Become part of a movement that makes a difference. CCORR is a state-wide initiative providing standards and inspections of sober homes that covers safety, cleanliness and the effectiveness of house programming. It is there to help you choose the level of care that will make your time at the home successful when you are struggling to maintain your recovery. CCORR goes the extra mile to share the community resources for counseling, interventions, and the community groups for recovery. Through a voluntary process, Certified Residences provide CCORR unrestricted access to interview management, staff and residents to verify implementation of policies, procedures and protocols previously documented by the house. Certification to the CCORR Standard ensures, for all stakeholders, provider compliance with measures demonstrated to enhance the quality of recovery support in a community-based, residential setting.


We recommend careful consideration of CCORR support levels before commencing your selection process. One level is not better than another; rather it offers distinctly different support characteristics and intensity of service. The effort you invest to familiarize yourself with these options helps to ensure your selection of the support level that is most appropriate for your needs given the unique DNA of every home.

The role that service needs and problem severity play in admissions decisions varies widely within and across levels of recovery residences (See illustration below). There are also recovery residences designed specifically for individuals with certain needs (e.g., co-occurring addiction and severe mental illness, veterans, mothers with children); however, some recovery residences may not be equipped to adequately meet these residents’ needs. Individuals with specific service needs seeking RRs should ask the provider about how these needs can (or cannot) be addressed within a particular residence.

Level I

Characteristics of Level I residences coincide most closely with Oxford Houses, which have been studied extensively by Jason et al. over the past 20 years. It is a self-governing house without paid staff, but proven effective.

Level II  


Sober living houses (SLHs) similar to those that are members of the Sober Living Network (SLN) in Southern California and some houses affiliated with the California Association of Addiction and Recovery Resources (CAARR) are good examples of Level II residences.  Like Level I residences, studies on these types of facilities have been limited.   One of the few studies on Level II residences was a recent study of houses in Northern California (Polcin, Korcha, Bond, & Galloway, 2010).  Researchers recruited 245 individuals entering Clean and Sober Transitional Living in Sacramento County, which includes 16 recovery homes.  The houses were located in a very high methamphetamine (MA) use area and 53% of the participants entered the houses with dependence on MA during the past year. Participants were interviewed within 2 weeks of entering the houses and then at 6-, 12- and 18-month follow-up.  Primary outcomes 
A Primer on Recovery Residences – FAQs: NARR  September 20, 2012 22 
included measures of alcohol and drug use and Addiction Severity Index (ASI) alcohol and drug scales.  Secondary measures consisted of other ASI scales and a variety instruments assessing criminal justice involvement, employment, and psychiatric problems.      Longitudinal analyses revealed two patterns for primary and some secondary outcomes over time.  One pattern involved residents entering the SLHs with moderate to high severity of problems, making significant improvements by 6 months, and then maintaining those improvements at 12 and 18 months.  Results from measures that assessed alcohol and drug use over a 6-month time period showed this pattern.  For example, alcohol and drug abstinence over a 6-month time period increased from 20% at entry into the SLH to 40% at 6-month follow up.  Abstinence improved even more at 12month follow-up (45%) and declined only a bit at 18 months (42%).   The other outcome pattern showed residents entering the SLHs with low severity of problems at baseline and then maintaining low severity at 6-, 12-, and 18-month follow up.  Findings from the ASI alcohol and drug scales were good examples of this pattern.  The average score on the ASI alcohol severity at baseline was 0.16 (se=0.02), and for drug severity at baseline, the average was 0.08 (se=0.01).  Because ASI values range from 0 to 1, these scores are very low.  There was therefore limited room to improve on these measures. Nevertheless, there were significant improvements at 6 months for both alcohol (mean=0.10, se=0.02) and drug (mean=0.05, se=0.01) scales.  Those improvements were maintained at 12 and 18 months.  Alcohol severity remained at 0.10 at 12 and 18 months, and drug severity also remained essentially unchanged, 0.06 at 12 and 18 months. It should be noted that improvements were maintained at 12 and 18 months despite the fact that most residents had left the SLHs.  By 18 months, about 90% of the residents had left, yet there was little regression of the earlier improvements.  Thus, the improvements noted were therefore not simply a function of residents being housed in a controlled environment.         The study also examined a variety factors that predicted outcome.  These included demographic characteristics and factors related to the philosophy of recovery in SLHs, such as involvement in 12-step groups and developing a social network supportive of abstinence.  Generalized Estimating Equations showed that involvement in 12-step groups was the strongest and most consistent predictor of good outcome. As expected, drinking and drug use in the participant’s social network predicted worse outcome.   Overall, a wide variety of demographic groups made improvement in the SLHs and only a few demographic characteristics predicted outcome.  The most notable exception was the relationship between age and abstinence.  Older age categories were over twice as likely to be abstinent than those aged 18-28.  Relative to residents who had not completed high school, those with at least a high school diploma were nearly twice as likely to be abstinent over the past 6 months and about half as likely to be arrested.   


Level III  


A good example of level III residences are “social model” recovery programs, which emphasize experiential learning, peer support, and 12-step recovery principles within a semi-structured group living environment. These programs are more structured than level II residences and include paid counselors who assist residents with case management services and the development of a formal recovery or treatment plan. 
from: A Primer on Recovery Residences – FAQs: NARR  September 20, 2012 23 
Typically, there are various life skills and other types of groups offered at the facility.  In many states, such as California, they are licensed by the state to provide treatment services.   As with Phase I and Phase II residences, the outcome studies on Phase III residences have been fairly limited.  Studies that have been conducted include the California Drug and Alcohol Treatment Assessment (CALDATA; Gerstein et al., 1994) and studies on social model recovery programs conducted by Kaskutas et al. (2003-2004, 2008) at the Alcohol Research Group.    The CALDATA study examined treatment outcomes among 1,858 clients in California who received methadone treatment, non-methadone outpatient, clinically oriented residential treatment (21 providers), or social model recovery programs (23 providers).  The study consisted of follow-up interviews approximately 15 months after leaving treatment. Clinically oriented residential programs included procedures such as psychiatric assessments, individual counseling, and treatment groups (e.g., therapeutic communities).  Social model recovery houses were oriented toward peer support, communal living, and practicing 12-step recovery principles. Borkman, Kaskutas, Room, Bryan, & Barrows (1998) compared the two types of residential programs and reported that residents in social model programs had longer stays and incurred lower costs.  Costs per treatment episode in the social model programs averaged $2,712, while costs per treatment episode in the clinical residential program averaged $4,405. Overall length of stay was associated with better outcome. Comparison of residents in the two types of residential programs who had comparable lengths of stay showed slightly better outcomes for the clinically oriented programs.  For example, residents who remained in treatment 4+ months in the clinically oriented program reduced the number of months they used substances by 63% while social model residents reduced the number of months of substance use by 52%.   Reductions in reports of criminal activity were slightly higher in social model programs (80%) than clinically oriented programs (74%).   Studies conducted by Kaskutas et al. (2003-2004, 2008) were stronger designs because they included longitudinal designs that compared measures collected at treatment entry with follow-up measures.  The 2003-2004 study consisted of a naturalistic comparison of outcomes for individuals in social model residential programs (N=164) with those in clinically oriented programs N=558).  The social model programs were detoxification and residential facilities, and the clinically oriented programs were a mix of inpatient and outpatient.  Individuals in the social model programs were more involved in 12-step meetings and reported fewer alcohol and drug problems at one-year follow up, but not problems between the two study conditions in other areas (e.g., family, medical, legal, and psychiatric).  The 2008 study randomly assigned clients to receive day hospital program treatment (n=154) or services in social model residential programs (n=139).  Although significant improvements were noted at 12 months for both groups, between group comparisons did not reveal significant differences.  Overall, clients tended to remain in the residential programs longer and costs were higher.

Level IV

Relative to other levels, Level IV residences include more structure, paid professional staff and on-site treatment services. Residential therapeutic communities (TCs) for drug treatment are a good example of Level four residences.  Large proportions (A Primer on Recovery Residences – FAQs: NARR  September 20, 2012 24) of residents in TCs are referred from the criminal justice system, and some are actually located in prisons.   TCs have a long history of participating in research, including large national studies assessing drug treatment outcome.  These studies include the Drug Abuse Treatment Outcome Study [DATOS] (Hubbard, Craddock, Flynn, Anderson, & Etheridge, 1997), National Treatment Improvement Evaluation Study [NTIES] (Center for Substance Abuse Treatment, 1996), Treatment Outcome Prospective Study [TOPS] (Hubbard et al., 1984), and Drug Abuse Reporting Problems [DARP] (Simpson & Friend, 1988).  Overall, these and other studies on TCs (e.g., Martin, O’Commel, Paternoster, & Bachman, 2011) show that clients make longitudinal improvements on substance use measures, arrests, illegal behaviors and employment.   When TCs have been compared to voluntary, control, or alternative treatment groups, the findings have been encouraging.  For example, DeLeon (1988) found that clients referred from the criminal justice system stayed in treatment longer than voluntary clients and had levels of improvement that were similar. Prendergast, Hall, Wexler, Melnick, & Cao (2004) conducted a randomized trial of 715 prisoners randomly assigned to either a therapeutic community program or to a no treatment group. At 5-year followup, the TC group had significantly lower rates of reincarceration, but not shorter times to first reincarceration.  As in most studies of TCs, longer lengths of treatment were associated with better outcome.   Martin, Butzin, & Inciardi (1995) studied 457 individuals participating in either an in-prison TC, a TC in the community, both types of TCs, or a no treatment comparison group.  Those attending the community-based TC or both types of TC had the best outcome (substance use and re-arrest). The in-prison TC had modestly better outcomes than the no treatment comparison group.    
Summary of Outcomes   There is obviously a significant need for additional research on residential recovery homes, especially those characterized by levels 1-3.  Studies on level 4 residences are more numerous because of the large number of studies examining outcomes within therapeutic communities.  Overall, the available studies across the different levels are encouraging.  Longitudinal studies of residents housed within each of the levels show improvements in a range of areas.  When comparisons have been made between recovery residences and appropriate alternatives, the results have shown recovery homes yield comparable or better outcomes.  Cost and cost-benefit analyses have been limited and to have yielded mixed findings.  
Borkman, T. J., Kaskutas, L. A., Room, J., Bryan, K., & Barrows, D. (1998). An historical and developmental analysis of social model programs. Journal of Substance Abuse Treatment, 15(1), 7-17. doi:10.1016/S0740-5472(97)00244-4 Center for Substance Abuse Treatment. (1996). National Treatment Improvement Evaluation Study (NTIES). Rockville, MD: U.S. Department of Health and Human Services