (gender, race/ethnicity, age, education) and the duration of stay at the recovery residence (on average, 149 days) to try to isolate the direct effect of residence characteristics on recovery outcomes. The authors looked at 6- and 12-month recovery outcomes simultaneously to obtain a broad snapshot of how residents do across the first year of recovery housing. Analyses statistically controlled for the time at which data were collected.


General characteristics of recovery residences and resident outcomes.

Among the participating recovery residence sites, 32% had residents on parole or probation, 74% had a live-in manager, 30% provided meals to residents, and 77% required residents to attend 12-step meetings and complete a drug test at intake. On average, facilities required 41 days of abstinence prior to intake and most operated on a 12-step based program. From baseline to each subsequent follow-up period, rates of abstinence and employment generally increased, while rates of arrest generally decreased.  

Houses affiliated with a larger organization and serving fewer residents were associated with better recovery outcomes.

Housing that was part of an organization or larger group of houses was associated with greater likelihood of drug and alcohol abstinence. Housing with an affiliated treatment program was associated with greater likelihood of total abstinence and employment. Facilities that had referral agreements with parole and probation were associated with lower odds of arrest and greater likelihood of employment. Likelihood of employment decreased among facilities housing 21 or more residents but increased among houses with 10 or fewer residents.

Alcohol abstinence and employment varied according to the Sober Living Network chapter (i.e., geographic region) that the house belonged to.

Relative to residents in LA Metro Chapter housing, residents in the San Fernando Valley/San Gabriel Valley Chapter and West LA Chapter housing showed a greater likelihood of alcohol abstinence. West LA Chapter housing was also associated with a greater likelihood of employment.

Abstinence varied according to the price of housing and types of residents served.

Houses charging greater than $600 per month and serving male-specific populations increased the likelihood of alcohol abstinence. Houses serving fewer residents on parole/probation increased the likelihood of drug abstinence and total abstinence.

12-step oriented programs were associated with better outcomes.

Houses that were almost entirely or completely 12-step oriented, as reported by the house manager/operatorwere associated with greater likelihood of total abstinence and employment.

Arrests and drug/alcohol abstinence varied according to the house rules and policies.

Requiring at least 30 days of abstinence prior to resident intake was associated with decreased likelihood of arrest, whereas requiring AA/NA attendance was related to increased odds of abstinence.

Figure 1.


Several factors were unrelated to recovery outcomes.

Having a manager living on-site, providing meals to residents, requiring drug testing at intake, and the extent to which programs followed a social model approach (peer-oriented recovery approach guided by the principals of mutual help) to recovery housing did not significantly predict recovery outcomes.


This study suggests that certain characteristics of recovery residences are associated with better recovery outcomes, at least among high-risk residents involved in the criminal justice system, who were the participants in this study. Regarding organizational characteristics, this study suggests that recovery residences are associated with better recovery outcomes when houses are affiliated with larger housing organizations, treatment programs, and parole/probation referral programs. Perhaps affiliations with these organizations help to promote greater implementation of operations and procedures that have therapeutic benefits for participants and/or improves matching of the resident to their optimal housing environment, thereby enhancing recovery outcomes. Affiliations with treatment facilities might promote smooth transitions to recovery residences without periods lacking continued care. Recovery residences with such affiliations might ultimately mean greater linkages among recovery support services, creating a more robust recovery-oriented system of care. Providing a well-connected recovery infrastructure that supports the diverse needs of individuals with substance use disorders, including transitions throughout the recovery process, can enhance recovery outcomes.  

With respect to the residence, the authors found that location of the sober living house influenced recovery outcomes, which might reflect differences in community resources. Access to 12-step meetings, other peer support groups, as well as supplemental mental and physical health care have the potential to impact recovery outcomes. Better recovery outcomes were also associated with houses that more greatly implemented 12-step principles, further emphasizing the importance of peer support and mutual help in recovery progress.  

When operational and program orientation factors were assessed, findings suggested that requiring 30-day abstinence prior to recovery residence intake might decrease the likelihood of resident arrests, but drug testing at intake may not impact these or other recovery outcomes. Facilities housing a smaller number of residents were associated with higher rates of employment, which could reflect the benefits of a low resident to staff ratio, with greater focus and attention afforded to each individual. Alternatively, facilities that take on a greater number of residents may be able to charge less, thereby allowing a more affordable option and attracting a larger population of unemployed residents. Although houses with higher monthly rates, male-only residents, and fewer residents on parole/probation predicted better substance use outcomes, these factors may have more to do with the attributes of the residents served by particular recovery residences and less to do with the houses themselves. For example, individuals that can afford more expensive housing may have more recovery capital to begin with, thereby enhancing their potential for successful recovery outcomes independent of the facility’s monthly rate. 

It is important to recognize that these data do not speak to the causality of observed relationships. It may be that either these recovery outcomes are a result of specific recovery residence characteristics, or that houses with specific organizational and operating characteristics are more likely to serve particular populations that, in turn, are more likely to obtain employment, remain abstinent, and avoid criminal arrest. Nonetheless, this study provides an important foundation for future work to further investigate these characteristics and their role in recovery facilitation.

  1. Multiple recruitment sites (i.e., recovery residences) were sometimes part of the same organization. Though the authors conducted statistical tests to ensure its limited influence on outcomes, this could ultimately restrict differences between sites and bias outcomes.
  2. This study was conducted in the context of a randomized controlled trial. Though condition assignment (no intervention or motivational interviewing and case management) was controlled for and outcomes were generally similar between conditions, clinical trial participation has the potential to impact recovery outcomes. In addition, data describing the facilities were partially collected during interviews aimed at determining their eligibility as a recruitment site, omitting several descriptive measures of the facilities that could be important predictors of recovery outcomes. These factors warrant more research.
  3. The authors were assessing a special population of residents with criminal justice involvement and heightened HIV/AIDS risk, with the majority being white male residents. Moreover, all houses assessed were located in California. It is unclear whether these results would generalize to other areas of the country.


  • For individuals and families seeking recovery: The current study examined relationships between recovery residence characteristics and residents’ recovery outcomes. Though this study does not speak to the direct effects of recovery residences on recovery outcomes, their potential benefits have been previously observed in studies looking at different types of recovery housing (e.g., Oxford Houses). Still, the programs and models of recovery vary widely between houses, and low-quality houses have received substantial media attention. Therefore, this study may be helpful for individuals and families when searching for a recovery residence. Results suggest better outcomes for residences with larger organization affiliations (e.g., parent organizations, treatment facilities, probation programs) and houses that implemented a 12-step program. These attributes might work to support standard house procedures, encourage resident engagement in recovery-focused mutual help programs, facilitate continued care transitions and a good match between residents and houses, that ultimately enhance recovery outcomes. The authors also found that houses with lower resident capacity, male-only residents and fewer residents on parole/probation, and that charged higher house fees and required 30+ days of abstinence to become a resident were all associated with better outcomes. However, these outcomes may have less to do with the house characteristics and more to do with residents (e.g., individuals with these characteristics, who can afford more expensive housing with lower resident capacity, might enter a residence with more resources at their disposal). More research is needed to determine the direct effects of facility characteristics on outcomes, but this study provides an important first step in understanding the recovery residence characteristics that may be the greatest help to individuals seeking recovery.
  • For treatment professionals and treatment systems: This study found several relationships between recovery residence attributes and residents’ recovery outcomes (substance use, criminal justice involvement, employment). Residences that were affiliated with a larger organization (e.g., parent organizations, treatment facilities, probation programs) and had a stronger 12-step orientation were associated with better recovery outcomes. Parent programs might work to promote standard house procedures, facilitate resident-house matching and transitions to continued care, thereby increasing chances for positive recovery outcomes. In line with the reported benefits of mutual help programs on recovery outcomes more generally, this study suggests that encouraging or requiring 12-step participation as part of continued care is also beneficial in a recovery residence context. Although the authors reported better recovery outcomes among more expensive houses, requiring 30+ days of abstinence, with lower resident capacity, male-only residents and fewer residents on parole/probation, individuals with these characteristics who can afford more expensive lower-capacity housing might enter recovery residences with more recovery capital and an advantage for successful recovery. Further investigation will help to determine the direct effects of these facility and resident characteristics, but this study provides an important first step in understanding therecovery residence characteristics thatmight contribute to optimal recovery outcomes.
  • For scientists: This study examined relationships between recovery residence characteristics and residents’ recovery outcomes. The authors found that better recovery outcomes were associated with recovery residences that were affiliated with a larger organization (e.g., parent organizations, treatment facilities, probation programs) and implemented 12-step programming. These attributes might work to facilitate recovery outcomes by promoting house procedures that confer therapeutic benefits, encouraging recovery-focused activities among residents, and expediting patient-house matching and continued care linkages and transitions. The authors also found that houses with lower resident capacity, male-only residents and fewer residents on parole/probation, that charged higher house fees and required 30+ days of abstinence predicted better outcomes. However, it is unclear whether these outcomes are attributable to housing or resident characteristics, given that individuals who can afford more expensive housing with lower resident capacity might enter continued care with greater resources and, thus, greater opportunity for successful recovery. Additionally, this research was conducted in the context of a randomized controlled trial among a special population of residents with criminal justice involvement and heightened HIV/AIDS risk, and all houses assessed were located in California. Further investigation is needed to identify the direct effects of housing characteristics on recovery outcomes and determine whether this study’s findings apply to other populations with various substance use disorders and/or recovery residences in other states. Randomized controlled trials are also needed to identify more clearly the direct effects of housing itself, independent of housing characteristics. 
  • For policy makers: Research studies like this one provide important preliminary information about the programmatic attributes of treatment and recovery facilities that are likely to yield the most effective outcomes. This study suggests that recovery residences might benefit from affiliations with parent programs and implementation of mutual help programs. The study also speaks to the importance of the residents to which the facilities cater and their characteristics, as well as the location of the facility. However, the relationship between these factors and recovery outcomes are somewhat unclear given that this study cannot speak to cause and effect. Additional funding is needed to more clearly identify the organizational, operational, and programmatic attributes that promote efficient and successful recovery. Investigating these issues can ultimately help to identify necessary/unnecessary facility attributes to promote efficient fund allocation for optimal yield and best practices and elucidate how recovery residences might reduce the financial burden related to substance use disorder over the long-term.


Mericle, A. A., Mahoney, E., Korcha, R., Delucchi, K., & Polcin, D. L. (2019). Sober living house characteristics: A multilevel analyses of factors associated with improved outcomes. Journal of Substance Abuse Treatment, 98, 28-38. doi10.1016/j.jsat.2018.12.004