Evaluating the Impact of a Peer Support Program on Participants’ Well-Being

Abstract

Background:

Loss of belongingness may be particularly pronounced for women veterans, representing a threat to long-term well-being. Improvements in social support through engagement in a structured peer support program may mitigate the negative effects of loss of belongingness on well-being.

Objective:

We assessed the impact of participation in a peer-led, structured, social support group-based network on outcomes related to well-being [i.e., belongingness, social support, quality of life, posttraumatic stress disorder (PTSD), depression]. Subgroup analyses examined relative impact among those who completed the intervention and those reporting clinical levels of PTSD and depression symptoms.

Methods:

We analyzed survey data consisting of reliable and valid measures collected at baseline, postgroup and 3-month follow-up among 393 participants in the Woven Veterans Network’s (WoVeN) group program.

Results:

We observed improvements in posttraumatic stress disorder (PTSD) symptoms over time. We observed additional benefits among those who received an adequate dose of the intervention (significant improvements on PTSD, belongingness) and those with clinical levels of mental health symptoms (significant improvements on PTSD, depression, belongingness, quality of life).

Conclusions:

Impacts on social support may have been masked due to ceiling effects given wide dispersion baseline social support in this sample. This social support network had particularly profound impacts on well-being for those veterans who suffered from conditions for which isolation and loneliness are particularly salient.

There is a growing recognition of the need to increase the focus of healthcare from more narrowly defined, symptom-driven outcomes to greater consideration of the well-being of the whole person.^1,2 This study sought to improve the well-being of a national sample of women veterans, a population at risk for poor well-being due to isolation and lack of belongingness after service.³ Specifically, we assessed the impact of participation in a peer-led, structured, social support group-based network (Women Veterans Network; WoVeN) on outcomes (belongingness, social support, quality of life) related to well-being. We further sought to understand the relative impact of this program on those women within the WoVeN program who are suffering from clinical levels of posttraumatic stress and/or depression symptoms as the presence of these mental health conditions is likely to increase the risk for poor well-being.³

Baumeister and Leary,⁴ in their belongingness hypothesis, posit that humans are driven to form and maintain meaningful and lasting interpersonal relationships. Belongingness is defined as “the experience of personal involvement in a system or environment so that persons feel themselves to be an integral part of that system or environment” (p. 173).⁵ The type of belongingness that may be most related to psychological distress is social belongingness.^6,7 The formation of relational bonds requires that one has repeated and frequent contact with several individuals within a positive, supportive framework. The impact of lack of belongingness on mental health and functioning is well-established and consequences of lack of belongingness include loneliness, increased risk for depressive symptomatology, social anxiety, isolation, and even suicidality.^8,9 The effectiveness of intervening in belongingness is less well understood.

Loss of belongingness may be particularly pronounced for the military population.¹⁰ The novel environment of military service exists within the larger context of separation from loved ones and former communities during servicemembers’ key periods of development.¹⁰ The development of unit cohesion throughout service is not only critical to the function of the military but has long been recognized as related to servicemembers’ well-being, mental health,^10,11 and sense of belongingness.¹² Indeed, relationships forged during military service are commonly described as some of the closest relationships across the lifespan.^13,14 Separation from service disrupts belongingness at multiple levels including the loss of close friendships and membership in a cohesive community. Loss of belongingness can be compounded by challenges inherent in re-integrating into civilian life including loss of military identity and sense of purpose and the loss of stability and structure afforded by the military. The presence of mental health conditions such as posttraumatic stress and depression secondary to exposures to trauma during service can further amplify isolation and lack of belongingness.¹⁵

A recent study has shown that women veterans may be at greater risk for poor mental health and declines in aspects of their well-being after separation from service including relationship functioning.¹⁶ Challenges in social functioning specifically may be amplified for women veterans given their low base rates in communities across the nation, lack of recognition for women’s military service by the general population, and barriers for women to benefit from existing veteran resources which are typically geared toward men.¹⁷ As a result, after separating from service, women veterans report difficulty establishing connections with the veteran population as well as with civilians.¹⁸

Cacioppo et al.¹⁹ posit that life circumstances can lead individuals to discover that their social and community connections are limited and that they are quite isolated. For individuals who already perceive themselves to be without community and bereft of social connection, cognitive processes, such as attentional, confirmatory, and memory biases, may contribute to withdrawal from future attempts at social connections leading to further isolation. This cycle might be particularly applicable for women veterans who find themselves on the periphery of both the veteran and civilian communities after separation from service and amplified for those women veterans suffering from PTSD and depression.¹⁹ The presence of these mental health conditions may independently impede the ability to foster social connections, further increasing risk for isolation and loneliness.¹¹

Social support has the demonstrable ability to mitigate the negative effects of loss of belongingness.¹¹ Interventions designed to mitigate the effects of loss of belonging and increase social support and overall quality of life are gaining increasing attention.¹⁹ Peer-facilitated interventions have the benefit of providing social support and connections/belongingness necessary for success in most aspects of functioning and are positively associated with mental²⁰ and physical health²¹ outcomes. Peer-led support groups tend to focus on a shared condition or problem or experience (eg, weight loss, reproductive health, grief) and can occur in isolation or in combination with clinical care.²² These groups have been steadily growing in number, likely due to their inherent key advantages including ready availability and accessibility in most communities, low or no cost to participants, and explicit focus on common goals including mental health difficulties (eg, drug and alcohol use programs such as Alcoholics or Narcotics Anonymous, severe mental illness and mood disorders).²² Although a growing literature, there are few methodologically rigorous studies (mostly conducted with civilian samples) evaluating the effectiveness of peer support programs perhaps due to methodological challenges such as variability in design, self-selection bias, difficulty tracking attendance in groups relying on volunteer staff and difficulty tracking receipt of additional care and resources by participants.^23–26 Despite these methodological challenges, gains in enhanced social support after participation in support groups are typically observed, with some exceptions, whereas significant gains in targeted mental health outcomes in these programs are less consistent. Peer-facilitated social support interventions have been developed across a variety of venues for veterans with different areas of need.²² The few studies assessing the efficacy of these programs within the veteran population have largely been conducted with clinical samples primarily comprised male veterans.²⁷ Despite the demonstrable loss of belongingness and elevated risk for associated negative outcomes for women veterans specifically, before the development of WoVeN, there had been no peer support network specifically designed to (1) meet the unique needs of women veterans, (2) encompass the range of topics identified by women veterans as central to improving their well-being, (3) provide content developed by national experts, (4) provide a structured curriculum led by trained peer leaders, and (5) provide a sustainable social support network at both the local and national levels.

CURRENT STUDY

We conducted an open trial designed to evaluate the impact of participation in WoVeN on constructs of interest. We hypothesized that participation in WoVeN would result in significant improvements in veterans’ sense of belongingness, quality of life, and perceptions of social support from friends. Although WoVeN is not a mental health treatment designed to target diagnosable conditions, we were particularly interested in understanding the impact of the program specifically among veterans who were suffering from clinically significant posttraumatic stress or depression symptoms given the potential for these symptoms to impair well-being. We hypothesized that those who suffered from clinical levels of distress would benefit more on primary outcomes from participation in the program as their needs may be greater as well as experience decreases in mental health symptoms. Finally, because posttraumatic stress and depression symptoms can disrupt one’s ability to form relationships, we evaluated the extent to which those with clinical levels of distress differed from those without on group cohesion.

METHODS

Participants

Of 1869 women enrolled in a WoVeN group and invited to participate in research, 662 completed at least the baseline (pregroup) survey representing a 35.42% participation rate. Of those participants, 393 completed at least 1 follow-up survey and were included in the final sample. Our sensitivity analysis demonstrated that those who completed follow-up surveys did not differ from those who did not (N=269) on demographic or military service characteristics.

Participants’ demographic and military characteristics are presented in Table 1 in the full sample and stratified by probable posttraumatic stress disorder (PTSD) and/or diagnosable depression status. On average, participants were 41 years old with 86.8% identifying as non-Hispanic and 50.1% identifying as White. Participants with probable PTSD and/or depression were slightly younger, overrepresented in lower SES categories, less likely to be working full time, and more likely to identify as disabled. Participants served across military branches, with the majority serving in the Army (44.5%).

TABLE 1.

Demographics Stratified by PTSD/Depression Status

	Total sample (N=393),* % or M (SD)	Clinical group (n=205),† ‡ % or M (SD)	Nonclinical group (n=93),† ‡ % or M (SD)
Age	48.1 (10.4)	46.9 (10.4)	50.0 (9.7)
Race and ethnicity
Black	33.6	30.2	29.0
White	50.1	52.5	54.8
Multiracial/other	16.3	17.3	16.1
Hispanic	13.2	12.3	14.0
Annual income
Less than $35,000	18.2	20.1	9.7
$35,000–$54,999	23.1	25.5	18.3
$55,000–$74,999	19.2	20.1	19.4
$75,000–$99,999	14.4	11.3	20.4
$100,000 or more	25.1	23.0	32.3
Relationship status
Single	22.3	24.0	15.1
Married/in a relationship	54.9	53.4	66.7
Separated/divorced/widowed	22.8	22.5	18.3
Work status
Full time	36.4	27.7	46.4
Part-time	4.3	4.3	3.6
Looking for paid work	3.4	3.3	2.4
Disabled	34.7	46.7	19.0
Other (caretaker/volunteer/student)	21.2	17.9	28.6
Primary military occupation
Combat arms	3.2	4.0	2.2
Combat support	30.5	37.5	26.4
Service support	66.3	58.5	71.4
Deployed overseas	62.1	63.3	67.0
Military branch
Army	44.5	48.0	42.4
Marine Corps	7.6	5.9	9.8
Navy	24.5	24.8	22.8
Air Force	22.7	20.3	23.9
Coast Guard	0.8	1.0	1.1

^*Observed sample completed baseline and at least 1 follow-up (post, 1 month, and/or 3 months).

^†N=95 were missing data for probable PTSD and/or depression at baseline.

^‡The clinical group consisted of participants with probable PTSD and/or depression, whereas the nonclinical group are those who do not meet clinical cutoffs.

PTSD indicates post-traumatic stress disorder.

Measures

To assess PTSD, we used the PTSD Checklist for DSM-5,²⁸ which is a 20-item questionnaire corresponding to the DSM-5 symptom criteria²⁹ for PTSD. We used the recommended clinical cutoff of 32 to distinguish those with probable PTSD from those without.³⁰ Cronbach’s alpha in the current sample was 0.96. The Patient Health Questionnaire-8³¹ assessed depression, with the recommended cutoff of 10 used to distinguish those with likely major depressive disorder from those without. Cronbach’s alpha in the current sample was 0.92. We used the Multidimensional Scale of Perceived Social Support³² 4-item friends subscale to assess perceived social support from friends. In the current sample, Cronbach’s alphas for the friends subscale was 0.95. The Quality of Life Inventory³³ assessed perceptions of quality of life via responses to the importance they attach to each of 16 life domains and their current satisfaction with each domain. Cronbach’s alpha in the current sample was 0.85. The General Belongingness Scale,³⁴ a reliable and valid 12-item measure, assessed participants’ general sense of belonging. Sample items include “I have a place at the table with others” and “I feel isolated from the rest of the world” (reversed). Cronbach’s alpha in the current sample was 0.95. The Group Cohesion Questionnaire,³⁵ a 25-item measure assessing group members’ sense of cohesion with the overall group and other members, assessed group cohesion. Cronbach’s alpha in the current sample was 0.94.

The WoVeN Peer Support Program

WoVeN was designed to target the specific and unique needs of women veterans. All aspects of the WoVeN program were informed by evidenced-based practices including content development, group size (6–8 members to facilitate intimacy and connection), and the training of peer leaders, which is manualized to maintain consistency across peer leader cohorts. WoVeN staff provides ongoing support to Peer Leaders through monthly consultation meetings.

The WoVeN structured group curriculum is manualized and consists of 8, 90-minute sessions beginning with an introductory meeting followed by 6 themed meetings (transitions, balance, stress relief, trust, connections, and esteem), and a final celebration. Each meeting includes a one-page activity that is designed to help participants collect their thoughts about their life experiences related to that session’s theme. The peer leaders are then trained to rely on the manual to generate discussion (using the provided open-ended questions as a guide and Socratic Questions as the technique) and to help the group members connect, share their experiences (military and otherwise), and provide support to one another. WoVeN groups are offered in person or virtually via web-based meeting platforms. Each meeting begins with a brief greeting followed by a peer leader-guided activity that is central to the theme and designed to provide the opportunity to reflect individually on life experiences. The ensuing group discussion is led by the peer leaders and facilitated by questions provided in the manual. Discussion questions are designed to help elicit cognitions that are central to well-being including current and historical barriers to developing relationships, challenges and successes across life events and experiences, and strategies for fostering change in positive directions. Military-related topics (eg, isolation after separation from service, lack of public awareness of women’s service) are included in the manual, but discussion questions do not center solely around participants’ military identity to increase social connections across the range of identities that participants may embrace.

Procedures

All women who enrolled in a WoVeN support group during the study period were invited to participate in the voluntary research (see Galovski et al., 2022 for a description of enrollment procedures). Research participants were invited to complete assessments at 3 time points: baseline, postgroup (up to 6 weeks after group completion), and three-month follow-up. Assessments were completed online through Qualtrics Research Suite (a secure, online platform used for collecting self-report data from research participants) or, if requested, paper and pencil survey. Participants received a $10 gift card for each survey completed. All human subjects procedures were approved by the Boston University Institutional Review Board.

Analyses

The analytic sample consisted of participants who completed a baseline survey and at least 1 follow-up survey. Participants’ demographic and military characteristics were described through descriptive statistics, examined both for the overall sample, and the clinical subsample defined by recommended cutoffs on measures of PTSD and/or depression.

Descriptive statistics on outcome variables were examined at all 3 time points. Linear mixed modeling was used to assess changes in outcomes over the 3 time points, with a compound symmetry covariance structure to account for the longitudinal observations. The models were run 3 ways: (1) main effect of time in the overall sample; (2) adding clinical group status and an interaction term between clinical group status and time; and (3) adding intervention completion status and an interaction term between intervention completion and time. Models 2 and 3 were conducted to assess for moderation of changes over time by the 2 group variables separately. For the interaction models, slopes for time in each group were calculated, along with the Time × Group slope, which indicates the change in slope between the two groups. Within-subject effect sizes for baseline to 3-month changes were calculated, and between-groups effect sizes for the interaction models were assessed using Cohen’s d_z. All analyses were conducted using SAS/STAT software, version 9.4 of the SAS System for Microsoft Windows (SAS Institute, Inc.).

RESULTS

Descriptive statistics for all outcome variables can be found in Table 2. We examined change over time for the full sample and saw no significant improvement on outcomes of interest with the exception of PTSD (β=−1.36, t(435)=−2.89, P<0.01; d=0.19). Depression and belongingness showed a trend (depression: β=−0.28, t(485)=−1.85, P=0.07; d=0.10; belongingness: β=0.56, t(506)=1.72, P =0.09; d=0.06). Full results are presented in Table 3.

TABLE 2.

Means and SDs of Outcomes by PTSD/Depression Status

Overall model
Outcomes	Pre, M (SD)	Post, M (SD)	Follow-up, M (SD)
General Belongingness	53.93 (16.50)	55.28 (15.88)	54.19 (15.84)
MSPSS Friends	4.97 (1.60)	5.00 (1.66)	4.96 (1.62)
QOLI Overall	1.62 (1.95)	1.65 (2.01)	1.69 (1.98)
PCL-5	37.72 (19.88)	35.13 (19.98)	35.23 (20.59)
PHQ-8	9.93 (6.70)	9.80 (6.95)	9.10 (6.62)
	Nonclinical group			Clinical group
Outcomes	Pre, M (SD)	Post, M (SD)	Follow-up, M (SD)	Pre, M (SD)	Post, M (SD)	Follow-up, M (SD)
General Belongingness	66.37 (14.56)	66.83 (13.43)	63.69 (14.26)	46.13 (13.15)	48.03 (13.11)	47.93 (13.61)
MSPSS Friends	5.66 (1.31)	5.68 (1.18)	5.71 (1.34)	4.61 (1.63)	4.60 (1.76)	4.52 (1.62)
QOLI Overall	2.92 (1.32)	2.71 (1.42)	2.65 (1.40)	0.86 (1.89)	1.09 (1.99)	1.10 (1.96)
PCL-5	15.89 (9.28)	17.72 (11.82)	18.43 (12.08)	49.56 (12.81)	45.25 (16.04)	44.17 (17.29)
PHQ-8	3.58 (2.74)	4.59 (4.21)	4.48 (4.03)	14.16 (4.98)	13.50 (6.03)	12.42 (5.69)

M indicates mean; MSPSS, Multidimensional Scale of Perceived Social Support; PCL-5, PTSD Checklist for DSM-5; PHQ-8, Patient Health Questionnaire-8; PTSD, post-traumatic stress disorder; QOLI, Quality of Life Inventory.

TABLE 3.

Results of Linear Mixed Models of Program Effectiveness for Clinical vs. Nonclinical Group

Overall model
Outcomes	Slope (95% CI)	t	P
General Belongingness	0.56 (−0.08, 1.20)	1.72	0.09
MSPSS Friends	0.01 (−0.06, 0.08)	0.38	0.71
QOLI Overall	0.03 (−0.06, 0.12)	0.62	0.53
PCL-5	−1.36 (−2.28, −0.44)	−2.89	<0.01
PHQ-8	−0.28 (−0.59, 0.02)	−1.85	0.07
	Nonclinical group			Clinical group
Outcomes	Slope (95% CI)	t	P	Slope (95% CI)	t	P	Interaction (95% CI)	t	P	Cohen’s d
General Belongingness	−0.87 (−2.12, 0.38)	−1.37	0.17	1.51 (0.65, 2.37)	3.47	<0.001	2.38 (0.87, 3.90)	3.09	<0.01	0.14
MSPSS Friends	0.03 (−0.11, 0.16)	0.37	0.71	0.00 (−0.09, 0.10)	0.04	0.96	−0.02 (−0.19, 0.14)	−0.28	0.78	−0.01
QOLI Overall	−0.11 (−0.30, 0.09)	−1.09	0.28	0.15 (0.02, 0.29)	2.26	0.02	0.26 (0.03, 0.49)	2.19	0.03	0.13
PCL-5	0.91 (−0.72, 2.54)	1.10	0.27	−2.73 (−3.84, −1.62)	−4.85	<0.0001	−3.64 (−5.61, −1.67)	−3.63	<0.001	−0.18
PHQ-8	0.42 (−0.15, 0.99)	1.45	0.15	−0.74 (−1.14, −0.34)	−3.65	<0.001	−1.16 (−1.85, −0.47)	−3.28	<0.01	−0.18

CI indicates confidence interval; MSPSS, Multidimensional Scale of Perceived Social Support; PCL-5, PTSD Checklist for DSM-5; PHQ-8, Patient Health Questionnaire-8; PTSD, post-traumatic stress disorder; QOLI, Quality of Life Inventory.

Completer Analyses

We were interested in the impact of WoVeN on those who received an adequate dose of the intervention (completers). We defined completers as those who attended at least 6 of the 8 WoVeN group sessions (56% of the sample, N=219). For noncompleters, the highest risk for drop-out was before the group even started (26% of noncompleters never started) and after session 1 (22%) or after session 2 (18%). Among completers, PTSD improvement over time remained significant (β=−1.96, t(420)=−3.15, P<0.01; d=0.23) and belongingness also significantly improved (β=1.15, t(487)=2.61, P<0.01; d=0.19). Improvement in depression neared significance (β=−0.39, t(467)=−1.91, P=0.06; d=0.14).

Clinical Subgroup Analysis

Given that we were particularly interested in the impact of WoVeN on those most in need of support, we examined the differential effects of the program on those meeting likely diagnostic criteria for PTSD and depression (N=205) compared with the nonclinical subgroup (Table 3). Results demonstrated that the clinical group improved significantly more on PTSD (interaction β=−3.64, t(391)=−3.63, P<0.001; d=0.18), depression (interaction β=−1.16, t(422)=−3.28, P<0.01; d=0.18), belongingness (interaction β=2.38, t(432)=3.09, P<0.01; d=0.14), and quality of life (interaction β=0.26, t(345)=2.19, P<0.05; d=0.13). There were no significant interactions on perceived support from friends (interaction β=−0.02, t(438)=−0.28, P=0.78; d=0.01). When we examined main effects within the clinical group only, we observed significant change on PTSD (β=−2.73, t(391)=−4.85, P<0.0001; d=0.34), depression (β=−0.74, t(422)=−3.65, P<0.001; d=0.26), belongingness (β=1.51, t(432)=3.47, P<0.001; d=0.26), and quality of life (β=0.15, t(345)=2.26, P<0.05; d=0.18). There was no main effect of time on support from friends (β=0.002, t(438)=0.04, P=0.96; d=0).

Finally, we conducted a completer analysis within the clinical group only (54%, N=111). These analyses demonstrated change over time on PTSD (β=−3.44, t(258)=−4.33, P<0.0001; d=0.42), depression (β=−0.82, t(271)=−2.76, P<0.01; d=0.30), belongingness (β=2.18, t(281)=3.73, P<0.001; d=0.46), and quality of life (β=0.21, t(225)=2.16, P<0.05; d=0.19).

Group Cohesion

Both the clinical and nonclinical groups reported high cohesion [clinical group: mean (SD)=120.8 (19.8); nonclinical group: mean (SD)=121.6 (20.0)] with no significant difference between-groups (t(181)=0.23, P=0.82).

DISCUSSION

This study examined the impact of WoVeN, a structured, peer-led support program, on aspects of well-being among women veterans. Our overall model suggested little change across primary outcomes of interest (quality of life, belongingness, and social support from friends) in the full sample of participants. These results were surprising given program evaluation data suggesting high levels of value and enjoyment, with a substantial impact across several domains of functioning.18 Examination of the baseline average for the friends subscale of our measure of social support revealed that, on average, the women were reporting perceptions of low support at baseline. However, the standard deviation was substantial, suggesting a wide range of baseline levels of perceived support. Essentially, a large portion of the participants reported good to high social support before the program. Overall effects of the program may have been masked due to ceiling effects for social support for this substantial number of participants. Alternatively, the overall lack of effect may reflect an insufficient dose of the intervention. In busy times and during a global pandemic, attendance in support groups may not be prioritized as demands of life arise. Interestingly, we did observe that when we evaluated change among WoVeN completers, belongingness significantly improved. Given that belongingness improved while perceptions of support from friends did not suggests that women might not develop close interpersonal relationships during their group experiences but may still benefit from being connected with their military community again.

Considering that we were not recruiting a clinical sample, we were surprised to see that over half of our sample exceeded likely clinical cutoffs for depression and/or PTSD. Although we had not hypothesized that the program would target posttraumatic stress in the full sample, we observed that posttraumatic stress symptoms decreased significantly over time, although the effect was small. This is consistent with prior research suggesting that increased support derived from community relationships can be a protective factor against PTSD.³⁶

Given the extent of clinical symptomatology, we were interested in understanding how the clinical subgroup benefitted from the program. Perceptions of group cohesion were high across the full sample and did not differ between subgroups suggesting that the presence of clinical symptoms did not hamper the clinical subgroups’ ability to integrate fully. Indeed, the program was particularly effective for those experiencing mental health symptoms as demonstrated by significant improvement in belongingness and overall quality of life in this subgroup. Although effects were small, those endorsing clinical levels of distress benefited more on primary outcomes compared with those who did not meet clinical cutoffs. Furthermore, although not anticipated, we observed significant decreases in PTSD and depression (again small effects) in our clinical subgroup. Perhaps most striking, those who received an adequate dose of the program (completers) showed the strongest effects on all primary outcomes as well as significant decreases on posttraumatic stress and depression (small to medium effects). These results suggest that the veterans who were most in need benefited the most on outcomes related to well-being, as well as experienced reductions in clinical symptoms.

This study is not without limitations. In the absence of a control condition, gains measured by standardized instruments cannot be solely attributed to the program. That said, the finding that completers benefited more than noncompleters supports the assertion that a higher dose of the content is related to more improvement on the targeted outcomes; however, firm conclusions about the dose-response relationship requires additional research. The study is further limited by the lack of fidelity ratings attesting to the adherence and competence of the peer leaders. Peer leaders were trained, the intervention is manualized, and regular consultation is available; however, the peer leaders are nonprofessional volunteers. Similar to other peer support programs, there is no way to confidently track our members’ participation in other programs. Likewise, it is difficult to track reasons for attrition or measure the impact of WoVeN on the women who chose not to participate in research. Future research might consider engaging peer leaders as study staff thereby increasing the ability to gather fidelity and participation data from them (eg, self-checklists to determine whether each of the elements of the protocol were administered) or conducting a full-scale, randomized trial with research-engaged participants and taped sessions which could later be rated by outside experts.

Despite these limitations, our results show that this type of social support network increases sense of belongingness for women veterans, supporting overall well-being. The impact appears to be particularly profound in improving belongingness and quality of life for those veterans who suffer from PTSD and/or depression, conditions for which isolation and loneliness are particularly salient. Being connected to a community of women veterans also appears to have a palliative effect on PTSS, particularly for those who receive a full dose. While clearly not a frontline intervention for PTSD, future research might consider testing the effects of this intervention in conjunction with evidence-based therapies for PTSD. Leveraging this type of program may also help women surmount barriers to care and provide a conduit for increasing women veterans’ access to resources.