Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Apr 1.
Published in final edited form as: Int J Drug Policy. 2024 Feb 21;126:104360. doi: 10.1016/j.drugpo.2024.104360

Effects of Strength of Relationship Ties in Recovery Homes: A Conundrum

Leonard A Jason 1, John M Light 2, Ted Bobak 3, Justin Bell 4
PMCID: PMC11056307  NIHMSID: NIHMS1969101  PMID: 38382353

Abstract

Recovery homes are a widespread source of support for those attempting to maintain abstinence. For those who are able to remain in these settings for at least 6 months, outcomes tend to be favorable; however, many leave prematurely. There is a need to better understand the social integration processes that play a major role in giving recovery home residents access to available recovery-related social capital that is associated with better outcomes. The current study involved Oxford House recovery homes in 3 states and examined the strength of relationship ties among house members. We found that those who associated with peers who have higher recovery scores tend to improve their own recovery scores over time. However, we also found that those with higher recovery scores tended to create “strong” ties with similarly high-scoring alters; likewise lower-scoring individuals preferentially formed strong ties with each other. These findings suggest a conundrum: recovery home residents most in need of relational support from more recovered housemates are the least likely to obtain it. We discuss possible pathways to creating more ties between high and low-recovered residents.

Many with substance use disorders require post-acute care in a supportive and cohesive setting (Cloud & Granfield, 2008; Moos, 2008; Vaillant, 1983; Zywiak, Longabaugh, & Wirtz, 2002). The odds of a successful recovery often depend on recovery supports that are available in one’s social environment. For example, Sliedrecht et al. (2019) found a significant recovery-supportive effect of social support in 44 out of 50 studies. Recovery homes are a widespread post-acute care residential setting (Jason et al., 2020). These settings also provide residents with critical post-treatment support (Bliuc, Best, Iqbal & Upton, 2017), and there is evidence that they specifically promote recovery (Brereton et al., 2014; Jason & Ferrari, 2010). Although there is evidence that a six-month stay is needed to promote lasting recovery (Jason et al., 2007), many residents leave earlier; indeed, the risk of departure is higher the less time an individual has been in residence (Jason, Bobak, Light, & Stoolmiller, 2023). The processes that influence whether residents maintain their residency in these settings for that critical 6-month period of time are not well understood.

There is a need for investigations of the micro-level, interpersonal social mechanisms that may lead to improved recovery outcomes in these sober living homes. By identifying the micro-social mechanisms, we might be able to determine the circumstances that promote ongoing recovery in such settings. Such mechanisms can be examined by analyzing predictors of the evolution of social relationships among residents. For example, Light et al. (2016) found that close friend relationships were more likely to form in dyads already characterized by a willingness to loan the other person at least $100. Jason et al. (2014) reported that willingness to loan at least $100 was predicted by 12-step involvement, suggesting that progress in one’s recovery may affect other residents’ willingness to accept him or her as a trusted member of the house social system. However, it is those individuals whose recovery is least well-developed who, arguably, most need access to the recovery social capital available in a sober living home environment. Thus, the question remains, how do individuals in early recovery integrate into this environment?

One framework for understanding the flow of facilitating resources that are available in recovery homes is social capital theory (Bliuc, Best, Iqbal & Upton, 2017; Granfield & Cloud, 2001). Social capital comprises the potential resources connected to a durable network of relationships of mutual acquaintance (Bourdieu, 1985). Kelly and Hoeppner (2014) have suggested that there is positive feedback between maintaining sobriety and recovery capital over time, and evidence continues to mount for this mechanism (Burns & Marks, 2013; Granfield & Cloud, 2008; Laudet & White, 2018). There is a need to better understand where recovery capital originates. In the course of addressing social integration for less-recovered sober living home residents, there is also a need for a model in which social relationships and recovery progress interact with each other over time.

Studies have demonstrated that social relationship dynamics affect individual recovery outcomes and vice-versa. Doogan et al. (2019) found that Oxford House recovery home residents with a high quality of life tend to nominate as friends or close friends with others with a lower quality of life. This finding suggests that Oxford House recovery homes may be able to encourage friendship ties to residents most in need of recovery support. Jason, Bobak, Islam, Guerrero, and Light (2022) later found that being friends with others who are more recovered improves one’s own recovery. Also, while seeking advice from other house residents generally is associated with negative outcomes, seeking advice from those who are more recovered leads to increases in the advice seekers’ recovery (Jason, Lynch, Bobak, Light, & Doogan, 2022). In sum, there is growing evidence that social relationship dynamics affect individual recovery outcomes and these outcomes affect social relationships.

In the current article, we explore the role of relationships among recovery home residents characterized as “strong” or “very strong”. Our interest in relationship strength stems from the possibility that a relationship may be strong, but not necessarily involve friendship per se, as the latter tend to be between equals. For example, an Alcoholics Anonymous-based “sponsor” relationship may often be asymmetrical, pairing a more experienced recovering individual with one less experienced, with the objective that the former offers advice and guidance to the latter during the course of his or her recovery (Stevens & Jason, 2015). Hence, it is possible that the strength of the relationship may be measuring intimacy and disclosure more akin to a sponsor-sponsee relationship, whereas friendships are thought of as more informal, symmetric, comfortable, and “social”. Thus, thestrength of a relationship tie could more accurately reflect a relationship wherein a more recovered resident provides access to recovery social capital to a less-recovered alter. Such a relationship may be more specifically directed towards the less-recovered individual’s recovery, and thus be a more effective resource than friendship for promoting such recovery. As such, it would represent an extension of the study by Doogan et al. (2019) mentioned earlier, which examined the dynamics and consequences of friendship formation. Based on results from that study, we expected to find a similar but even stronger tendency for higher-recovery factor individuals to preferentially form strong relationships with lower-recovery factor alters (or “heterophily”, that is, preference for someone unlike oneself), along with a similar positive-influence effect on the latter’s recovery, thus providing evidence for a strong pathway to recovery social capital for more in-need residents.

Method

Settings

The study was conducted in a set of Oxford Houses (OH), an organization comprising over 3,000 homes that are self-governed with no professional staff. Each house is rented and gender-segregated, housing individuals in recovery. Residents can stay in these settings as long as they remain abstinent, pay weekly rent, and follow house rules. Housing expenses average about $150 per week. Residents enter OH from a variety of acute care and criminal justice settings, including in-patient treatment centers, incarceration, and drug court programs. Some enter OH directly from homelessness or other precarious living situations (Jason et al., 2007; Whipple et al., 2016). Individuals looking to live in OH must be voted in by current residents. An 80% majority is required to accept a new resident (Jason & Ferrari, 2010).

OHs in three sections of the country were involved in this study (East, Southwest, and West). Longitudinal data were collected in 42 houses (See Jason et al., 2020 for more details). Data were collected every four months for two years, and participants were paid $20 for each interview. Self-governing environments such as Oxford Houses are often specifically intended to provide the motivation and methods required for recovery from substance use disorders.

Measures

Race/ethnicity was broken down into four categories: White (78.8%), Black (8.6%), Latinx (10.1%), and all others (2.5%). In preliminary analyses, using White as the reference group, the contrasts for Black, Latinx and all others were negligible. Accordingly, ethnic contrasts were simplified to either not-Black (reference group including White, Latinx, etc.) vs. Black. OHs are gendered and, accordingly, gender was included as a house-level predictor. We also evaluated residents’ employment status as employed, unemployed, or other forms of income (disability, student, retirement). We classified residents as either having a high school education, GED, or less vs. those with some college, technology school, or college degree. Prior time in OH residence was assessed at participants’ first wave.

To calculate total months in residence, we combined the resident-reported number of months in residence before their first baseline survey with the number of months in residence during the study. For logistical reasons, actual departure dates were not collected, so departure dates were approximated by adding two months to the date of the resident’s last survey. To reduce skewness and protect against high-leverage outliers, we log transformed this variable.

The recovery factor (RF) behavior variable was a factor score based on a confirmatory factor analysis of several recovery capital indicators (Jason et al., 2020). The analysis included each resident’s recovery factor score over time. This measure was calculated from the following instruments: wages, quality of life (WHOQOL Group, 1998), self-efficacy (Sklar et al., 1999), self-esteem (Rosenberg, 1965), stress (Cohen et al., 1983), social support (Cohen & Wills, 1985; Cohen, Mermelstein et al., 1985), sense of community (Jason et al., 2015), and hope (Snyder et al., 1996). The continuous scores were recoded to ordinal categories ranging from 0 to 4.

Social Network data were collected using a Social Network Instrument (SNI; Jason & Stevens, 2017; Jason et al., 2014; Jason & Stevens, 2017; Light et al., 2016). A whole network approach was used whereby all residents of the house rated each other on different relationship domains, and in this study, we focused on the strength of relationship. Strength of relationship was operationalized by asking, “Overall, how strong would you relate your relationship with this person?” and was considered a tie (value 1) if it was rated very strong or strong, but not a tie (value 0) if it was related weak, none, or negative. In a whole network, each resident rates all other residents, and from these ratings, we were able to establish measures of Density (dividing existing ties or connections by all potential ties), Reciprocity (proportion of ties that are reciprocated), Transitivity (describes triads and thus network clustering) and Three-Cycles (the absence of hierarchy) (See Jason et al., 2020, for more details).

Statistical Approach

We used the R package RSiena (see https://www.r-project.org) to estimate Stochastic Actor-Oriented Models (SAOMs) (Snijders, van de Bunt, & Steglich, 2010). This modeling framework conceives of social environments as consisting of one or more sets of time-varying relationships among recovery home residents, as well as one or more time-varying behaviors (changeable individual characteristics), which evolve as a function of both endogenous and exogenous predictors. For instance, endogenous network characteristics can include tendencies toward density, reciprocity, and transitivity of a given relationship. Individual characteristics can be endogenous (like the recovery factor in this study), while others, such as employment status and current length of residence, can be time-varying and exogenous, and others still may be exogenous and fixed (such as sex). SAOMs can be thought of as a set of interrelated multinomial logistic regressions (one for each endogenous network and behavior variable), where the overall model is a variable-dependent stochastic process. Model estimation was carried out using an adapted method of moments approach with a Robbins-Monro algorithm (for details, see Snijders et al., 2010).

The tendency for heterophily on RF in the selection process can be tested by a negative interaction between RF ego and RF alter in the network (strong tie) part of the model, controlling also for the main effects of RF ego and RF alter. An average alter effect was included in the behavior (RF) part, to see whether having strong ties with alters whose RF is higher than one’s own predicts RF increase over time. A selection table (Snijders, 2022) and associated graph are presented, showing model predictions related to the heterophily hypothesis.

Results

Of 714 eligible participants, 93% (N = 666) agreed to participate, of whom 602 are included in the current study1. Regarding sex, 51% were male and 49% were female, with an average age of 37.0 (SD = 10.5). Most of the sample was White (78.8%), while 10% were Latinx and 8.5% were Black. Asian American, Alaskan Native, American Indian, and Pacific Islanders accounted for the remaining 2.7%. Characteristics of the sample are located for the first interview in Table 1a and by wave in Table 2.

Table 1a.

Sample Characteristics at first interview.

Total Number of Residents 627
Age 37.7 (10.8)A
Sex - Female 0.48 (0.50)
African American 0.08 (0.28)
Time in Residence (months) 6.13 (9.26)
High School Grad or more 0.44 (0.50)
Number prior SUTx episodes 4.62 (5.54)
Number of waves participated (of 6 1.91 (1.36)
Open in a new tab

Table 2.

Stochastic Actor-Oriented Model Results—method of moments estimation

Effect Name Parameter Est SE P value 95% Confidence Interval Convergence t-ratio
Network Dynamics
constant rate (period 1) 3.31 0.64 <.01 (2.03 4.58) 0.03
constant rate )(period 2) 2.85 0.87 <.01 (1.14 4.55) −0.02
constant rate (period 3) 4.02 0.98 <.01 (2.09 5.94) 0.01
constant rate (period 4) 2.33 0.66 <.01 (1.04 3.62) 0.00
constant rate (period 5) 5.81 2.54 0.02 (0.83 10.79) −0.03
outdegree (density) −0.37 0.24 0.13 (−0.85 0.11) −0.02
reciprocity 1.36 0.23 <.01 (0.92 1.80) −0.03
transitive triplets 0.33 0.08 <.01 (0.18 0.49) −0.03
3-cycles −0.62 0.13 <.01 (−0.88 −0.37) −0.03
employ alter 0.66 0.19 <.01 (0.29 1.03) −0.04
Length time ego 0.31 0.06 <.01 (0.19 0.43) 0.01
sex ego −0.28 0.16 0.07 (−0.59 0.03) 0.00
RF alter −0.07 0.17 0.68 (−0.40 0.68) −0.05
RF ego −0.03 0.16 0.85 (−0.34 0.28) 0.05
RF ego x RF alter 0.58 0.28 0.04 (0.03 1.13) −0.02
Behavior Dynamics
rate RF (period 1) 1.51 0.31 <.01 (0.90 2.12) 0.00
rate RF (period 2) 1.20 0.26 <.01 (0.68 1.72) 0.04
rate RF (period 3) 1.55 0.31 <.01 (0.94 2.16) −0.03
rate RF (period 4) 1.37 0.27 <.01 (0.84 1.90) −0.01
rate RF (period 5) 1.77 0.40 <.01 (0.99 2.55) 0.04
RF linear shape 0.22 0.08 <.01 (0.07 0.37) −0.02
RF quadratic shape −0.49 0.08 <.01 (−0.65 −0.33) −0.01
RF average alter 0.53 0.20 0.01 (0.14 0.92) 0.00
RF: effect from RS.Blk 0.66 0.24 0.01 (0.19 1.13) −0.01
Open in a new tab

Overall Convergence t-ratio: 0.12

Table 2 shows model results with p values and parameter estimates, based on 3961 iterations during the estimation routine. Model convergence was .13, and all individual parameter convergence t ratios were .05 or less (Schweinberger, 2012). These values indicate good convergence (Ripley et al., 2023).

Network Dynamics

The network part of the model in Table 2 has the strength of relationship as the endogenous (outcome) variable. The outcomes reflect symmetric predictors of formation or maintenance (positive parameters) and dissolution (negative parameters) of these ties over time.

Endogenous network effects involve network evolution tendencies that need to be controlled so as not to be confounded with the primary hypothesis-related effects but can nevertheless sometimes be informative in their own right. These include outdegree, reciprocity, transitive triplets, and 3-cycles. The Outdegree (density) effect is primarily an indicator of a tendency toward a higher or lower proportion of nonzero “strong” ties relative to a midpoint of 50%. The fact that it is nonsignificant for strength of relationships suggests that these nominations are neither particularly likely to form nor particularly unlikely; rather the overall probability is indistinguishable from 50–50.

In contrast, Reciprocity was significant and positive (b=1.36, [0.92, 1.80]), indicating a tendency for relationship strength ties to be reciprocal. We also found a significant effect for Transitive triplets (b=0.33, [0.18, 0.49]), which suggests that having at least one strong relationship is one possible path to others. The 3-cycle effect was also significant (b=−0.62, [−0.88, −0.37]) which points to some movement towards hierarchical and non-closed triads. Taken together, the transitivity and 3-cycle effects suggest that strength of relationship ties has some tendency to fragment residential networks into mutual strength of relationship subgroups; however, there is also a concomitant tendency for certain individuals to have strength of relationship ties to others with no expectation of receiving the same treatment in return.

Other covariates besides the hypothesized RF variable are useful to include if there is the possibility they could be correlated with RF, and thus present an alternative interpretation of any RF effects. These included employment and length-of-stay at ego’s first survey (i.e. previous time as a resident in that home). Employment of the alter was significant (b=0.66, [0.29, 1.03]) but not of the ego (not shown), indicating that employed residents are more attractive targets for strong ties. The significant length-of-stay effect suggests that the longer someone has lived in the home prior to their first survey in the study, the more likely they are to form or maintain strong ties. The length of time in the OH variable was significant for the ego (b=0.31, [0.19, 0.43]) indicating that the ego’s (the chooser) time in residence is predictive, and those with longer residence were more likely to offer (out-choices) and maintain stronger relationships. Having more time in residence may allow individuals to get to know and develop stronger relationships with other residents; additionally, it is a useful control covariate, since it is likely to be correlated with (yet conceptually distinct from) recovery.

We included sex in the model because experience has suggested that network dynamics often differ for male and female residences. Sex approached significance and was a negative value (b=−0.28, [−0.59, 0.03]), which we judged close enough to a conventional .05 cutoff to require inclusion in the model. This finding means that males tend to create stronger relationships more frequently than females.

The ego and alter RF effects were not significant, but their interaction was (b=0.58, [0.03, 1.14]). This result is the opposite of what was hypothesized, implying that there is homophily (not heterophily) on RF for this type of relationship—that is, those with higher RFs appear to form strong relationships with each other, as do with lower RFs.

Figure 1 shows the model-based predicted relatively likelihoods (in units of the model objective function, cf. Snijders et al., 2010, for details) of strong ties being reported by an ego with a particular RF score (different-colored lines as per the legend) with alters with a particular RF score (horizontal axis values). The fundamental takeaway from Figure 1 is that individuals with higher RF scores tend to prefer other higher-RF individuals for strong ties, and also lower-RF individuals are most likely to nominate other lower-RF residents as strong ties. This pattern is referred to as “homophily” (a preference for someone like yourself). The graph is slightly asymmetric horizontally, with the confluence of the four ego-RF lines coming to the right of the RF scale midpoint of 2.5. This suggests that, although both high-RF and low-RF individuals’ preferences for strong ties with similar alters factor into the observed pattern, low-RF individuals’ preferences contribute somewhat more.

Figure 1.

Figure 1.

Open in a new tab

Relative likelihood of a strong/very strong tie forming between egos (line colors) and alters (horizontal axis values) with different recovery factor (RF) scores.

Behavior Dynamics

We found that RF for the average alter was significant (b=0.53, [0.15, 0.90]), which is a measure of how ego’s score is affected by whether alters’ scores are on average greater or lesser than ego’s. The positive parameter means that individuals who nominate alters with higher RFs than their own tend to increase their RFs over time. Similarly, nominating lower-RF alters will tend to decrease their RFs.

For the racial variable, a significant and positive value (b=0.66, [0.17, 1.16]), means Blacks are more likely to improve more on RF than other racial groups. The average alter effect thus holds within racial groups, and race does not moderate the effect (results not shown).

Discussion

This study addressed the question of how the formation of strong or very strong relationships was affected by a recovery home resident’s state of recovery. We found that those who associate with peers who have higher recovery factors tend to do better over time. But we also found that very strong relationships are most likely to occur between individuals who are equally recovered. Such preference for others similar to one’s self is termed homophily.

This finding was somewhat unexpected, given a previous study on the same data set by Doogan et al. (2019), which found heterophily (preference for someone unlike one’s self) on a rating of personal quality of life. This difference may reflect the more narrow focus on quality of life; although improvement in the latter is certainly a hoped-for element of recovery, it is not unreasonable that other elements alone or in combination conspire to impede the formation of new, recovery-supportive relationships, which as the earlier-described model shows, contribute to improvement in recovery while in residence. For example, less-recovered individuals may be less trusting of others generally, less willing to be accountable for responsibilities that go with OH residence, or more deeply enmeshed in their own problems and hence less sociable. Such factors could also contribute to being less attractive candidates for strong ties from their more-recovered counterparts.

We describe this finding as presenting a conundrum because this study also found—consistent with previous with friendship (Jason, Bobak et al., 2021) and advice-seeking (Jason, Lynch et al., 2021) as the relationship types—that having a relationship with a more-recovered alter tends to lead to improvements in a resident’s recovery. However, finding homophily on our recovery index with respect to strong relationships implies that more recovered residents tend to form such relationships with each other, rather than with other less-recovered residents, even though the latter are presumably more in need of recovery support. Moreover, the symmetry of this finding is also consistent with tendencies for less-recovered individuals to form strong relationships with each other, which is predicted by our model to present a barrier to recovery.

Taken together, these results may provide at least a partial explanation for why so many recovery home residents fail to maintain residence for the 6-month minimum that research indicates is necessary to reap the benefits of such residence for sustained recovery (Jason et al., 2007). The question then becomes, how are (particularly) new, less-recovered residents able to form supportive relationships with their more-recovered housemates, and how might such relationship formation be promoted? A recent study (Jason, Bobak, Light, & Stoolmiller, 2023) may offer a promising direction. That study found that relatively new OH recovery home entrants who report having an “important person” (described as someone who was “important to you” and with whom they had had contact in the previous 3 months) living in an Oxford Home (not even necessarily the same home as the resident doing the reporting) will tend to stay longer than residents without such an important person. This finding may point the way toward a deeper understanding of pathways to residence retention. For example, perhaps having such an important person makes a resident feel responsible to someone other than themselves for continuing to pursue recovery. Another possibility is that this important person has been able to supply some pre- or ongoing socialization to help a new resident navigate their home’s social system and find a way to form supportive relationships. In either case, perhaps a more planned/programmatic approach to helping a new resident fit in would be beneficial to retention.

It is also possible that other individual characteristics may prove beneficial for early-recovery new residents. These might include substance abuse treatment history, recovery orientation, family support, personality characteristics consistent with sociability, and others. Previous studies, for instance, implicate ‘recovery orientation’, or whether a resident follows an explicitly abstinence-based philosophy, as a possible influence on the acceptance of a new resident. Oxford House is strongly influenced by abstinence-based groups like Alcoholics Anonymous, and research consistently finds a vast majority of OH residents attend these groups (Groh et al., 2009, Bell et al., 2022). When new residents violate the tenets of abstinence-oriented recovery, they may have more difficulty forming relationships. For example, research on 12-step group members finds negative attitudes toward users of the medications (Andraka-Christou et al., 2022). Like any social group, various implicit and explicit norms govern the social relations of Oxford House. Future studies should explore characteristics that allow new residents to fit into the environment of Oxford House.

The bottom line, however, seems to be that a “successful” (extended to at least six months) stay in a sober living home like Oxford House appears to depend primarily on a resident’s ability to form recovery-supportive relationships with other residents. If this ability is lacking in the population of potential residents, a logical possibility to consider is to offer some form of extra facilitative support, with the hope that such support can improve home retention and concomitant recovery outcomes.

We also found that longer-time residents are more likely to propose or maintain strong-relation ties, but the length of stay did not affect how likely someone was to receive such a tie proposal (model not shown). This pattern could mean that newcomers comprise a number of these longer-time residents’ incoming ties (which would attenuate any length-of-stay homophily), but that these ties are not typically reciprocated. This would in turn suggest that the longer-time residents do indeed prefer strong ties with similarly longer-time alters, but they are attractive to both longer and shorter-time alters, the latter under some as-yet-unknown conditions.

There are several limitations in this study. First, the findings pertain to those living in self-governing Oxford House recovery homes, and the outcomes found in this investigation might not generalize to other types of recovery homes. Second, we did not examine what occurred following departure from the recovery homes, and as a full recovery may require as much as 5–10 years (Laudet, Savage, & Mahmood, 2002), there is a need to examine recovery home benefits in this longer-term context. Our findings suggest that men tend to create stronger relationships than women and that Blacks improve more than non-Blacks, but these findings need to be explored in more depth in future studies.

Our article does not indicate how many people with a low RF change to higher RF and develop stronger relationships or how many with lower RFs barely change their score over time. It is possible that certain aspects of the RF, such as hope, predict how likely someone is to develop relationships. The hope measure in the RF might begin to get at mental health issues that are prevalent in this population and affect people’s capacity to build strong relationships in houses and improve their RF scores. It is also possible that there are other recovery resources (e.g., personality, self-help groups, sense of community) that might help us analyze changing supportive relationships in the OHs.

In summary, those who associate with peers who have higher recovery factors tend to do better over time. However, we found a marked tendency for those with higher recovery scores to associate with each other, rather than with their more in-need, lower-recovery housemates. The challenge for investigators is to better understand conditions upon which those with lower recovery factors can form ties with those with higher recovery factors, which appears necessary to enhance their recovery. Ultimately, this knowledge might help investigators know how this recovery home experience might be improved or broadened to other types of settings.

Table 1b:

Sample characteristics by waveB.

Wave: 1 2 3 4 5 6
Number of Residents 229 175 198 176 186 183
Number of Residents per house 5.45 4.40 5.23 4.90 5.26 5.37
African American 0.09 0.08 0.09 0.08 0.05 0.08
Recovery Factor Score 2.60 2.64 2.64 2.67 2.70 2.81
SORC Tie Density 0.70 0.79 0.74 0.74 0.76 0.77
SOR Reciprocity 0.74 0.83 0.78 0.75 0.83 0.79
SOR Transitivity 0.75 0.82 0.76 0.83 0.78 0.83
SOR Average Degree 6.02 5.53 5.96 5.78 6.43 6.69
Jaccard IndexD --- 0.20 0.12 0.18 0.13 0.19
Open in a new tab
A.

Mean (standard deviation)

B.

Mean, proportion, or value based on participating individuals

C.

Strength Of Relationship network ties

Proportion of (non-missing) changed ties from the previous wave.

Highlights.

  • The social integration processes that play a major role in giving recovery home residents access to available recovery-related social capital that is associated with better outcomes.

  • We found that recovery residents who associated with peers who have higher recovery scores tend to improve their own recovery scores over time.

  • Yet, recovery home residents most in need of relational support from more recovered housemates are the least likely to obtain it.

  • We discuss possible pathways to creating more ties between high and low-recovered residents.

Acknowledgments:

The authors appreciate the financial support from the National Institute on Alcohol Abuse and Alcoholism (grant number AA022763). We also acknowledge the help of several members of the Oxford House organization, and in particular Alex Snowden, Casey Longan, and Howard Wilkins.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

There are no conflicts of interest with this paper.

Ethics approval

DePaul University Institutional Review Board

1

34 residents entered an OH for the first time at wave 7, and we were not able to determine whether they relapsed given that the last data collection occurred during wave 7; 15 were forced out because their OH closed. before wave 7 and before they filled out a second survey; 15 were missing their reason for leaving and had only filled out 1 survey before leaving.

Contributor Information

Leonard A. Jason, DePaul University

John M. Light, Oregon Research Institute

Ted Bobak, University of Washington.

Justin Bell, DePaul University.

References

  1. Andraka-Christou B, Totaram R, & Randall-Kosich O (2022). Stigmatization of medications for opioid use disorder in 12-step support groups and participant responses. Substance Abuse, 43(1), 415–424. 10.1080/08897077.2021.1944957 [DOI] [PubMed] [Google Scholar]
  2. Bell JS, Islam M, Bobak T, Ferrari JR, & Jason LA (2023). Spiritual awakening in 12-step recovery: Impact among residential aftercare residents. Spirituality in Clinical Practice, 10(4), 337–349. 10.1037/scp0000296 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Best D, Irving J, & Albertson K (2016). Recovery and desistance: what the emerging recovery movement in the alcohol and drug area can learn from models of desistance from offending. Addiction Research & Theory, 25(1), 1–10. [Google Scholar]
  4. Best D, Vanderplasschen W, & Nisic M (2020). Measuring capital in active addiction and recovery: The development of the Strengths And Barriers Recovery Scale (SABRS). Preprint. [DOI] [PMC free article] [PubMed]
  5. Bliuc A, Best D, Iqbal M, & Upton K (2017). Building addiction recovery capital through online participation in a recovery community. Social Science & Medicine, 193, 110–117. [DOI] [PubMed] [Google Scholar]
  6. Block P (2018). Network Evolution and Social Situations. Sociological Science , 45, 402–431. http://dx.doi.org/doi: 10.15195/v5.a18 [DOI] [Google Scholar]
  7. Bourdieu P (1985). The forms of capital. Handbook of theory and research for the sociology of education. Richardson JG. New York, Greenwood, 241(258), 19. [Google Scholar]
  8. Brereton KL, Alvarez J, Jason LA, Stevens EB, Dyson VB, McNeilly C, & Ferrari JR (2014). Reciprocal responsibility and social support among women in substance use recovery. International Journal of Self-Help & Self-Care, 8, 239–257. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Burns J & Marks D (2013) Can recovery capital predict Addiction Problem Severity?, Alcoholism Treatment Quarterly, 31, 3, 303–320. [Google Scholar]
  10. Cloud W, & Granfield R (2008). Conceptualizing recovery capital: Expansion of a theoretical construct. Substance Use & Misuse, 43(12–13), 1971–1986. [DOI] [PubMed] [Google Scholar]
  11. Cohen S, Kamarck T, & Mermelstein R (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385–396. [PubMed] [Google Scholar]
  12. Cohen S, Mermelstein R Kamarck T, & Hoberman H (1985). Measuring the functional components of social support. In Sarason IG & Sarason BR (Eds.), Social support: Theory, research and application (pp. 73–94). The Hague, Holland: Nijhoff Martinus. [Google Scholar]
  13. Cohen S, & Wills TA (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98, 310–357. doi: 10.1037//0033-2909.98.2.310 [DOI] [PubMed] [Google Scholar]
  14. Doogan NJ, Light JM, Stevens EB, & Jason LA (2019). Quality of life as a predictor of social relationships in Oxford House. Journal of Substance Abuse Treatment, 101, 79–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Granfield R, & Cloud W (2001). Social context and “natural recovery”: The role of social capital in the resolution of drug associated problems. Substance Use & Misuse, 36, 1543–1570. doi: 10.1081/JA-100106963 [DOI] [PubMed] [Google Scholar]
  16. Granfield R, & Cloud W (2008). Conceptualizing recovery capital: Expansion of a theoretical construct. Substance Use & Misuse, 43, 1971–1986. [DOI] [PubMed] [Google Scholar]
  17. Groh DR, Jason LA, Ferrari JR, & Davis MI (2009). Oxford House and alcoholics anonymous: The impact of two mutual-help models on abstinence. Journal of Groups in Addiction & Recovery, 4(1–2), 23–31. 10.1080/15560350802712363 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hubbard RL, Craddock SG, & Anderson J (2003). Overview of 5-year followup outcomes in the drug abuse treatment outcome studies (DATOS). Journal of Substance Abuse Treatment, 25(3), 125–134. doi: 10.1016/s0740-5472(03)00130-2 [DOI] [PubMed] [Google Scholar]
  19. Jason LA, Bobak T, Islam M, Guerrero M, Light JM (2021). Exploring possible network properties facilitating recovery for residents of sober living homes. International Medicine, 3(4), 122–128. 10.5455/im.79688 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Jason LA, Bobak T, Islam M, Guerrero M, & Light JM (2022). Social integration in recovery living environments: A dynamic network approach. Journal of Community Psychology, 50(3), 1616–1625. 10.1002/jcop.22739 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Jason LA, Bobak T, Light J, & Stoolmiller M (2023). Understanding length of stay in recovery homes. Manuscript submitted for publication.
  22. Jason LA, & Ferrari JR (2010). Oxford House recovery homes: Characteristics and effectiveness. Psychological Services, 7, 92–102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jason LA, Guerrero M, Lynch G, Stevens E, Salomon-Amend M, & Light JN (2020). Recovery home networks as social capital. Journal of Community Psychology, 48, 645–657. 10.1002/jcop.22277 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Jason LA, Guerrero M, Salomon-Amend M, Light JN & Stoolmiller M (2021). Personal and environmental social capital predictors of relapse following departure from recovery homes. Drugs: Education, Prevention & Policy, 28(5), 504–510. 10.1080/09687637.2020.1856787. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Jason LA, Guerrero M, Salomon-Amend M, Lynch G, Stevens E, Light JM, & Stoolmiller M (2021). Advice seeking and loaning of money related to relapse in recovery homes. Journal of Community & Applied Social Psychology, 31, 39–52. 10.1002/casp.2486 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Jason LA, Guerrero M, Salomon-Amend M, Stevens E, Light JN & Stoolmiller M (2021). Context matters: Home-level but not individual-level recovery social capital predict residents’ relapse. American Journal of Community Psychology, 67(3–4), 392–404. 10.1002/ajcp.12481 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Jason LA, Light JM, Stevens EB, & Beers K (2014). Dynamic social networks in recovery homes. American Journal of Community Psychology, 53, 324 –334. 10.1007/s10464-013-9610-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Jason LA, Lynch G, Bobak T, Light JM, & Doogan NJ (2022). Dynamic interdependence of advice seeking, loaning, and recovery characteristics in recovery homes. Journal of Human Behavior in the Social Environment, 32(5), 663–678. 10.1080/10911359.2021.1947930 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Jason LA, Olson BD, Ferrari JR, Majer JM, Alvarez J, & Stout J (2007). An examination of main and interactive effects of substance abuse recovery housing on multiple indicators of adjustment. Addiction, 102(7), 1114–1121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Jason LA, Stevens E, & Ram D (2015). Development of a three-factor psychological sense of community scale. Journal of Community Psychology, 43(8), 973–985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Jason LA, Wiedbusch E, Bobak T, & Taullahu D (2020). Estimating the number of substance use disorder recovery homes in the United States. Alcoholism Treatment Quarterly, 38(4), 506–514. 10.1080/07347324.2020.1760756 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kelly JF, & Hoeppner B (2015) A biaxial formulation of the recovery construct. Addiction Research & Theory, 23(1), 5–9, DOI: 10.3109/16066359.2014.930132 [DOI] [Google Scholar]
  33. Laudet AB, Savage R, & Mahmood D (2002). Pathways to long-term recovery: a preliminary investigation. Journal of Psychoactive Drugs, 34(3):305–11. doi: 10.1080/02791072.2002.10399968 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Laudet AB, & White W (2010). What are your priorities right now? Identifying service needs across recovery stages to inform service development. Journal of Substance Abuse Treatment, 38(1), 51–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Light JM, Jason LA, Stevens EB, Callahan S, & Stone A (2016). A mathematical framework for the complex system approach to group dynamics: The case of recovery house social integration. Group Dynamics: Theory, Research, and Practice, 20(1), 51–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Majer JM, Beasley C, Stecker E, Bobak TJ, Norris J, Nguyen HM, Ogata M, Siegel J, Isler B, Wiedbusch E, & Jason LA (2018). Oxford House residents’ attitudes toward medication assisted treatment use in fellow residents. Community Mental Health Journal, 54(5), 571–577. 10.1007/s10597-017-0218-4 [DOI] [PubMed] [Google Scholar]
  37. Moos RH (2008). Active ingredients of substance use-focused self-help groups. Addiction, 103(3), 387–396. [DOI] [PubMed] [Google Scholar]
  38. Polcin DL (2009). A model for sober housing during outpatient treatment. Journal of Psychoactive Drugs, 41(2), 153–161. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Ripley RM, Snijders TAB, Boda Z, Voros A, & Preciado P (2023). Manual for RSiena. Oxford, UK: Department of Statistics, Nuffield College. Available at: http://www.stats.ox.ac.uk/~snijders/siena/RSiena_Manual.pdf [Google Scholar]
  40. Rosenberg M (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press. [Google Scholar]
  41. Schweinberger M (2012). Statistical modelling of network panel data: Goodness of fit. British Journal of Statistical and Mathematical Psychology, 65, 263–281. 10.1111/j.2044-8317.2011.02022.x [DOI] [PubMed] [Google Scholar]
  42. Sklar SM, Annis HM, & Turner NE (1999). Group comparisons of coping self-efficacy between alcohol and cocaine abusers seeking treatment. Psychology of Addictive Behaviors, 13, 123–133. doi: 10.1037//0893-164X.13.2.123 [DOI] [Google Scholar]
  43. Sliedrecht W,Waart R, Witkiewitz K, & Roozen HG (2019). Alcohol use disorder relapse factors: A systematic review. Psychiatry Research, 278, 97–115. [DOI] [PubMed] [Google Scholar]
  44. Snyder CR, Sympson SC, Ybasco FC, Borders TF, Babyak MA, & Higgins RL (1996). Development and validation of the State Hope Scale. Journal of Personality and Social Psychology, 70(2), 321–335. [DOI] [PubMed] [Google Scholar]
  45. Snijders TAB (2022). R Script SelectionTables.r. Retrieved from https://www.stats.ox.ac.uk/~snijders/siena/SelectionTables.r
  46. Snijders TAB, van de Bunt GG, & Steglich CEG (2010). Introduction to stochastic actor based models for network dynamics. Social Networks, 32, 44 – 60. 10.1016/j.socnet.2009.02.004 [DOI] [Google Scholar]
  47. Stevens EB & Jason LA (2015). Evaluating Alcoholics Anonymous sponsor attributes using conjoint analysis. Addictive Behaviors, 51, 12–17. doi: 10.1016/j.addbeh.2015.06.043 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Vaillant GE (1983). The Natural History of Alcoholism, Harvard University Press, Cambridge, MA. [Google Scholar]
  49. Wasserman S, & Faust K (1994). Social Network Analysis: Methods and Applications. Cambridge, UK: Cambridge University Press. [Google Scholar]
  50. Whipple CR, Jason LA, & Robinson WL (2016). Housing and abstinence self-efficacy in formerly incarcerated individuals. Journal of Offender Rehabilitation, 55(8), 548–563. 10.1080/10509674.2016.1229713 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. World Health Organization Group (1998). The World Health Organization quality of life assessment (WHOQOL): development and general psychometric properties. Social Science & Medicine, 46(12), 1569–1585. [DOI] [PubMed] [Google Scholar]
  52. Zywiak WH, Longabaugh R, & Wirtz PW (2002). Decomposing the relationships between pretreatment social network characteristics and alcohol treatment outcome. Journal of Studies on Alcohol, 63(1), 114–121. [PubMed] [Google Scholar]

RESOURCES