Women’s and Men’s Experiences in Group Therapy for Substance Use Disorders: A Qualitative Analysis

Abstract

Background and Objectives:

This thematic analysis of qualitative interviews from participants in a Stage II randomized controlled trial examined women’s and men’s experiences in group therapy for substance use disorders (SUDs).

Methods:

Interviews were conducted with 77 women and 38 men after completion of either the gender-specific Women’s Recovery Group (WRG) or mixed-gender Group Drug Counseling (GDC). Interviews were coded for themes using a deductive approach with a coding scheme modified from the Stage I trial. Satisfaction was measured quantitatively post-treatment.

Results:

Participants had high satisfaction scores with no significant differences between groups. Women in GDC rated group gender composition as less helpful than those in WRG. In the GDC group, women more frequently discussed the theme of self-perception (e.g., feelings of comfort, safety, shame) compared to men. Men overwhelmingly expressed the benefits of having women in the group, whereas women expressed advantages and disadvantages of mixed-gender groups and preference for single-gender groups. Guilt and shame were discussed by women and men; however, only women discussed stigma and its important role in their addiction and recovery.

Discussion and Conclusion:

Men more frequently endorsed the helpfulness of mixed-gender groups than did women while women appreciated the enhanced support in single-gender SUD groups. Issues of stigma are especially salient for women.

Scientific Significance:

Men and women express differences in their experiences of SUD group therapy. Only women endorse stigma as an obstacle to their treatment and recovery. Tailoring treatment to meet women’s and men’s needs may enhance engagement, retention, and clinical outcomes.

Introduction

Gender differences in substance use disorders (SUDs) have important implications for treatment. Clinical characteristics such as co-occurring psychiatric disorders, trauma histories, and relationships with partners and children can play a significant role in women’s addiction and SUD treatment outcomes.¹ Gender-specific treatments have been developed to address these and other specific needs of women with SUDs with evidence of enhanced treatment outcomes,^2,3 as well as reports of feeling safer and more comfortable than in mixed-gender settings.^1,2

Single-gender and gender-specific treatments for women were developed following increased recognition in the 1970’s of gender differences in substance use, course of addiction, and social factors that affect recovery.¹ Gender-specific treatment is not synonymous with single-gender treatment; the former includes gender-specific content in addition to a women-only environment or group composition.¹ The theoretical rationale for gender-specific groups for women with SUDs has been well-described including gender disparities and barriers to addiction treatment entry, retention, and engagement, as well as differential gender distribution of characteristics that are predictors of SUD treatment outcomes such as psychiatric comorbidity, trauma histories, education and employment. ^2,4 Research on gender-specific treatments for women with SUDs has shown that these treatments are efficacious. Several randomized controlled trials (RCTs) of gender-specific SUD group treatments have been conducted with comparison conditions that include a gender-focused psychoeducation group,⁵ gender-neutral treatment,⁶ and mixed-gender group treatment.^7,8 Three of the studies found equivalent substance use outcomes for treatment and control groups, ^5,6,8 while a smaller Stage I trial of one group therapy found superior outcomes for women in the gender-specific group compared to women in the mixed-gender group at 6 months post-treatment.⁷ An additional RCT focused on women with alcohol use disorder compared female-specific cognitive behavioral therapy in individual and group format and found equivalent outcomes but did not use a mixed-gender control condition.⁹

Most of these single-gender group therapies were compared with other single-gender conditions.^5,9 However, the Women’s Recovery Group studies specifically compared the gender-specific group therapy for women to a mixed-gender group condition.^7,8 In the small semi-closed Stage I trial of the Women’s Recovery Group (WRG), substance use outcomes were superior at 6-months post treatment for the single-gender group treatment compared with mixed-gender group therapy,⁷ while in the larger 2-site rolling-group Stage II trial, substance use outcomes were positive but efficacy was equivalent to the mixed-gender group therapy control. However, additional secondary analyses of data from the Women’s Recovery Group (WRG) study demonstrated that participants exposed to groups with high levels of verbal affiliative statements had better SUD treatment outcomes at 6 months post-treatment completion, with the best treatment outcomes among women who experienced this high affiliation in the gender-specific WRG group treatment.¹⁰ This analysis provided some support for the study hypothesis that the all-women group composition of the WRG would enhance comfort and support, increase verbal affiliation and group cohesion, and this would be a potential mechanism to enhance substance use treatment outcomes.^7,10

While substance use disorder treatment was developed initially using mostly male populations and did not take gender-specific characteristics of women into account, neither was it necessarily gender-specific for men.^2,11 There is preliminary support for the use of gender-specific group therapy for men;¹² however, findings related to the benefit of an all-male group are mixed. In a study of a mindfulness-based relapse prevention intervention, group gender composition moderated treatment outcomes, with the greatest reduction in drug use days when the group was comprised of one third or more women.¹³

While it is clear that gender-specific factors can play a role in both men’s and women’s substance use and treatment response, less is known about how certain gender-related factors, such as group gender composition and gender-specific content, relate to patients’ experiences in treatment. Several attributes may contribute to the effectiveness of single-gender group therapy for women. Women have reported that the interpersonal environment and opportunity to share similar experiences enhances their sense of comfort in treatment.¹⁴ One hypothesized mechanism of action is group cohesion,¹⁵ which has been shown to have higher ratings in all-women’s groups compared to all men’s groups.¹⁶ More broadly, women often state preferences for all-women group therapy,¹⁷ and converging evidence shows that matching patients with their preferred treatment can increase retention and improve clinical outcomes.^18,19 Increased understanding of both men’s and women’s needs, experiences, and preferences is an essential step in enhancing substance use treatment. Qualitative research can be especially helpful in elucidating patients’ experiences and relationships to their treatment outcomes and satisfaction.^20,21

The current study builds upon a previous qualitative analysis of semi-structured interviews from the Women’s Recovery Group (WRG) Study from the Stage I trial. The WRG²² is an evidence-based, 12-week, manualized, relapse prevention group therapy for women with SUDs.^7,8,22 The WRG was designed with two key elements: all-women group composition and women-specific group content. Content for each session is based on gender-specific substance use antecedents, consequences, and treatment outcomes, including: the effect of substances on women’s health; women’s relationships and recovery; violence and substance use; co-occurring disorders (e.g., mood, anxiety, and eating problems) and substance use; recovery skills; stigma, shame, and disclosure; being a caregiver and being in recovery; self-help groups for women; and achieving a balance (see Greenfield et al., 2007⁷ and Greenfield, 2016²² for a more detailed description of group content). The WRG was initially developed and compared to mixed-gender Group Drug Counseling (GDC) in a small Stage I trial, and then in a larger two-site randomized controlled Stage II trial. In both the Stage I and Stage II trials, women in the WRG and GDC demonstrated clinically relevant reductions in substance use after 12-weeks of group treatment.^7,8 However, women in the WRG Stage I trial continued reductions in substance use at 6-months post treatment whereas GDC women did not. In the Stage I trial, 28 semi-structured qualitative interviews of women participants were coded to explore their experiences in single-gender versus mixed-gender group therapy.²¹ Using a grounded theory approach, this qualitative analysis demonstrated that compared with women in GDC, women in the WRG reported a sense of safety, support, and ease of communication with other women.²¹ However, this study was limited by its small sample size (n = 22, WRG; n = 6, GDC) and analysis of interviews from only women participants.

In the Stage II WRG trial, semi-structured interviews with both women and men were conducted to examine participants’ experiences in the group treatments. The goal of the current analysis is to compare women’s experiences in gender-specific versus mixed-gender treatment, as well as to better understand differences between men’s and women’s self-reported treatment experiences. To achieve this, we examined by gender and group therapy condition, the differences in 1) ratings of satisfaction and helpfulness of the group, and 2) frequency with which specific themes were discussed. Next, we conducted a qualitative examination of the themes that differed significantly by gender and group therapy condition.

Methods

Participants

All study procedures were approved by the Mass General Brigham Institutional Review Board. The WRG Stage II trial (N = 158) compared outcomes for women in the two group therapies: WRG and GDC. Recruitment materials were targeted to individuals trying to achieve or maintain abstinence from substances who were willing to participate in group therapy. Participants were included in the study if they were 18 years of age or older, met DSM-IV criteria for substance dependence, and used substances in the past 60 days. Participants were excluded from the study if they were diagnosed with certain Axis I psychiatric disorders (e.g., psychotic disorders), had a current medical condition that would prevent regular group attendance, and were current intravenous drug users (see Greenfield et al. 2014⁸ for full description of methods). The mean number of groups attended by all participants was 7.92 (SD= 3.31).²³ Of the 158 participants enrolled in the Stage II trial, 115 (42 WRG women, 35 GDC women, 38 GDC men) completed exit interviews; interviews were conducted at either 6- or 9-month follow-up visits. All enrolled participants were contacted for follow-up visits regardless of their attendance in the treatment groups. Participants who completed exit interviews attended significantly more groups (M = 8.99, SD = 2.72) than those who did not complete exit interviews (M = 5.07, SD = 3.08; t = −7.77, df = 156, p < 0.001). Unlike an earlier analysis using data from the Stage I exit interviews,²¹ men were interviewed and included in the current analyses.

Measures

Group satisfaction and experience was measured at the end of treatment using the Group Overview Questionnaire (GOQ).⁸ Participants were asked how helpful aspects of the group were on a scale of 0 to 4 (0 = not at all helpful; 4 = extremely helpful), and how much they liked the group overall on a scale of 0 to 4 (0 = not at all; 4 = liked a lot).

Exit interviews were individual, semi-structured interviews, ranging from 9 – 63 minutes in duration (average length = 22 minutes), and conducted by study staff. Participants were asked about their satisfaction with and experience in the group therapies, and how they compared with previous group treatment. Exit Interview questions were the same as asked in the Stage I trial.²¹

Coding Process

Initially we reviewed four Stage II transcripts using the coding taxonomy that was developed for the Stage I study to ascertain duplication of themes or new themes. The Stage I coding taxonomy was developed using grounded theory and included 55 axial codes of group attributes valued by participants that were then categorized into five theoretical codes to provide a meaningful description of the data.²¹ After initial review of this sample of Stage II transcripts, the taxonomy was edited such that similar axial codes were condensed (e.g., “feelings of being guarded” and “women feeling constrained” were collapsed into the theme “guarded, constrained, holding back”) and additional codes were added to capture themes related to male participants. One research assistant (MER; Rater 1) and one doctoral-level researcher (DES; Rater 2) coded thirteen transcripts to establish interrater reliability (measured by achieving a Cohen’s kappa of at least .70 on five consecutive transcripts). After each transcript, Rater 1 and Rater 2 discussed and iteratively edited the taxonomy. A final version of the coding scheme was developed and approved by the PI (SFG) before assigning any final codes. The final, modified coding scheme was refined to 30 themes that were categorized into five theoretical codes: 1) Perceptions of Atmosphere of Group, 2) Perceptions of Group Members, 3) Self-Perceptions, 4) Gender Differences, and 5) Therapist-related (Table 1). Themes were coded if they were present regardless of valence (e.g., “feeling safe” was coded if a participant discussed feeling safe in the group or feeling unsafe). The transcripts were coded using QSR International’s NVivo 11 Pro,²⁴ a qualitative analysis software that allows users to upload and manage a variety of source materials.

Table 1.

Final Themes and Theoretical Codes

Theme	Example Quote
Perceptions of Atmosphere of Group
Communication (i.e., open and honest; effective and productive communication; listening and contributing)	“I think it’s easier to be honest. That it’s easier to speak. And that it’s easier to really and I think it comes with the honesty piece—the emotions come out.”
Confrontation and conflict	“I just said well—her trigger was having money. Mine wasn’t money. So I just mentioned that. Well, that—and she really got mad at me. And I was—I’m very sensitive. I said, “I’m not coming back.”
Feeling alone	“There was a fair amount of things that I felt like I didn’t relate to the other women on.”
Intimacy/closeness/connectedness/cohesiveness (i.e., feeling close to group members OR not feeling alone)	“…we kind of all had some common threads with our addiction. We were all able to relate to each other really well, even though we’re from fairly different walks of life. There was a lot of commonality.”
Missing attributes/suggestions for improvement	“I think that—maybe it could have been another half hour longer.”
Participation, discussion, and sharing (referring to frequency, not quality)	“There were some members that weren’t as engaged as others, but that didn’t ruin it for everyone else, those that wanted to share did share.”
Shared or received tips or advice (i.e., shared tips/advice, received tips/advice)	“And everybody in the group was very caring. If someone had a husband problem or children’s problem everyone offered some sort of have you tried this. Have you tried that? It became much more than just a woman’s group.”
Specific topics (mentioning any one of the fourteen topics, or referring to topics overall)	“certainly the topics that we covered […] they were definitely appropriate and they were interesting. They were helpful for more self-knowledge.”
Structure (including format, size, length, elements, duration of group)	“We stuck to the agenda and if somebody starting going off a little bit, it was brought right back to the agenda, which I like that.”
Understanding/empathy	“It’s always nice to be with someone else that understands what is happening”
Perceptions of Group Members
Differences and diversity among group	“Just to hear different age groups, different people kind of going through the same thing, trying to help each other in the same kind of ways.”
Different stage of recovery	“I liked that it was kind of—for the few people that were in it, that it was diverse. You had very different people in different stages of their alcoholism or recovery. It just makes it interesting.”
Gender composition of group	“I think it’s easier to be honest [in all-women’s group]. That it’s easier to speak. […] the emotions come out.”
Learned from the group (i.e., learned from others, the leader, the group in general)	“It’s helpful to learn from other peoples’ experiences as well, things that you wouldn’t have thought of on your own.”
Relationships (i.e., thinking about group members, liking group members, engages with members outside of group)	“The group I had had so much camaraderie and almost a friendship, towards the middle of the group or whatever the subject was, people got down to bare bones.”
Similar stage of recovery	“In particular one of the guys in my group we had very similar using histories and seriousness of problem, so it really was helpful to see him every week and we started at the same time and ended at the same time, so I was in there the entire time with him and our problem with drinking was more severe than most of the other people there. So it definitely helped to have somebody in the group who happened to be at the same stage I feel like I am.”
Similarities among group	“These people are like me. We have these commonalities. Age. Teenage kids. Long-term relationships. The drinking in the midst of a long-term relationship. That whole thing.”
Supportive/considerate	“There was a lot of interaction, a lot of exchange and a lot of empathy […] I think that we all shared that and supported each other.”
Self-Perceptions
Comfortable	“I felt more comfortable sharing certain topics with an all-female group.”
Feeling Guarded	“I would hold back on things in a mixed group”
Issues of Guilt, Shame, and Stigma	“I think women when they use substances tend to have a lot of similar issues that – I think there’s just I think a lot more guilt attached to it”
Safety	“It’s also a little safer to me, there are probably some things that I wouldn’t have felt comfortable talking about if it was a mixed group.”
Self-reflection, awareness, motivation, and self-change (as a result of the group)	“It kind of just forced you to think and really go through — okay, now like what did I do this week, how was I really feeling, how was I thinking…”
Use of information	“The literature was great. I took it home with me and I did, on occasion, look back to just refresh my memory on some of the things.”
Gender Differences
Gender differences	“For some reason, I think women in general just have different experiences than men going through substance abuse. Especially women who are mothers.”
Importance of opposite gender feedback	“I definitely feel like I can always relate more to women, but I like […] hearing men’s aspects as well”
Men’s group suggestions	“I think probably relationships and how substance abuse affects relationships with women and—because I do notice there’s a lot of anger with men towards their exes, ex-wives, ex-girlfriends and you may want to focus, ‘Well, it can’t be all her fault.’ Because I think men have a difficult time, talking about their relationships with women.”
No difference in men and women (i.e., a statement like: “There is no difference in x, y, or z”)	“I strongly suggest that they keep it mixed because, to me, the disease doesn’t pick whether it’s a female or a male.”
Therapist-Related
Gender of therapist	“I don’t know if it matters, but I liked that it was a female running the group, I mean it worked out nice, I liked that.”
Therapist	“The group leader. That really made it work well. The fact that there was a topic she followed and always kept us back on track.”

Procedure

Rater 1 coded 100% (n = 115) of the transcripts. Thirty-five transcripts (30%) were randomly selected for Rater 2 to code. The average kappa coefficient across the 35 dually coded transcripts was .81, suggesting moderate to high interrater reliability. After each transcript was dually coded, Raters 1 and 2 discussed any disagreements in coding until they reached 100% agreement on the assigned codes.

Results

Quantitative Analyses

Demographics and Background Characteristics

Table 2 provides a summary of the demographic and substance use characteristics of the sample. The majority of participants were white (96%) and well-educated (54% with college graduate or postgraduate education). Eighty-four participants (73%) reported alcohol as their primary substance of use, with cocaine as the next most common (8%). At baseline, participants reported an average of 18 days of substance use out of the past 30 days. Women were more likely than men to report a history of trauma (defined as any sexual, physical, or emotional abuse; 60% vs 32%, respectively).

Table 2.

Demographics and Substance Use Characteristics (N = 115)

Demographics
Age (years), M (SD)	47.2 (12.4)
Range	20–79
Sex, n (%)
Male	38 (33.0%)
Female	77 (67.0%)
Ethnicity, n (%)
Non-Hispanic	114 (99.1%)
Race, n (%)
White	110 (95.7%)
Marital status, n (%)
Married/living with partner	47 (40.9%)
Widowed	2 (1.7%)
Divorced/separated	33 (28.7%)
Never married	33 (28.7%)
Education, n (%)
Less than high school	8 (7.0%)
Graduated from high school	22 (19.1%)
Some college	23 (20.0%)
Graduated from college	37 (32.2%)
Postgraduate	25 (21.7%)
Occupational status, n (%)
Employed full time	27 (23.5%)
Employed part time	14 (12.1%)
Works within the home	4 (3.5%)
Unemployed	46 (40.0%)
Retired	6 (5.2%)
Disability	15 (13.0%)
Student	3 (2.6%)
Substance use characteristics
Substance use days (past 30 days), M (SD)
Alcohol use days	14.8 (10.4)
Days of any substance use (including alcohol)	17.9 (9.4)
Drug use days (excluding alcohol)	5.0 (9.2)
Primary substance use days	15.3 (10.4)
Heavy drinking days	12.4 (10.8)
Drinks per drinking day	10.0 (6.6)
Primary substance, n (%)
Alcohol	84 (73.0%)
Marijuana	4 (3.5%)
Sedatives	3 (2.6%)
Opioids	4 (3.5%)
Cocaine	10 (8.7%)
Other	10 (8.7%)

A large majority of participants had previous experience in mixed-gender groups (WRG women = 86%, GDC women = 94%, GDC men = 97%). Experience in single-gender groups was lower, but still highly prevalent (WRG women = 52%, GDC women = 79%, GDC men = 63%). Most participants (89%) indicated that their past experiences with single- and mixed-gender substance use recovery groups were through mutual help groups (e.g., Alcoholics Anonymous, Narcotics Anonymous) and not through professionally-led recovery groups.

Group Satisfaction and Helpfulness of the Group Gender Composition

Scores on the GOQ indicated that participants were generally satisfied with both groups (WRG: M = 3.8, SD = 0.5; GDC: M = 3.7, SD = 0.6). There were no significant differences in ratings of satisfaction between women in the two groups (WRG vs. GDC), or between men and women in GDC. However, women in the GDC rated the gender composition of the group as significantly less helpful than women in the WRG (t = 2.15, df = 72, p < .05); there were no significant gender differences in ratings of helpfulness of the gender composition between men and women in the GDC group.

Comparison of Frequency of Themes by Group Therapy Condition and by Gender

To examine differences in the frequency with which themes were discussed by women GDC participants versus women in the WRG as well as gender differences in themes discussed in the mixed-gender group (GDC men vs. GDC women), we compared the frequency of the five theoretical codes using independent sample t-tests. We used Bonferroni correction to minimize the risk of type I errors, and thus the two-tailed significance threshold was set at 0.005 (0.05/10).

Out of five theoretical codes (Table 1), only the theoretical code ‘Self-Perceptions’ differed by gender (t=3.11, df=56.9, p= 0.003), with women in the GDC more frequently discussing themes categorized in this code than men in the GDC. There were no significant differences in frequency of the five theoretical codes for women by group therapy condition (WRG vs. GDC).

Qualitative Analyses

We have followed the guidance outlined by Neale and colleagues²⁵ regarding reporting quantitative information in qualitative research. In particular, we have limited reporting of counts to questions that were assessed for all participants, avoided terms that have a specific quantitative meaning (e.g., ‘majority’, ‘minority’, ‘most’), and when necessary, used appropriately non-specific terms (e.g., ‘a few’, ‘several’, ‘some’, ‘many’).

Women’s Perceptions of Gender Composition in Single- Versus Mixed-Gender Groups

To explore the finding that women in the GDC group rated the gender composition of the group as less helpful than women in the WRG, we examined the theme ‘Gender Composition of Group’. Table 3 provides example quotes from participants to further illustrate the findings below.

Table 3.

Perceptions of Gender Composition in Single- Versus Mixed-Gender Groups

Benefits of Single-Gender Groups
WRG Female	GDC Female		GDC Male
“Surprisingly, I liked the fact that it was women….talking with different struggles that are gender specific I think would not have happened — in a mixed gender group. It would have been more about, I think, the alcohol itself, and less about the reasons why.”	“I was actually hoping I’d be in an all-women’s group because when I go to AA, the women’s stories appeal to me much more…. I just find that in a women’s group that you hear people that have so much in common with you in the sense that they don’t stop by a bar on the way home after work. They maybe run home to take care of their kids but they drink while they’re doing it.”		“I do go to all-men’s groups, which I do find a little bit more open…’”
“I think it’s easier to be frank and more open than I would be in a mixed group…..The past, physical abuse; I wouldn’t talk about those things in a mixed group.”	“You feel like the conversation gets on to a deeper emotional level when it’s all women… I just feel more comfortable and I feel that there’s more like, I don’t know how to explain it, like kinship.”		“I think in the men’s group, depending, I think they would be able to share easier because we know we can be foolish, where maybe in front of the women, we always want to be as an adult.”
I don’t think I would have opened up…I think women have different issues, have different understandings about life…the themes and the material we went over….is more appropriate for women…I wouldn’t have gone if it had been a mixed group.”	“I think that there is a difference in male/female experiences with addiction….there are different reasons, different circumstances and psychologically different needs that men and women have…so this is your opportunity to speak and be heard and I think that women – we do that, we support each other and we listen…for some women it’s a safer place to have just all women and that’s important too.”
“I really empathized with the women, and felt that they empathized with me. I think that there was a feeling of camaraderie in the single sex groups that was really important. And I could see it being important to other people.”	“…from my experience…people will become more vulnerable, and more open and honest because they’re all women. So they’re more likely to say what’s on their mind…”
“I didn’t think it would make a difference to me, but it did. It really did. I thought upon the things that we talked about and it just hit home…was focused more on women…I think I just feel a little more open with women…”	‘I would have preferred if it wasn’t coed…because a lot of the things they talked about…I couldn’t really relate to…men I guess I feel they don’t share as …deep of an emotional level as women do…’
“I really felt it was important for me. I’m an addict and I’m a mother who lost her children. So I just feel the connection, women to women, more intimacy versus being in a room with males…I don’t feel that intimacy or that female-to-female connection as if I would if a guy or more guys were in the room…”	“I think women when they use substances tend to have a lot of similar issues….you tend to get into some of the abuse issues or the abandonment…”
“I think you can speak more when are around the same sex or the same gender about your problems and about things that are – something as complex as alcoholism I believe there are a lot of different issues that arise that sometimes could be uncomfortable…things you did in your past, or things that you have experienced while being under the influence…that are easier to discuss in an all-female environment…”
“I think women tend to have different problems, different issues, different reasons, different times that they would drink, trying to hide it during the day, wine in the coffee cup in the morning and hiding it from the kids, stuff like that…”
“I definitely think that, and this was present in the group, the feeling that you’re not going to be judged and ..that hasn’t happened to me often.”
“I feel like women are more comfortable speaking around other women. I feel like sometimes they feel like they’re going to be judged by a man for certain responses or certain scenarios or stories.”
Benefits of Mixed-Gender Groups
GDC Female		GDC Male
“…[It was] helpful….to see the challenges that guys deal with and how frightened they are, and [how] hard it is in marriage…or a guy who’s supposed to be this tough guy and not able to express himself to his wife….to have that was incredibly helpful.”		“I’ve been in groups where it’s just all males and no females, and it kind of gets, sometimes off of the subject that you’re talking about.”
“I liked the mixed-gender just because….it gives you a different viewpoint.”		“Having females there could start a conversation because, again, males like to show off at times…”
		“Having women involved…softens it…you can try to macho it up a little bit but you’re not going to go overboard…”
		“I also think women are better listeners, more empathic listeners…I thought it was nice to have a couple of women. I was probably more open than I might have been in front of other guys…[women] are more empathic”
		“Girls tend to open up quicker and that, you know, after they kind of get the ball rolling, then a guy will be more likely to talk on the matter…”
		“I like the mixed gender approach better too because there’s a common bond with the problem, but there’s different ways of analyzing and coping with it or responding to it, different viewpoints….It was just interesting to get a woman’s opinion of the exact same situation I’ve been in and find out what her thoughts were about it…”
		“…from a woman’s perspective it seemed like it was being able to give input and feedback on things that, you know I’m having this problem…It seems like the comments weren’t things that would have been generated by a group of men…things that I did find helpful, so I think it was good.”
		“The men are better behaved, sort of, when they’re in a group with women…”
		“…it makes for more respectful dynamics in the group…I think it is a more respectful atmosphere when you have mixed-gender. I really liked listening to the women’s perspectives on different things. I found that very intriguing…”
		“…I really like this because there were a few women that I got a lot out of what they were saying and just to hear their perspective on what I was saying I thought was really helpful…”

Women in the GDC expressed both advantages and disadvantages of mixed-gender groups. Some women appreciated mixed-gender groups because they found opposite gender feedback helpful as well as learning more about men’s experiences. However, GDC women frequently expressed a wide range of benefits to their past experiences in single-gender groups, including being able to relate to others easily and feeling a sense of comfort and strong group connection.

Women in the WRG group overwhelmingly described the all-women’s gender composition as a facilitator for discussing gender-specific issues that affect their recovery. In addition, women in the WRG group described the importance of being able to speak freely without feeling judged by men. While many women in the GDC group expressed a preference or desire to be in a single-gender group, GDC men often endorsed the helpfulness of mixed-gender groups compared with single-gender groups because of the opportunity to get a different perspective, gain an understanding of women’s experiences with addiction, and the sense that women were more empathic and facilitated group discussion. In particular, several men described feeling more comfortable opening up about relationship issues in a mixed-gender group compared to an all-male group. Conversely, many women in the GDC group said that they felt more comfortable discussing relationship issues in all-female groups. While several men in the GDC group described advantages with having women in the group, a few men expressed a sense that they might have been more able to talk about certain topics more openly in a single-gender group.

Gender Differences in Self-Perceptions Among Women and Men in GDC

To better understand the unique experiences of women and men, we conducted a close examination of the six themes that comprised the Self-Perceptions theoretical code: 1) Feeling Comfortable, 2) Feeling Guarded, constrained, holding back, 3) Having feelings of Safety, 4) Self-reflection, awareness, motivation, and self-change, 5) Stigma, shame, and guilt, and 6) Use of information. Among these six themes, differences in the ways that women and men in the GDC groups discussed their experiences emerged within one theme: Stigma, shame, and guilt. Table 4 outlines these differences with illustrative quotes.

Table 4.

Gender Differences in Stigma, Shame, and Guilt Among Women and Men

WRG Female	GDC Female	GDC Male
“It is still something I struggle with, especially as a mother, to admit the things that I did. And just to think to myself, people are going to think I am the scum of the earth…I think that when you hear of a female alcohol or drug addict the immediate perception is…she must be like a total neurotic crazy lady. Whereas a man—it’s almost more acceptable for a man to be an alcoholic and get this idea of this sort of business man who drinks his lunch.”	“I think outside these walls—for me at least—every day I feel very ashamed and embarrassed that I have this problem.”	“Guilt and shame are my two top feelings. I wish I could believe it’s a disease”
“… Women who have no control over their urges are somehow less women, whereas men are just kind of, you know, boys…boys will be boys….my perception and I think it’s maybe a societal perception that women who drink are just…bad, sloppy people; and men who drink are…out to have a good time…”	“…the women all have common issues around…the shame and the guilt…”	“I try to close my eyes to [certain things] in recovery and…the shame I have towards my family is just terrible.”
	“I don’t always want to do the shame and guilt thing….I think men tend to release the shame and guilt a lot easier than women. We tend to kind of drag it along with us for a longer amount of time….”
	“it’s less acceptable for a woman to have a drinking problem”
	“I think it’s easier for women to get sucked down by their illness, because it’s more of a negative for women in society, the way we’re looked at.”
	“I notice I was clean for a long time and I have a really hard time with society today….”

Experiences of guilt and shame were discussed by both women and men. For example, participants of both genders described the shame and guilt that they felt associated with their addiction. While discussion of shame and guilt was frequent for both men and women in the GDC group, only women raised the issue of stigma. Addiction stigma is represented by negative societal attitudes toward individuals with alcohol and drug problems.²⁶ Women expressed feeling societal judgment about their substance issues and that society’s perceptions of women with addiction are harsher than those of men with addiction.

Women in the WRG expressed feelings of shame and guilt and, commensurate with women in the GDC group, also discussed societal stigma and judgment of their addiction especially as related to their roles as women and mothers.

Discussion

This qualitative analysis of exit interviews of women and men who participated in the Stage II Women’s Recovery Group study replicated certain findings from qualitative interviews with women participants from the smaller Stage I trial. This study also extended these earlier findings by examining and comparing experiences in the group treatments between men and women participants in the GDC. Main outcomes from this Stage II trial showed that women in both groups had clinically relevant reductions in substance use post-treatment and maintained those reductions throughout the follow-up period.⁸ Consistent with these outcomes, we found high ratings of satisfaction among women in both the gender-specific WRG and mixed-gender GDC.²¹

Despite equally high ratings of satisfaction, women in the GDC rated the gender composition of the groups as less helpful than women in the WRG. Of note, there were no significant differences in number of treatment sessions attended for women in WRG versus GDC;⁸ therefore, these results cannot be explained by greater treatment exposure in the WRG group. There is scarce research elucidating the mechanisms that contribute to SUD group therapy effectiveness²⁷ including whether group treatment leads to enhanced social support outside of treatment itself. Previous research has ascribed effectiveness of group therapy to processes including in-session factors²⁸ such as group cohesion.²⁹ The WRG is the first study to measure cohesion by the observable phenomenon of verbal affiliation¹⁰ and higher levels of verbal affiliation were found in the WRG compared with the mixed-gender GDC.³⁰ Although we did not specifically ask participants during the exit interview if they were in contact with other group members, nine participants (all women) spoke about exchanging phone numbers, meeting with group members outside of group, or staying in touch with group members after the group ended. Of these nine women, seven were in the WRG group. In a previous study, we demonstrated that women in the WRG group were exposed to higher levels of verbal affiliative statements³⁰ and a subsequent analysis showed exposure to groups with high verbal affiliative predicted better SUD outcomes at 6-months post treatment especially in women exposed to high-affiliation WRG groups.¹⁰ Verbal affiliation is one potential mechanism of enhanced support in the gender-specific WRG group, which may have fostered connections among group members. It is possible that enhanced affiliation within the group facilitated increased contact with other group participants outside of treatment and enhanced recovery support; however, this was not assessed. Another possibility is that greater affiliation within the gender-specific WRG group provided a supportive experience for women that enhanced their learning and consolidated skills that facilitate relapse prevention and improve SUD treatment outcomes. Psychotherapy-by-time effects have been demonstrated for individual treatments³¹ and were reported in the Stage I trial of the WRG possibly due to women’s reduced feelings of stigma, increased sense of support, and identification of gender-specific triggers among other elements.⁷ Additional research is needed to investigate the processes by which increased in-session affiliation in the gender-specific group may enhance women’s SUD treatment outcomes.

This study also replicated the smaller trial results insofar as women in both groups more frequently expressed preferences for single-gender group treatment. In the Stage I trial, women expressed feeling that in single-gender SUD group therapy they were more able to have their needs met, were supportive of one another, and that they could be more honest, empathic and comfortable as well as speak about topics relevant to women.²¹ In the present study, women in both groups endorsed preferences overall for women-only groups. They more frequently endorsed that single-gender groups promoted greater feelings of comfort, empathy, support and a strong group connection as well as a wide range of other additional benefits to single-gender groups. Similarly, women in the WRG said they preferred single-gender groups because of their experience that these groups provided exposure to narratives with which they had more in common, felt a deeper emotional connection with women group members, as well as a feeling of kinship.

The qualitative data confirmed our understanding that women in the WRG felt they benefited from the all-women’s group composition for enhanced intimacy, support, and empathy and ability to be open and honest about their addiction and other life circumstances. In addition, women in the GDC expressed preferences for all-women’s groups because they felt there is more comfort, kinship, and safety. Our prior studies of verbal affiliation document that there are 66% more verbal affiliative statements in the WRG compared with the mixed-gender GDC³⁰ and these interviews describe women’s experiences of this verbal affiliation. The qualitative data extended our previous findings that the non-judgmental atmosphere of the WRG enabled women to discuss the stigma they have felt related to being a woman with a substance use disorder.

The present study differed from the earlier Stage I trial in that it had the opportunity to compare the experiences of men and women in the mixed-gender GDC. While both men and women in GDC described appreciating opposite-gender feedback and gaining insight into the opposite-gender’s experiences as advantages of mixed-gender treatment, men expressed a feeling that women were more empathic, more helpful at facilitating discussion, and more respectful than men. Men also said that they preferred talking about relationship issues in mixed-gender groups, whereas women more often said they were more comfortable discussing relationship issues in single-gender groups. Overall, the men’s comments regarding mixed-gender treatment focused more on what women offered the group rather than what men contributed. This is consistent with previous research that concluded that men in mixed-gender groups may be more likely to experience a supportive group environment than women in mixed-gender groups.³² As the focus of the study was a comparison of women randomized to a gender-specific WRG versus a mixed-gender control condition, we do not have a randomized comparison for men in GDC with a single-gender group therapy. In this study, men were asked to compare their experiences with past group treatments, most of which was in mutual help groups. We heard from both men and women that some of the elements they liked about both of the study groups (WRG and GDC) was that it was professionally led and more structured than mutual-help groups. A small pilot trial (N=10) of a single-gender men’s group therapy that was professionally led showed that men endorsed high levels of satisfaction and felt that the all-male setting provided them an opportunity to express feelings and experiences they would not be able to express in a mixed-gender setting.¹² A study comparing men in mixed-gender treatment with a gender-specific group therapy for men would be informative in assessing satisfaction, preferences, and outcomes.

We did not see any gender differences in the GDC group in text that was coded as feeling guarded, constrained, or holding back. That said, three women described not wanting to discuss sexual abuse in mixed-gender groups. It is possible that other women had trauma-related experiences and similar feelings but did not want to disclose this information during the exit interview. Most often, text that was coded as guarded, constrained, or holding back, involved participants of both sexes describing not wanting to discuss certain topics in groups that are mixed-gender; however, participants did not always elaborate on what those topics were.

Issues of shame and guilt were salient for both men and women in this study. Prior research has identified that patients can benefit from addiction treatment that helps reduce feelings of shame.³³ Both men and women in the Stage II trial expressed that acceptance and the non-judgmental stance of their peers in both GDC and WRG were helpful attributes of these group treatments in reducing feelings of shame. That said, shame and stigma are related but distinct constructs and stigma, as opposed to shame, has been demonstrated for other health conditions such as sexually transmitted diseases and HIV/AIDS as barriers to seeking care.³⁴ Stigma is defined as “an attribute or label that sets a person apart from others and links the labeled person to undesirable characteristics.”³⁴ These societal stigmatizing attitudes are well understood by the people who are the targets of these attitudes and are often internalized and can be a barrier to seeking treatment.³⁵ Experience of societal stigma has long been discussed as a barrier to women’s accessing and seeking treatment for substance use disorders.³⁶ A recent systematic review found that studies that have measured drug use-related stigma using quantitative methods showed no gender differences; however, when examined in qualitative studies, drug use-related stigma was more prevalent in women in nearly all of the articles reviewed.³⁷ In our qualitative analysis of interviews from this Stage II trial, it is notable that while both men and women in the GDC reported feelings of shame and guilt, only women discussed the issue of societal stigma. In particular, the women in this study expressed stigma related to two of the themes identified in the Meyers et al.³⁷ review, (1) the “double” stigma of being a woman and an individual with an SUD, and (2) societal expectations of womanhood and their impact on drug use-related stigma. These findings further support the importance of qualitative research in understanding gender differences in the experience of stigma for individuals with SUDs. Moreover, these results extend the previous literature by highlighting the role of gender-specific group therapy in helping women discuss stigma-related issues of being a woman with an SUD without feeling judged.

There are several limitations to this study. The Stage II trial was conducted in a majority white sample with participants with high levels of education. The US Census Bureau data estimated that at the time of the study enrollment, 85% of the Massachusetts population was White;³⁸ therefore, our sample had a larger majority of White participants compared to the statewide data. Given the intersectionality of gender and race/ethnicity, we do not know whether these findings would generalize to more diverse populations. Although there were too few minority participants to conduct statistical comparisons of satisfaction scores, we explored the transcripts of the 5 non-White participants (all women, 4 identified as Black and 1 identified as Asian) and did not detect any differences in the themes compared to what was reported for the overall sample. We did not ask participants if they preferred same-race groups, and none of the participants discussed issues related to race or preferences for same-race groups in their exit interviews. With regard to education, we examined satisfaction scores for participants with high school or less education (M = 3.7, SD = 0.7) compared to those with more than a high school education (M = 3.8, SD = 0.5) and found no differences in satisfaction scores (t = −0.6, df = 109, p = 0.58). Examination of the transcripts for individuals with a high school or less education did not reveal any discussion of issues related to education level of themselves or their peers in the group.

Finally, although we attempted to interview all study participants, participants who completed the exit interviews attended more groups on average than those who did not complete the exit interviews, a difference that is statistically significant. We employed multiple strategies to retain participants in groups.²³ However, participants who had difficulty attending the groups may also have had difficulty attending the research visits for similar reasons (e.g., transportation issues, severity of illness). It is also possible that those who attended fewer groups may have been less satisfied with the treatment and/or have different experiences than what was reflected from the interviews of participants who attended more groups. Understanding the experiences of individuals who have low group attendance is important and future research should determine additional methods to engage these participants in research visits.

In spite of these limitations, to our knowledge, this is the first study to conduct in-depth interviews with men and women with SUDs about their experiences and satisfaction in mixed-gender compared with single-gender group therapy. This study’s findings demonstrate that while women are highly satisfied with either empirically supported gender-specific or mixed-gender SUD group therapy, on balance, if given the choice, women prefer gender-specific group therapy for women with SUDs citing enhanced empathy, comfort, support, gender-specific relevance and safety that these groups afford them. However, men in this study preferred mixed-gender group therapy citing additional empathy and support afforded by the women group participants as well as an appreciation for understanding the experiences of women with SUDs. Given that there is some evidence that empathic and supportive statements made in mixed-gender groups are not bidirectional and that men are more likely to be the recipient of these statements,³⁰ women who engage in mixed-gender groups might benefit from adjunctive gender-specific treatment. It is important to note that women highlight not only the single-sex aspect of this group as important but also endorse the gender-specific nature of this professionally led group compared with single-sex groups such as those they had experienced through mutual support programs. However, not all women will have access to in-person gender-responsive group treatments. Technology-based programs may be able to fill this gap in a cost-effective way.³⁹ In addition, it will be important that future SUD treatment research focus on whether gender-responsive treatment approaches enhance participant engagement and retention in treatment in addition to other clinical outcomes. This study also demonstrates that shame is a common and powerfully expressed emotion by both women and men in addiction treatment; however, women expressed that the societal stigma they experience as women with addiction is a factor in their addiction and recovery. The qualitative data from this study extended our previous findings that the non-judgmental atmosphere of the WRG enabled women to discuss many aspects of their experience of addiction including the stigma they have felt related to being a woman with a substance use disorder. Perception of societal stigma is a powerful barrier for women in accessing addiction treatment. Assessment of interventions targeted at stigma reduction for women with SUDs and entrance and retention in treatment would be valuable given the large addiction treatment gap.