Exploring support provision for recovery from substance use disorder among members of a sober active community

Abstract

Social support systems are often the most important factor in initiating and sustaining recovery from substance use disorders (SUDs). The Phoenix is a non-profit organization whose mission is to create communities and host events that harness the transformational power of social connection to promote SUD recovery. Through online surveys and in-depth interviews, this study assessed factors related to support provision within Phoenix members’ social networks. Online surveys measured participants’ demographic information, when they started attending Phoenix programming, and how frequently they attended The Phoenix. During interviews participants were asked to identify who supports them in their recovery. For each network member listed, the participant indicated their relationship to the person, the person’s gender, if that person was in recovery, if that person was a Phoenix member, and how often that person provided support to the respondent (Never to Always). Multilevel modeling explored factors related to more support provision across 723 support dyads reported from 79 participants. Participants (n = 79; 76% non-Hispanic white; 48% male; M^age = 38.27 years) reported an average of 9.15 members in their support networks (range 2–15). After controlling for network size, Phoenix members reported the most support provision from mentors (β = 0.356, p = 0.001), people in recovery (β = 0.451, p < 0.001), and fellow Phoenix members (β = 0.303, p = 0 .001). The longer someone had been a member of The Phoenix, the more likely they were to report greater support provision from their network members (β = 0.064, p = 0.03). This study makes two important contributions. First, while it is understood that social support broadly defined is important for recovery, this study provides specific characteristics of social networks that could yield greater social support provision. Second, because findings show that fellow Phoenix members provided participants more support, and that participants who had been involved in The Phoenix for longer experienced greater support, this study suggests The Phoenix could be an effective environment for creating the needed support systems for recovery from SUDs. More rigorous study designs could empirically test participation in The Phoenix as a laudable strategy for combatting relapse and supporting long-term recovery from SUD.

Introduction

In 2021, 1-in-6 Americans aged 12 and older (40.3 million people) met the criteria for a substance use disorder (SUD)¹. Substance use can have negative effects at individual (e.g., health issues including overdose, organ damage, mental health disorders, infectious diseases, and increased risk of accident);², interpersonal (e.g., domestic violence, child neglect or abuse, and strained relationships)^3,4, and community (e.g., diminished quality of life through increased crime rates, reduced property values, and limited economic development)^5,6 levels. According to the National Institute on Drug Abuse (NIDA) and the Substance Abuse and Mental Health Services Administration (SAMHSA), the estimated economic cost of SUDs and other harmful substance use in the United States in 2019 was around $740 billion annually¹. This estimate covers expenses related to healthcare, lost productivity, criminal justice costs, and social services.

Though professional treatment can help people abstain from or manage their SUD⁷, only 6% of those who need treatment successfully access care and a staggering 40–60% relapse within one year after treatment¹. Recovery is defined as a process of change through which individuals improve their health and wellness, live a self-directed life, and strive to reach their full potential^8,9. Traditionally, SUD recovery has been supported via professionally directed clinical services (e.g., medical detoxification, counseling, medications) and peer-led mutual-help organizations (MHOs; e.g., Alcoholics Anonymous, Narcotics Anonymous, SMART Recovery)¹⁰. Although these pathways remain critical, they do not fully address the diverse needs of individuals in recovery, as there is no one-size-fits-all model¹¹.

To bridge this gap, community-based recovery support services emerged, offering broader and more flexible recovery options. Notable among these are Recovery Community Centers (RCCs)—such as the Connecticut Community for Addiction Recovery (CCAR) and the Recovery Café Network—which provide peer-driven, holistic support tailored to individuals’ needs. Similarly, collegiate recovery programs (CRPs) offer dedicated recovery resources within academic environments, while digital recovery support services (D-RSS) have expanded access to recovery support, addressing barriers related to geography, cost, and social constraints¹². Unlike traditional pathways, many community-based resources are not tied to a single recovery approach but are inclusive of diverse recovery pathways, focusing on strengthening individuals’ recovery capital—the personal, social, and environmental resources that support long-term recovery¹³. To highlight the scale and diversity of these efforts, the Association of Recovery Community Organizations (ARCO) provides a global map of organizations (https://facesandvoicesofrecovery.org/programs/arco/members-on-the-map/), including The Phoenix, which we describe in greater detail below. These emerging pathways—including RCCs, CRPs, and D-RSS—reflect a shift toward holistic, flexible, and community-based recovery support models as they emphasize personalized and accessible approaches that align to lived experiences of those in or pursuing recovery.

In recognition of (1) the large number of people who pursue recovery outside of formal treatment and (2) the need for ongoing support following acute care, community-based recovery support services often take a holistic, peer-based, and long-term approach that is accessible to people in recovery throughout their lives^14,15. The origin of the social model recovery occurred in the 1930s among the Alcoholics Anonymous (AA) 12-step/12-traditions mutual help groups founded by Bill Wilson and Dr. Bob Smith aimed to fight alcoholism and then expanded to full neighborhood recovery homes in the 1970s believing that individuals in recovery can help others in recovery^16,17. Inspired by AA, other 12-step SUD mutual aid fellowships follow similar principles are established later and grew over time, including the Narcotics Anonymous (founded in 1953), the Pills Anonymous (1972), the Cocaine Anonymous (1982), the Marijuana Anonymous (1989), the Crystal Meth Anonymous (1994), and the Heroin Anonymous (2004) providing mutual support to encourage abstinence.

A recent systematic review found that community-based and non-clinical services were more promising approaches to address SUD compared to more formal, short-term, and clinical services¹⁵. For example, a longitudinal study found that the social networks within support groups such as AA programs in a northern California county served as encouraging mediators of decreasing alcohol consumption¹⁸. The effectiveness of social support as a mediator has also been demonstrated in another two-year study of the Double Trouble in Recovery group in New York City with the findings that longer participation in the cohort in the first year was associated with lower substance use in the second year, proving a straightforward association between social support and substance use^19,20. On the long-term side, recovery residences are also an encouraging mechanism to support individuals with addiction to both initiate and maintain long-term addiction recovery²¹. Studies have shown that community-based recovery centers like the Oxford House have successfully provided an alternative way for individuals seeking help from substance abuse with effective but inexpensive support networks²². However, despite an emerging body of literature supporting community-based recovery support and the growing popularity of such services, many community-based programs lack empirical evidence for their effectiveness²³.

The 2023 U.S. Surgeon General’s Advisory suggested that social connection influences overall well-being directly, while social isolation and loneliness increase the risk of premature death significantly, indicating that people will live longer when more socially connected but will face poor health and other negative outcomes when socially disconnected²⁴. Belonging to a supportive social network is among the strongest predictors of sustained remission from addiction, and is a likely reason why community-based recovery support is an effective way to address SUD^25,26. Research consistently suggests people recovering from SUDs who have strong social support systems are less likely to re-initiate and/or sustain substance use, and are more likely to report positive physical and mental health, helping to foster lifelong recovery from addiction^14,27. Among a sample of 500 + people recovering from SUDs, shifting from social isolation to social connectedness was a principal factor associated with their transition from addiction to recovery and positive changes in the composition of their social networks²⁸. Thus, strengthening connectedness to others, including being a part of a supportive community of peers and advocates of recovery, is critical for people with SUDs.

Studies have shown how one’s social environment and connections can influence recovery. Recovery from SUD can be seen as a socially mediated transition resulting in changes in social environments and social circles²⁸. Mutual-help organizations and community organizations can aid in fostering positive social environments and facilitating positive social network changes, as well as significantly reducing the chance of relapse^29,30. Through consistent engagement in mutual help organizations, an individual’s social circle can shift from problematic social connections (i.e., social connections inhibit recovery) to supportive social connections (i.e., social connections that foster recovery)^30,31. The shift in social connections for someone in recovery can lead to a total rebuilding of their social networks through the emergence of their “recovery identity”²⁸. Having a strong support network while in recovery can contribute to forming a long-term identity change, thus supporting sustained recovery and abstinence^32,33, and belonging to supportive communities is an ideal way to foster the connections and social networks needed for these outcomes. While research consistently confirms the important role of social support in recovery from SUDs, specific characteristics of social ties that yield the most support are undetermined. To foster the right kinds of social support, it is important to better understand what characteristics of various social connections foster recovery support for people overcoming SUDs.

The Phoenix

This study was completed in partnership with The Phoenix, a non-profit organization and community-based recovery support service. The mission of The Phoenix is “to build a sober active community that fuels resilience and harnesses the transformational power of [social] connection…”³⁴, p. 3]. Since its launch in 2006, The Phoenix has served over 556,950 people and operates programs in 280 counties across all 50 states, with a growing number of virtual programs available through its digital platform.

The Phoenix offers a diverse range of free events, including group fitness classes such as CrossFit, yoga, kickboxing, and boxing; outdoor activities like hiking, golf, and paddleboarding; and creative and cultural programs, including recovery meetings, poetry slams, book clubs, concerts, and arts and crafts. Members can attend as many events as they choose with no restrictions on the type or number of activities. Participation is cost-free and open to anyone with at least 48 h of continuous sobriety, including individuals at any stage of recovery, as well as allies, family members, supporters, and people living sober lifestyles.

The range of events is driven by the interests and creativity of Phoenix members and volunteers, who play a central role in shaping and delivering programming in a variety of formats. Many events are hosted in partnership with local organizations—such as CrossFit gyms, yoga studios, and public parks—The Phoenix also operates physical “brick-and-mortar” locations. At the time of this study, these were located in Denver, CO, Wichita, KS, and Boston MA. These facilities include multipurpose fitness spaces (e.g., weightlifting, cardiovascular equipment, boxing, and yoga), community meeting areas, kitchens, locker rooms, and office space. Designed to encourage social interactions, these locations provide opportunities for connection between events, although they are not open 24 h a day.

The Phoenix has experienced rapid growth, driven in large part by its volunteer network. By the end of 2024, volunteers, supported by regional managers, led over 65% of all Phoenix events. To further expand its reach, The Phoenix provides on-demand digital content—such as meditations, workouts, and yoga—through its app, ensuring flexibility and accessibility for individuals with diverse needs. Growth was also spurred in 2023 when The Phoenix launched a first-of-its-kind online “marketplace” for recovery resources, connecting members to other recovery supports, including CCAR, SMART Recovery, Ben’s Friends and more, through its digital platform.

The Phoenix supports multiple recovery pathways, complementing existing options such as clinical services, MHOs and RCCs, while also standing alone as a distinct recovery option for some. Uniquely, it is not time-bound, allowing individuals to engage as and when it suits their recovery, or life, journey. Many Phoenix members also participate in mutual help groups and utilize additional recovery resources. The Phoenix’s programs function as a form of ongoing support that complements traditional treatment methods, and as such, offers a flexible and inclusive approach to recovery. While programs like The Phoenix are anecdotally known to facilitate recovery through community and social connectedness, to our knowledge these relationships have not been empirically evaluated. This study aims to address that gap.

Study purpose

The purpose of this study is to assess specific factors related to greater support provision within Phoenix members’ social networks. We specifically examined what characteristics are associated with more supportive social ties within participants’ networks. This study fills two notable gaps in the literature. First, it explores social support experienced through a sober active community, potentially providing evidence for the effectiveness of these types of pathways to recovery on yielding critically needed social support. Second, using Social Network Analysis (SNA), this study examines specific factors and mechanisms related to more supportive social connections, unveiling a more nuanced understanding of what kinds of relationships might result in more recovery support for someone overcoming a SUD.

Methods

Participants and procedure

The research team collaborated with staff members of The Phoenix to recruit participants through their mobile app who were: (1) Phoenix members (verified through a membership database), (2) currently in recovery from a SUD, (3) participated in a group-based exercise program (e.g., CrossFit, yoga) offered through The Phoenix, and (4) were 18 years of age or older. People become a member of The Phoenix by downloading and creating an account on the mobile app. They can interact with others in the community as well as sign up for online and in-person events. Alternatively, they can sign-up for events in-person where they provide their information to create a personal account. Because physical activity and exercise are often used as adjuncts to addiction therapy due to their positive impacts on physical and mental health³⁵ and because The Phoenix exists as a sober active community, where physical activity is a key component, we focused on members involved in group-based activities offered through The Phoenix.

Study procedures began with The Phoenix posting in their mobile app about the study, which included a link to learn more and contact information for the research team. After clicking the link for more information about the study, which included details about the study’s purpose, protocols, and any risks and benefits for participating, potential participants provided their electronic informed consent (i.e., they had an option to select “I agree to participate in this study” or “I do not agree to participate in this study”) and were led to an online Qualtrics survey measuring demographic information, history with substance use and recovery, and duration and frequency of participation with The Phoenix. Upon completion of the online survey, participants were given the opportunity to schedule an hour-long interview with a member of the research team using online appointment scheduling software. All interviews were conducted via teleconferencing software. We used Network Canvas software³⁶ to collect rich social network data from each participant. 137 Phoenix members completed the online survey, and of those, 79 completed follow-up interviews. Participants were sent a $40 gift card after completing surveys and interviews.

Measures

Demographic information was collected from participants in the online survey. Participants reported their age (in years), gender (man, woman, transgender, non-binary/non-conforming, prefer not to answer, other), race (select all that apply: White or Caucasian, Black or African American, American Indian/Native American or Alaska Native, Asian, other), and ethnicity (Are you Spanish, Hispanic, or Latino origin, yes or no? ).

Phoenix program duration and frequency

Respondents indicated when they began participating in group exercise programming with The Phoenix, with answer options less than 6 months ago, 6 months-1 year ago, 1–2 years ago, 2–3 years ago, 3–4 years ago, 4–5 years ago, and more than 5 years ago. They also indicated how frequently they attend group exercise programming, with answer choices less than once per week, 1–2 times per week, 3–4 times per week, and 5 or more times per week.

Egocentric network variables

Egocentric network data were collected via name generator, name interpreter, and edge interpreter questions³⁷. Name generators require the respondent (from here on referred to as the “ego”) to provide a list of people they are connected to in some way. In this study, we asked egos three name generators, including: (1) “list up to five people you consider most supportive of you in recovery right now,” (2) “list up to five people you feel close to, but currently make recovery more difficult in some way,” and (3) “list up to five people who may not make recovery harder or easier for you, but are still important to you in your life right now.” As a result, egos nominated 1–15 members of their egocentric networks, henceforth referred to as “alters.”

Name interpreters require the ego to provide information about each alter they nominated in the name generator questions. Egos answered name interpreter questions related to the following: (a) their relationship to each alter (e.g., parent, friend, significant other, mentor); (b) each alter’s gender (man, woman, transgender, non-binary/non-conforming, other); if, to the ego’s knowledge, the alter was (c) in recovery and if they (d) participate in The Phoenix (“Yes,” “No,” or “I don’t know”); and (e) to what extent each alter supports the ego in their recovery (“Never”, “Sometimes”, “Often”, and “Always”).

Edge interpreter questions, also known as “Alter-Alter Ties,” assess the presence of social connections between an ego’s alters. For each set of alters (i.e., dyad) nominated in the name generator question, we asked, “are these people likely to interact with one another if you were not present?” Egos could say “yes” or “no” for each dyad.

Data collected from name generators, name interpreters, and edge interpreters result in variables measured at two analytic levels. Level 1 variables constitute characteristics of the alters and ties (e.g., alter’s relationship to ego, support provision from alter to ego). Level 2 variables constitute ego characteristics (e.g., an ego’s age), as well as aggregated characteristics of the ego’s network (e.g., proportion of the ego’s network that is in recovery).

Level 1 variables: alters and ties

Characteristics of alters and ties included each alter’s: gender (0 = man, 1 = woman; there were no other genders assigned to alters in these data); recovery status (0 = not in recovery, 1 = in recovery); Phoenix participation (0 = not a Phoenix participant, 1 = participates in The Phoenix); relationship to the ego (1 = friend, 2 = kin, 3 = significant other, 4 = mentor); and recovery support provision (0 = Never supports ego in their recovery, 1 = Sometimes supports ego in their recovery, 2 = Often supports ego in their recovery, and 3 = Always supports ego in their recovery).

Level 2 variables: egos and networks

Gender, race and ethnicity, age, duration of Phoenix participation, and frequency of Phoenix participation were measured at ego-level. Aggregate network characteristics included the size of ego’s network (i.e., number of alters in ego’s network), the proportion of ego’s network that was in recovery, and the network density (i.e., the proportion of ties present between alters compared to the ties possible; scores of 0 indicate no ties between alters, scores of 1 indicate all possible ties between alters exist. Level 2 network variables were computed using the egor package in R³⁸.

Analytic strategy

To identify ego, alter, and network-level factors related to recovery support provision from alters, we conducted a multilevel model using the multilevel package³⁹ within R programming language and software. Multilevel modeling is an ideal analytic strategy when conducting egocentric network analyses due to its ability to account for the variance between and within ego networks^37,40. Based on intraclass correlation coefficients and likelihood ratio tests, we computed a random-coefficient multilevel model. Random-coefficient models use Level-1 alters nested in Level-2 egos, and account for dependence by including a random intercept for each ego, as well as a unique slope for each ego based on an alter-level variable. In this case, we used a random-coefficient model to predict support provision from alters and adjusted slope based on whether an alter was in recovery.

Independent variables in the random coefficient model included: (1) ego’s age, gender, duration of Phoenix participation, and frequency of Phoenix participation; (2) alters’ relationship to ego, gender, recovery status, and participation with The Phoenix; and (3) network size, proportion of the network that was in recovery, proportion of the network that were Phoenix participants, and network density.

Results

Descriptive statistics

Almost half (48.1%, n = 38) of egos were men, most identified as non-Hispanic White (75.9%, n = 60), and the average age was 38.27 years (SD = 8.76). Approximately one third (29.1%, n = 23) of egos had only been participating at The Phoenix for 6 months or less, and half (49.4%, n = 39) reported attending group exercise with The Phoenix 1–2 times per week. Egocentric networks ranged in size from 2 alters to 15 alters, with an average of 9.15 alters per network (SD = 2.18). On average, egos reported that 38.8% of alters in their network were also in recovery (SD = 18.9%). However, 10.1% (n = 8) of egos reported no alters in their networks who were in recovery, and one ego reported all of their alters were in recovery. The mean density score of egocentric networks was 0.45 (SD = 0.23), which suggests on average, just under half of possible ties existed within ego’s network. See Table 1 for a summary of descriptive statistics on egos.

Table 1.

Ego sample characteristics table (n = 79).

Ego-level variable name	n	%	M	SD
Age			38.27	8.76
Gender
Female	36	45.6
Male	38	48.1
Non-Binary	5	6.3
Race
White (Non-Hispanic)	60	75.9
Hispanic or Latinx	11	13.9
Black or African American	2	2.6
Other	3	3.8
Multiracial	3	3.8
Duration of Phoenix attendance
<6 months ago	23	29.1
6 months – 1 year ago	18	22.9
1–2 years ago	7	8.9
2–3 years ago	7	8.9
3–4 years ago	10	12.6
4–5 years ago	7	8.9
5 + years ago
Frequency of Phoenix attendance
<1×/week	21	26.6
1–2x/week	29	36.7
3-4x/week	17	21.5
5 + x/week	2	15.2
Network size			9.15	2.38
Network density			0.4478	0.23

Egos nominated 723 total alters, with 35.3% (n = 255) of alters being in recovery, and 20.1% (n = 145) being Phoenix participants. Over half (52.8%, n = 382) of alters were women, 43.8% (n = 317) were identified as friends, 37.2% (n = 269) were family members, and 11.9% (n = 77) were mentors, sponsors, or coaches to the ego. See Table 2 for a summary of descriptive statistics on alters.

Table 2.

Alter sample characteristics table (n = 723).

Alter-Level Variable Name	n	%
Gender
Female	382	52.8
Male	306	42.3
Non-Binary	6	0.8
Prefer not to answer/unsure	29	4.1
Race
White (Non-Hispanic)	521	72
Hispanic or Latinx	94	13
Black or African American	38	5.3
Asian or Pacific Islander	9	1.2
Multiracial	5	0.8
Prefer not to answer or unsure	56	7.7
Alter relationship to ego
Parent	95	13.1
Spouse or Partner	50	6.9
Sibling	82	11.3
Friend	317	43.8
Roommate	9	1.2
Coworker or Classmate	25	3.5
Mentor	61	8.4
Extended Family	47	6.5
Child	37	5.1
Alter in recovery
Yes	255	35.3
No or I don’t know	468	64.7
Alter Phoenix Participant
Yes	151	20.9
No	572	79.1

Multilevel model

Multilevel regression analysis assessed recovery support provision from alter to ego, examining what factors might explain greater support through network ties. The random coefficient model (see Table 3) showed egos who had a longer duration participating in The Phoenix (b = 0.07, p = 0.04) reported greater recovery support provision from alters. Alters who were mentors (b = 0.36, < 0.001), were in recovery (b = 0.46, p < .001), and who participated in The Phoenix (b = 0.29, p < 0.001) provided ego more recovery support. None of the network-level variables significantly explained support provision from alters.

Table 3.

Random coefficient multilevel model assessing recovery support provision from alter to ego.

Predictors	β	t	p
Ego’s age	0.000	0.00	0.997
Ego’s race (ref: white)	− 0.124	− 0.607	0.546
Ego’s gender (ref: female)	0.015	0.128	0.898
Ego’s duration of Phoenix attendance	0.064	2.249	0.028
Ego’s frequency of Phoenix attendance	0.120	1.544	0.128
Alter gender: female	0.062	0.942	0.347
Alter relationship: kin	0.081	1.088	0.277
Alter relationship: mentor	0.356	3.237	0.001
Alter in recovery	0.451	4.181	0.001
Alter Phoenix participant	0.303	3.291	0.001
Network size	0.023	0.859	0.394
Network density	0.171	0.609	0.545
Proportion of network members also in recovery	− 0.24	− 0.741	0.461

Discussion

The purpose of this study was to examine recovery support provision within the social networks of people participating in The Phoenix, a non-profit sober active community aimed at promoting recovery and wellbeing for people with SUDs. Findings from this study suggest that participation in a sober active community like The Phoenix might help someone experience more recovery support from their social networks, which could lead to longer, more successful recovery from SUDs.

Social network analysis revealed that participation in The Phoenix could yield social support needed for SUD recovery. Specifically, we found that the longer a participant had been a member of The Phoenix, the more support they received from their alters. Previous research points to the importance of social support and social networks in the recovery process^27,30,33 and how community-based recovery support services create opportunities for people to find the support they need^7,29. Research indicates that participating in meaningful activities and engaging with prosocial groups that support an individual’s recovery efforts promotes long-term recovery^41,42. As such, it was unsurprising that those who had a longer tenure with The Phoenix were also those who reported the most support provision from their network members. Additionally, research shows that when recovering from an addiction, some people have to recreate much of their social environment and social networks in support of their recovery and sobriety^25,28. Spending more time as a member of The Phoenix, and simultaneously spending more time in recovery, could translate to someone updating their social networks to be more supportive over time.

Beyond Phoenix membership being related to recovery support provision, our study also found that it was fellow Phoenix members in participants’ networks who were among the greatest support providers, offering more evidence that the social connections created within community-based recovery support services yield the recovery capital (i.e., the breadth and depth of resources that can be drawn upon to initiate and sustain recovery from SUD)⁴³ and support needed for recovery and wellbeing^41,44,45. In line with The Phoenix’s mission to create a supportive and safe community for people overcoming SUDs³⁴, this finding provides evidence that the social connections created within The Phoenix are likely to result in recovery support and increased recovery capital, and validates continued efforts to foster community among Phoenix members.

In addition to engagement with The Phoenix and its members, our study found that mentors and other people in recovery provided greater social support to participants. Specifically, anyone an ego labeled as a coach, sponsor, or mentor was more likely to provide greater recovery support. In this case, mentors, coaches, and/or sponsors represent a more experienced person who provides guidance, support, and encouragement for the ego. Given that research consistently shows mentors, sponsors, and peers can positively impact individuals recovering from SUD – particularly through peer recovery support and sponsorship⁴⁶ – we are not surprised by this finding. In their literature review, Reif and colleagues (2014) found that peer recovery support can lead to reduced relapse rates, increased treatment retention, improved relationships with providers and social supports, and increased satisfaction with treatment experience⁴⁶. This aligns with other research that shows individuals who experience successful long-term recovery often report being actively engaged in fellowship that includes working regularly with a sponsor/mentor^47,48. Moreover, since sponsors, mentors, and coaches are often individuals in recovery themselves, this could explain why others in recovery were among the most significant sources of support within participants’ social networks. Extant research shows that connecting with others in recovery fosters support, understanding, comradery, and accountability^41,49,50, and that having support from various social networks has positive effects on wellbeing^26,51. As such, it could be that connecting with a variety of people in recovery, including a sponsor, members involved in the sober active community, and others, could yield more support within networks.

Strengths and limitations

A primary strength of this study is the robust use of egocentric network analysis to examine support provision with participants’ social networks. Using Network Canvas and conducting in-depth interviews allowed for a rich collection of network data, and a deeper understanding of the types of relationships that yield greater support provision for people with SUDs. This study addresses important gaps in the literature including specific types of social relationships that translate to greater support and adds empirical evidence for community-based recovery support as a laudable strategy for facilitating recovery from addiction. However, conclusions drawn from these data are limited by the cross-sectional study design, convenience sampling technique, and smaller sample size. While statistical analyses were adequately powered because they were conducted at the dyadic level (n = 723), only 79 people are represented in this sample, all of which are members of The Phoenix. Generalizing these findings to people with SUDs outside of The Phoenix are cautioned, and more research is needed to determine if similar network dynamics yield recovery support for people in other contexts, including other recovery community organizations. Moreover, findings can only be interpreted as correlational due to the cross-sectional design. It could be that people who self-select into programs like The Phoenix already have the support they need. A longitudinal design could better determine if supportive social connections are created overtime through The Phoenix. That said, this study was an important first step in establishing evidence for The Phoenix’s effectiveness in providing a social environment conducive to recovery.

Implications for future research and practice

This study adds to a growing body of literature in support of sober active communities to aid in the addiction recovery process. Our findings suggest the possibility that longer involvement in a sober active community results in greater recovery support provision for participants recovering from SUDs, and that connections with fellow community members are associated with greater support. Given cost, disparity in access, and meager success rates of formal treatment, community approaches like The Phoenix are needed as an additional and complementary pathway to recovery. Results support sober active communities and programs like The Phoenix foster connections between members that are likely to provide the support needed for long-term recovery. Additionally, this study supports the continued use of social network analysis to understand social dynamics related to optimal recovery strategies. Longitudinal network studies could better describe the types of relationships needed for people at various stages of their recovery, and how to optimize social networks to prevent relapse and support wellbeing. Sociometric studies could also reveal network positions specifically within sober active communities that leads to greater recovery outcomes, both for individuals and network-wide. Additionally, future studies could measure various types of support such as emotional support (e.g., “Who do you go to for personal matters?”), instrumental support (e.g., “Who provides you with tangible help such as giving you a ride?”), or even informational support (e.g., “Who do you turn to for useful information to solve a problem you may have?”), as well as any negative support (i.e., anyone who may make life more difficult or hinders recovery progress) to have a more comprehensive understanding of how social support operates within these networks.

Conclusion

Social networks are critical for addiction recovery, and this study provides preliminary evidence for sober active communities and The Phoenix as a means for building social networks that yield the support that helps to overcome addiction. This study contributes added evidence for the importance of community-based recovery support services in fostering meaningful social connections between people in recovery, aligns with previous research that suggests the importance of a person in recovery connecting with a sponsor/mentor as well as with others in recovery, and substantiates future research using social network analysis to assess the social support mechanisms needed for addiction recovery across the lifespan.

Author contributions

M.P. led study conception, data collection, analysis, interpretation of results, writing, and manuscript submission; A.F. assisted with data collection and writing; S.P. assisted with data collection and critical revisions to the manuscript; S.L. assisted with data collection and critical revisions to the manuscript; Z.K. assisted with writing and critical revisions to the manuscript; K.H. assisted with study conception, participant recruitment, stakeholder partnerships, and critical revisions to the manuscript; T.P. assisted with study conception, interpretation of results, and critical revisions to the manuscript.

Funding

This study was financially supported through the Texas A&M Health Science Center Seedling Grant program.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

All study procedures were approved by the Texas A&M University IRB (#IRB2023-0202 M) in accordance with the standards detailed in the Declaration of Helsinki, and each participant provided their digital informed consent prior to data collection.