The Efficacy of the Hold Me Tight Relationship Education Program Delivered by Paraprofessionals in South Africa

ABSTRACT

While there is a growing body of evidence demonstrating the efficacy of Hold Me Tight (HMT) as a relationship education intervention for couples, more studies are needed to test its efficacy in under‐resourced communities. In this mixed‐methods study, we explored the efficacy of HMT when delivered by paraprofessionals in South Africa. After training paraprofessionals, we assessed the efficacy of their delivery of HMT to individuals who were a part of the couple dyads participating in six HMT groups. When comparing pre‐test to posttest outcomes, participants showed significant improvements in emotional control, relationship trust, relationship satisfaction, attachment anxiety, attachment avoidance, and psychological distress; using a Bonferroni correction, emotional control and relationship satisfaction remained significant. Nine participants completed qualitative interviews and the findings indicated that participants had a positive HMT experience and it enabled them to focus on their relationship in new ways. Some cultural and gender differences and concerns were identified.

1. Introduction

Emotionally Focused Therapy (EFT; Johnson, 2019) is an evidence‐based treatment for couple relationship distress that produces robust and lasting effects (Spengler et al. 2024). Developed more recently, Hold Me Tight (HMT; Johnson 2008) is a couple relationship education program based on EFT principles that has been evaluated in several studies to date, often with HMT facilitators who are master's‐level mental health professionals. These studies include an examination of the efficacy of non‐distressed couples in the United States and Canada (Kennedy et al. 2019), an adaptation of HMT in a 1‐day format (Morgis et al. 2019), and a test of HMT with self‐referred and clinician‐referred couples (Conradi et al. 2018). In addition, HMT's efficacy has been explored in culturally diverse contexts such as a Chinese‐language version of HMT for Canadian‐Chinese couples (Wong et al. 2018), Iranian married women diagnosed with breast cancer (Hedayati et al. 2023), couples in the Netherlands facing Huntington's Disease (Petzke et al. 2022), and married students in Iran (Borjali and Sadr 2021). In one of the earliest HMT studies, Lesch et al. (2018) explored the use of HMT with Black South African couples. In the Lesch et al. study, HMT was facilitated by mental health professionals and certified EFT therapists, and participants were interviewed before and after the intervention to explore their experiences of participating in HMT. Findings revealed that participants reported largely positive experiences and enhanced relationship quality. The authors suggested the need to train local, lay individuals to conduct HMT within their communities (e.g., individuals who share participant demographics) instead of relying on mental health professionals from the outside. They suggested that the training of lay, local professionals, if successful, could augment the dissemination of HMT, while preventing a power differential or hierarchy between facilitators and participants (Lesch et al. 2018).

Facilitators of relationship education programs are typically not therapists, but rather individuals specifically trained to deliver these programs (Markman et al. 2022). Interventions delivered by paraprofessionals, individuals who possess some degree of skill and training, but who lack professional certification such as a master's degree or mental health license, may be especially important in the dissemination of effective treatments in under‐resourced contexts, especially as they are likely to be able to relate to those who live in similar contexts (Lesch et al. 2018). Relying on paraprofessionals can be useful in under‐resourced contexts where access to care is limited. Paraprofessionals are involved globally in providing interventions for a wide range of mental health difficulties (e.g., Diebold et al. 2020; Mehta et al. 2019; Reddy et al. 2021). More specifically, these paraprofessionals have been widely used in under‐resourced settings around the world to help with a range of health‐related and mental‐health issues (e.g., Cappella and Godfrey 2019; Linsk et al. 2010; Muriuki and Moss 2016; Wiggs et al. 2021). In this study, we explored the efficacy of HMT delivered by paraprofessionals, to couples living in under‐resourced communities in South Africa.

HMT, as well as other relationship education interventions, are effective for a wide range of problems including improvements in mental health/psychological distress as well as relationship distress (Doss et al. 2022; Kennedy et al. 2019; Markman et al. 2022). Studies of HMT include a focus on outcomes such as relationship distress, attachment anxiety, attachment avoidance, relationship trust, and mental health difficulties (e.g., Conradi et al. 2018; Kennedy et al. 2019; Markman et al. 2022). Lesch et al. (2018) report that while no comprehensive studies of relationship distress in South Africa exist, a range of studies indicate a high rate of relationship distress in South Africa. There are also elevated levels of common mental health difficulties including depression, substance abuse, and anxiety. Lesch et al. (2018) further report that the divorce rate has increased in South Africa, although highly accurate data are not available. It is evident that a well‐studied relationship education program like HMT could be highly valuable in a South African setting, especially in helping with both mental health and relationship problems. Only one study (Lesch et al. 2018) has examined the implementation of HMT in Africa/South Africa; the mental health professionals who served as facilitators in this study did not reside in the local community where HMT was delivered.

1.1. The Current Study

This was the first study to examine HMT delivered by paraprofessionals in participants' local communities. Specifically, in this study, we tested the efficacy of HMT delivered by paraprofessionals to couples in South Africa. Given the scarcity of master's level mental health professionals on the African continent, it is important to evaluate ways in which interventions can be delivered by paraprofessionals. Given that South Africans suffer many health and mental health difficulties (e.g., HIV/AIDS, tuberculosis, depression, trauma, substance abuse), and that large segments of the population live in severely under‐resourced communities, an intervention with the potential to both strengthen relationships and prevent mental health problems is desperately needed.

Paraprofessional participants received training in HMT over 2 weekends. During the first weekend, they participated with a partner in the HMT intervention itself. Our training team consisted of researchers and interventionists from South Africa and the USA, and our goal was to train religious and community leaders (i.e., paraprofessionals) to deliver HMT in their local communities. The first weekend occurred in the Western Cape region of South Africa, and was designed to give potential facilitators the experience of receiving HMT themselves. All but one paraprofessional attended with a spouse/partner and one attended with a sister. Our goal was to teach HMT to the paraprofessionals by immersing them fully in the HMT curriculum as a participant first. Two mental health professionals who were experienced in delivering the HMT intervention were the lead trainers for this event, and they delivered the HMT training as they typically would to any group of participants. The trainers were White, South African women who spoke both English and Afrikaans. Since the paraprofessionals did not have formal training in the facilitation of psychosocial groups such as HMT, aside from experiences working with groups in their employment or religious settings, the second weekend was spent teaching the paraprofessionals how to facilitate a HMT group including recruiting and screening potential participants, orienting participants to the program, teaching the HMT content, dealing with challenging group dynamics, and seeking support and supervision. This part of the training was facilitated by a team of seven experts, including one of the facilitators who delivered HMT to the paraprofessionals on weekend one, another locally respected EFT/HMT therapist and scholar, and five clinicians and scholars from the USA, two of whom grew up in South Africa and were familiar with the local culture. The trained paraprofessionals went on to offer HMT in their local communities following the curriculum and processes they learned over the course of these 2 weekends. We provided support and supervision to them as they went through the process of running their groups. See Blow, Timm, Lesch, Wittenborn, de Bruin, Anderson, White VanBoxel, & Tseng (in press) for the full training protocol.

This study explored the efficacy of paraprofessionals' delivery of a relationship education program in an under‐resourced setting and addressed the following research questions:

• 1.Did participants benefit from receiving the HMT intervention delivered by paraprofessionals? More specifically, our quantitative exploration tested the following hypothesis:

  Members of groups conducted by paraprofessional facilitators will demonstrate significant improvements at posttest on relationship satisfaction, relationship trust, emotional control, attachment anxiety, attachment avoidance, and psychological distress compared to pre‐test.

• 2.What were the experiences of group members receiving HMT from paraprofessionals, and were any cultural challenges encountered? This question was answered using qualitative methods.

2. Methods

2.1. Design and Procedures

This is a mixed‐methods study that describes our evaluation of the efficacy of HMT delivered by paraprofessionals to couples in South Africa using quantitative assessments, followed by qualitative interviews of a subsample of couples exploring their experiences with HMT in groups run by paraprofessionals. Quantitative data assessed outcomes pre‐ and post‐ intervention on a number of variables including relationship satisfaction, relationship trust, emotional control, attachment anxiety, attachment avoidance, and psychological distress.

The study was approved by IRBs in both the USA and South Africa. Participants received a R125 (about $8 USD) shopping voucher for each survey completed, and R300 (about $16 USD) a couple for completion of a qualitative interview. For the qualitative portion of this study, a subset of group members volunteered to be interviewed for 1 h. The purpose of the interviews was to gain a deeper understanding of the experiences of group members in the HMT groups run by paraprofessional facilitators, along with any cultural issues that arose in their groups.

2.2. Participants

A total of 34 couples who were members of six HMT groups began the program that was set up and facilitated by paraprofessionals; some dropped out or did not complete the study assessments resulting in a final data set of 61 individuals. Group members were recruited by the paraprofessionals who provided the research team with a list of participants before their HMT group commencement. The majority of HMT participants were contacted by the study team via email with survey completion instructions which they were able to complete online. Some participants who lived in areas with poor internet access completed their surveys via paper and pencil and those were uploaded into the Qualtrics database by the study team. Participants completed the posttest within 2 weeks after the facilitator notified the team that their HMT training was complete. Nine group members were interviewed using the Zoom video‐conferencing platform. These were volunteers who indicated that they were willing to be interviewed. We attempted to interview dyads, although some spouses were unwilling to take part or were unavailable. The first and fourth authors conducted all the interviews. Interviews were transcribed and all identifying information was removed from the transcriptions. Figure 1 details the enrollment of individuals into the study.

Figure 1.

Flowchart of the HMT intervention using paraprofessionals.

2.3. The Intervention

HMT is a relationship education intervention focused on strengthening couple relationships. While different formats have been developed, this study was based on the framework described in Johnson (2008). In this format, HMT is delivered in a couple group format with each group consisting of approximately six couples. HMT is delivered in eight sessions with the following topics: (1) love and attachment, (2) the demon dialogues, (3) finding the raw spots, (4) revisiting a rocky moment, (5) HMT conversation, (6) forgiving injuries and trusting again, (7) tender touch and synchrony sex, and (8) keeping love alive and caring for your relationship (Johnson 2008). All paraprofessionals we trained personally participated in HMT over 1 weekend and then received education on how to deliver HMT over a second weekend. To increase treatment fidelity, all paraprofessionals who ran their own groups offered the same HMT curriculum following the provided HMT treatment manual, which we translated into Afrikaans for those more comfortable with this language. Support with facilitating HMT was provided to paraprofessionals by members of the HMT study team who were available on WhatsApp and via phone. On one occasion, group facilitators met in a group format to provide and obtain peer support and tips for running a group by sharing their experiences, successes, and challenges. Those who successfully implemented their own groups delivered the program in the 8‐week format, with the exception of one group that was delivered over the course of 1 weekend.

2.4. Measures

Participants completed measures before and after their HMT groups. This study was conducted in South Africa and over 50% of participants reported Afrikaans as their primary language. As a result, we translated all assessment measures into Afrikaans. The translations were completed by members of the study team with proficiency in both English and Afrikaans. In the survey we gathered demographic information (e.g., gender, age, race/ethnicity, religion, yearly income, marital status, number of children, home language, employment status, and highest education) and the following standardized measures.

2.4.1. Relationship Satisfaction

To assess relationship satisfaction, participants completed the Couple Satisfaction Index (CSI‐16; Funk and Rogge 2007) before and after the HMT intervention. The CSI‐16 asks questions about their relationship, including their perceptions of relationship quality, happiness, comfort, enjoyment, cooperation, and overall feelings.

We used the sum of responses, with higher scores indicating higher relationship satisfaction. The CSI has good internal consistency, with Cronbach's alphas ranging from 0.84 to 0.99 in a prior study (Funk and Rogge 2007), and 0.68 (pre) and 0.74 (post) in this sample.

2.4.2. Relationship Trust

The Relationship Trust Scale (RTS; (Holmes et al. 1990) was used to assess changes in couples' levels of trust in their relationship before and after the HMT intervention. The RTS is a 31‐item measure of interpersonal trust in intimate relationships. We used the total score for this measure to indicate trust in the relationship, with higher scores indicating a stronger level of trust between partners. In this study, the alpha was 0.89 (pre) and 0.94 (post).

2.4.3. Emotional Control

The Courtauld Emotional Control Scale‐Revised (Feeney 1995; Watson and Greer 1983) was used to assess emotions in relationships and how participants may react to their partner when they are feeling a particular emotion (e.g., angry, anxious, sad). This scale was assessed both before and after the HMT intervention. This 21‐item measure has three subscales, and each subscale has seven questions. We used the sum of the responses, with higher scores indicating greater emotional control. In this study, alphas were 0.88 (pre) and 0.93 (post).

2.4.4. Attachment Anxiety and Avoidance

Attachment anxiety and avoidance were assessed with the 12‐item short version of the Experiences in Close Relationships (Fraley et al. 2000; Wei et al. 2007). This measure includes two dimensions of attachment—avoidance and anxiety. Higher scores indicated higher levels of attachment avoidance and attachment anxiety. The scale has good internal consistency with the alpha scores for this sample at 0.67 (pre) and 0.66 (post).

2.4.5. Psychological Distress

The Brief Symptom Inventory (BSI‐18; Derogatis 2001) was used to assess psychological distress before and after HMT. The BSI‐18 is an 18‐item questionnaire that is used as a brief screening of psychological symptoms that may be of concern to an individual. Participants were asked to respond to each item in reference to how much a specific problem distressed or bothered them during the prior 7 days. They responded on a Likert scale ranging from not at all to extremely. Scores across all items were summed with higher scores indicating more psychological distress. Alphas for the BSI‐18 in this study were 0.85 (pre) and 0.88 (post).

2.5. Data Analysis

2.5.1. Quantitative Data Analysis

All quantitative analyses were conducted in SPSS version 29 (IBM Corp. Released 2023). Before the primary analyses, preliminary checks were performed to ensure that the data met the assumptions for the analyses we planned, including normality, linearity, and homoscedasticity. Descriptive statistics and frequency distributions were used to examine demographic characteristics of participants. Finally, we conducted two‐tailed paired‐samples t‐tests to assess differences between pre‐ and posttest scores on relationship satisfaction, relationship trust, emotional control, attachment anxiety, attachment avoidance, and psychological distress in the study sample. Thirty‐five participants completed both pre‐ and post‐tests, and these 35 assessments were used for the paired‐samples t‐tests. Although each paired‐samples t‐test examined a distinct outcome, conducting multiple t‐tests within the same data set increases the risk of Type I error (Tabachnick and Fidell 2019). To account for this error, we applied a Bonferroni correction (Armstrong 2014) to adjust the threshold for statistical significance.

2.5.2. Qualitative Data Analysis

Thematic analysis (Braun and Clarke 2006; Braun and Clarke 2021) was used to understand participants' experiences of participating in the groups. Thematic analysis involves six steps: (1) familiarization with the data, (2) generating initial codes, (3) searching for and generating themes, (4) reviewing themes, (5) defining and naming themes, and (6) writing the report (Braun and Clarke 2006). The first author led the data analysis. He was the interviewer in the majority of the interviews, and he read each transcript twice before beginning coding. Coding was guided by each research question and transcripts were read with two questions in mind, (1) what were the experiences of group members in the groups run by paraprofessionals? and (2) did any cultural challenges arise? In reviewing each transcript, the lead author highlighted and labeled lines and responses from participants related to each research question. Next, each of these passages were reread at least two times and larger themes were created based on these responses. This approach allowed for a deeper understanding of participant viewpoints on each question asked. Main themes were generated based on this review, and then the fourth author provided a check on the themes, and a discussion was held where there was disagreement between the two coders; the discussion continued until there was agreement. In analyzing the interviews, all voices were considered relevant and particular attention was paid to ideas that were different from the larger group, or which were communicated with intensity of emotions. This was particularly relevant to themes related to gender and sexuality.

Lincoln and Guba's (1985) trustworthiness criteria of credibility is important when it comes to the rigor of a qualitative study. Regarding credibility, it was important to bracket the experiences of the lead analyst in this study as we interpreted the qualitative findings. The lead author was born in South Africa and lived there for 27 years. He worked as a church minister in Cape Town for 7 years before moving to the USA to further his education. It is possible that he overidentified with the study participants. At the same time, he was able to connect with and had a shared understanding of the challenges described by those in the study. Having a second coder not from South Africa (the fourth author) review the themes added credibility to the findings.

3. Results

3.1. Demographics for HMT Group Participants

Table 1 summarizes the demographic information of participants who engaged in the HMT groups, and the demographics of those who completed both pre‐ and posttest surveys. In total, 61 people who participated in groups facilitated by paraprofessionals completed some study assessments, while 35 people completed both surveys and were included in the final analysis. Participants identified as women (51%) and men (49%). Approximately 85% were married. Participants were all South African and, according to South African classifications, identified as: Black (6%), Colored (57%), and White (37%). Regarding income, 66% reported falling in the lower to middle class income group. Approximately 91% of participants had at least one child. Participants described themselves as Christian (94%) or Muslim (3%), while 3% did not disclose their religion. Regarding employment, 86% reported employment in one job, 26% reported working two or more jobs, and 14% reported no employment. In terms of education, 74% earned at least a high school diploma.

Table 1.

Sample characteristics of group members.

Sample characteristics	Group members (N = 61)	Group members in t‐test (N = 35)
Mean age in years	40	41
Gender %
Men	44%	49%
Women	56%	51%
Marital status %
Married	74%	85%
Unmarried	5%	6%
Long‐term committed relationship	2%	9%
Other	7%	0%
Highest education level %
No high school completion	30%	26%
Grade 12/Std10/NTCIII	21%	23%
Post matric diploma	13%	20%
Bachelor's level degree	3%	6%
Post graduate diploma	3%	6%
Honors degree	3%	3%
Higher degree (e.g., masters, PhD)	0%	0%
Other	10%	11%
Missing	17%	5%
Self‐identified population group %
Black	18%	6%
Colored	59%	57%
White	23%	37%
Monthly household income %
No income	2%	0%
R1–R6400	32%	32%
R6401–R12,800	21%	20%
R12,801–R25,600	13%	14%
R25,601–R51200	16%	26%
R51,201–R102,400	3%	6%
R204,801 or more	0%	0%
Missing	13%	2%
Number of children %
0	5%	3%
1	8%	9%
2	26%	42%
3 or more	43%	40%
Religion %
Christian	85%	94%
Muslim	2%	3%
Missing	13%	3%
Employment %
No employment	16%	14%
Employed with one job	61%	60%
Employed with two or more jobs	15%	20%
Retired	3%	6%
Volunteer	2%	0%
Missing	3%	0%

Nine individuals from four couples who participated in HMT were interviewed in qualitative interviews which were conducted in English (one had a translator present to assist with Afrikaans interviewees). All were middle‐aged; two were Black, five were Colored, and two were White. In terms of gender, five identified as women and four as men. Participants were middle to low income, and all were Christian.

3.2. Research Question 1

Paired‐samples t‐tests were conducted to evaluate the impact of HMT on group members' scores on target variables. For emotional control, relationship trust, relationship satisfaction, attachment anxiety, attachment avoidance, and psychological distress, there were significant differences in pre‐ and posttest scores (see Table 2). For emotional control, there was a statistically significant decrease from pre‐ (M = 64.12, SD = 16.31) to posttest (M = 52.46, SD = 14.16), t (25) = 4.22, p < 0.001. The eta squared statistic (0.83) indicated a large effect (Cohen 1988). For relationship trust, there was a statistically significant increase from pre‐ (M = 162.97, SD = 27.67) to posttest (M = 174.29, SD = 25.22), t (33) = −2.60, p < 0.05. The eta squared statistic (0.45) indicated a small to medium effect (Cohen 1988). For relationship satisfaction, there was a statistically significant increase from pre‐ (M = 61.20, SD = 7.28) to posttest (M = 65.91, SD = 7.11), t (34) = −3.89, p < 0.001. The eta squared statistic (0.66) indicated a medium effect (Cohen 1988). For attachment anxiety, there was a statistically significant decrease from pre‐ (M = 23.56, SD = 6.14) to posttest (M = 21.47, SD = 5.95), t (31) = 2.45, p < 0.05. The eta squared statistic (0.43) indicated a small to medium effect (Cohen 1988). For attachment avoidance, there was a statistically significant decrease from pre‐ (M = 16.03, SD = 6.94) to posttest (M = 13.53, SD = 4.70), t (31) = 3.18, p < 0.05. The eta squared statistic (0.56) indicated a medium effect (Cohen 1988). For psychological distress, there was a statistically significant decrease from pre‐ (M = 27.43, SD = 7.74) to posttest (M = 24.31, SD = 6.89), t (34) = 2.14, p < 0.05. The eta squared statistic (0.36) indicated a small effect (Cohen 1988).

Table 2.

Paired‐samples t‐tests.

	Group members (N = 35a)
	Pre‐HMT		Post‐HMT		t	p	d
	M	SD	M	SD
Relationship trust	162.97	27.67	174.29	25.22	−2.60*	< 0.05	−0.45
Emotional control	64.12	16.31	52.46	14.16	4.22***	< 0.001b	0.83
Attachment‐anxiety	23.56	6.14	21.47	5.95	2.45*	< 0.05	0.43
Attachment‐avoidance	16.03	6.94	13.53	4.70	3.18*	< 0.05	0.56
Relationship satisfaction	61.20	7.28	65.91	7.11	−3.89***	< 0.001b	−0.66
Psychologicaldistress	27.43	7.74	24.31	6.89	2.14*	< 0.05	0.36

^a Of the 61 group members who completed part of the HMT assessments, only 35 completed both pre‐ and post‐tests. These 35 participants were included in the paired‐samples t‐tests.

^b To adjust for multiple comparisons, a Bonferroni correction was applied (adjusted α = 0.0083); only Emotional Control and Couple Satisfaction remained significant at the corrected threshold.

^* p < 0.05;

^*** p < 0.001.

3.2.1. Bonferroni Correction

To account for the increased risk of Type I error due to conducting multiple paired‐samples t‐tests, a Bonferroni correction was conducted. With six paired‐samples t‐tests, the adjusted significance threshold was set at p < 0.0083 (0.05/6). After the adjustment, two outcomes remained statistically significant: emotional control (p < 0.001) and relationship satisfaction (p < 0.001). Relationship trust, attachment anxiety, attachment avoidance, and psychological distress did not meet the adjusted threshold, and their results should be interpreted with caution.

3.3. Research Question Two

The qualitative research question posed was, “What were the experiences of group members receiving HMT from paraprofessionals, and were any cultural challenges encountered?” Group members interviewed were all highly positive about their experiences participating in an HMT group. They all gave positive reports about their group facilitators. They particularly liked how vulnerable and authentic their facilitators were, and this meant a lot to them.

NJ (female, Black) stated,

“…she's a flexible facilitator in the sense that, you know how other facilitators can get worked up when things don't go according to plan or stuff. I think she is relaxed in her facilitation and allows things to emerge and stuff like that. That relaxed posture makes her able to be flexible as things change or situations change and stuff like that. And also, I think that just being able for her to be vulnerable as well in terms of sharing her own personal stories. I think that's good. It invites others to also be vulnerable in the space and something that they can share. So that's something I've really appreciated.”

Two main themes emerged from the interviews: first, the HMT format provided an opportunity to focus on their relationship and provided an opportunity to process relationship issues and, for some, growth. Second, for some participants, cultural and gender challenges were encountered.

3.3.1. Opportunity to Focus on Their Relationship

Some couples reported that HMT was a pleasant surprise and that it was helpful for their relationship. One individual (WA, male, Colored) described how he and his wife were “skeptical at first, but as it went on, we learned a lot, and it was fantastic.” Couples reported that relationship issues were easy to avoid in everyday life, but that the HMT format almost “forced” them to focus on their relationships and issues that may have been percolating below the surface. WA (male, Colored) described how they were able to “talk about topics that they had previously been afraid to address.” CM (female, White), a different participant, stated something similar:

“When we had our discussions [at home], we sort of veer completely off the subject from what we were supposed to be talking about… So, that [HMT] was a good experience for us. It made us talk about things that were bugging us… where we wouldn't have if we were at home.”

This same participant (CM, female, White) stated:

“We were going through a little bit of a patch, and it [HMT intervention] was good. It was good that it came at the right time for us. I found the material very intense at the time when we were doing it…In the workshop we were out of the home situation in this space where we were almost forced to think and talk about our relationship, not about what the kids did in school today… and there was none of that, that could interfere, but we were in this space, it was just about us.”

Another participant (NJ, male, Black) talked about how they benefited from the group in that it gave them a space to share personal experiences with partners and other participants:

“[It] created a comfortable space to talk about issues that you hardly talk about, and it sits in your head. You don't get an opportunity to share some things with your spouse [at home] and you debate certain issues by yourself. But that just created a space to open up amongst ourselves as a couple, but also sharing experiences from other couples. So, times [were opened up] where people there were able to share what they were discussing. So, there were opportunities to hear from other couples' experiences, as we were not just in this challenge by ourselves, but other people are going through the same challenges too.”

3.3.2. Cultural and Gender Challenges Encountered

There were very few cultural challenges, but there were concerns that should be noted. Even though these voices were in the minority, they were of a cultural group whose norms may differ from other ethnic groups, and the concerns they raised are important for HMT implementation in this context. There was one couple who were highly tuned in to issues of gender, religion, and sexuality, and they provided a great deal of in‐depth feedback about their specific experiences and concerns, and several of the points below can be attributed to them. While these were minority voices, the points they raised were powerful and should be carefully considered by those setting up these kinds of groups in diverse cultural contexts. There were three main cultural concerns raised by these participants.

3.3.2.1. Cultural Mismatch in Metaphors and Illustrations

One couple described not feeling like the pictures or videos captured the local context. The male in the couple (NJ, Black) stated:

I think the pictures and materials were created in Canada. So, the faces that I got to see was Canadian faces not a reflection of South Africa needs. So that was one area that if one is to explore something like this in Africa, I know that it would try and find some materials that will make African people connect with them. There was also the accent that was used in the videos and some of us, myself and my wife may have understood, but maybe not everyone.”

3.3.2.2. Challenges Related to Sexuality

Topics such as sexuality created discomfort for some or were viewed as potential sources of discomfort, while others found these topics to be helpful. One individual stated that the module on sexuality created a great deal of discomfort in the group they were in (this was a Black group in a church setting with both young and older couples). It was not only the topic of sexuality, but it was also if couples displaying any kind of affection to each other that also created discomfort, especially in front of older people. He (NJ, Black) stated:

“The one thing that was difficult, culturally, [was the] session on sex or intimacy…It was extremely awkward.”

He expanded on the topic of age and awkwardness around sexuality. He (NJ, Black) stated:

“The age differences created more awkwardness. Now having people that would normally be someone's parent in the same room, talking about sex and in our culture, it's just a bit of a no go zone that just creates a bit of a tension, but she [the facilitator] made it somehow, and we did the activity, but it was where we did not stay long. It was touch and go.”

3.3.2.3. Challenges Related to Gender

One female group participant talked about concerns related to gender, especially related to the group facilitator who was a woman. She (NJ, female, Black) said,

“I'm always feeling like, I think that women should talk about it [sex] separately from men and men should have their own conversation about what's happening in the bedroom. It's just always difficult to talk about it when they have other couples, other men in the room.”

One group in a Black religious community was led by a woman. A couple, who were members of this group described how, while they were comfortable with her in the lead, that they had concerns about some other group members. One participant (NJ, female, Black) said:

“Really, really, it's an issue culturally for a woman to stand before men, as a teacher, so to speak, in that role about a sensitive issue about marriage and sex, about all these things. So, I think it really takes a lot to have her sustain that because she's a woman but also, she's a young woman at the same time, right. I really take my hat off to her…”

She (NJ, female, Black) continued,

It comes up in subtle ways. But I think it also has to do with us being Christians and Christianity and gender and most of these men were pastors and now they have to, you know, because she was quite strong. I mean she was taking the lead and her husband was more in the background and there was always this gender dynamic and I think for most of us, there's always this expectation that the men would lead….I don't know how other people experienced it, but I was always aware of gender in that space and the extent to which people open up or decide to, but I think in the end things came together nicely.

4. Discussion

In this study, we explored the efficacy of HMT delivered by paraprofessionals. Regarding group participants' quantitative findings, there were significant positive improvements in the following variables: emotional control, relationship trust, relationship satisfaction, attachment anxiety, attachment avoidance, and psychological distress. Emotional control and relationship satisfaction remained significant after applying a Bonferroni correction. Participants' qualitative responses indicated they experienced HMT positively and it allowed them the opportunity to focus on their relationship in ways that they would not have without the program. There were also some cultural and gender differences which should be considered in the future.

Findings indicate that HMT delivered in a South African context is efficacious, which aligns with prior studies of HMT in other contexts (e.g., Kennedy et al. 2019). The Kennedy et al. study was one of the earliest and largest HMT studies to date (n = 95 couples) and participants showed improvement in relationship satisfaction and relationship trust shortly after completing HMT. In the Kennedy et al. study, participants lost their gains at the 6‐month assessment point, but we were unable to assess this in our study. Their study also did not show significant changes in the attachment variables of avoidance and anxiety. Our findings are similar to those of Conradi et al. (2018) whose couples improved on a number of variables including relationship satisfaction and psychological complaints. The results of our study suggest that paraprofessionals are able to deliver HMT to couples in under‐resourced communities with good outcomes for psychological distress and relationship related variables. The remainder of this discussion is organized around four themes: the changes to relationship variables as a result of HMT, the changes to psychological distress as a result of HMT, the promise of paraprofessionals to alleviate distress in South African communities, and cultural challenges.

4.1. Improvements in Relationship Outcomes

Numerous studies have shown that HMT is an efficacious intervention for improving relationship dynamics, at least in the short term (e.g., Kennedy et al. 2019). Markman et al. (2022) recently reviewed the progress in relationship education research for many types of relationship education programs. Although their review did not include HMT, it does conclude that relationship education programs are effective for a wide range of couples and that these interventions are being utilized with more diverse and underrepresented populations. Improvements in relationship functioning, including relationship trust, satisfaction, and attachment variables following HMT have been shown in previous studies in the United States (Morgis et al. 2019), Canada (Kennedy et al. 2019; Wong et al. 2018), the Netherlands (Conradi et al. 2018), and Iran (Borjali and Sadr 2021). Study participants interviewed reported that changes were possible in relationships, just because of the space provided for them to address their issues while taking part in HMT. These couples suggested that they would not take the time to do this without the formal program and facilitator support.

4.2. Improvements in Psychological Distress

It is important to note that HMT was associated with improved psychological distress. While HMT is not promoted as a mental health intervention, it improved psychological distress symptoms in participants in our study. This is notable for two reasons. First, individuals may be more likely to engage in a relationship‐enhancing intervention than a mental health treatment, and this may be even more important in contexts where mental health treatments are not readily available to individuals such as in under‐resourced settings. Second, interventions that can be delivered by paraprofessionals can increase access to services in under‐resourced regions. In their review, Markman et al. (2022) discuss the promise of relationship education programs as a treatment for mental health noting that the vast majority of studies they included in their reviews showed significant effects for mental health‐related variables.

4.3. The Promise of Paraprofessionals

Our findings suggest that paraprofessionals are able to achieve outcomes that are similar to those achieved in other HMT offerings around the world by mental health professionals. When approached, community leaders were very eager to become trained facilitators, and they were motivated to succeed (Blow et al. in press). HMT programs offered by lay professionals in their local communities can be helpful for both relationship problems and psychological distress.

4.4. Cultural Challenges

Our goal in this study was to deliver the HMT approach as intended by the developer without any significant cultural adaptations other than translating the manual into Afrikaans. In the qualitative interviews, we asked participants about HMT and its fit with their cultural views and values. While many participants interviewed reported a strong cultural match, others raised concerns related to sensitive topics such as sexuality and intimacy, especially discussing these topics in a group format. This also raised some concerns about culturally appropriate approaches to gender roles in program implementation. The topic of sexuality created discomfort in some groups, especially as younger people were talking about sexuality in front of elders. This was only an issue in some cultural groups, while others appeared to be quite comfortable with this format. While topics of intimacy are important in HMT curricula, considerations are important when these are implemented across cultural groups. More studies of how to introduce these topics in different cultures are needed (Bernal et al. 2009).

4.5. Implications

This study has numerous implications for strengthening relationships and improving mental health. Findings demonstrated that HMT delivered by paraprofessionals is efficacious in South Africa. Although our findings were positive for both relationships and mental health outcomes, these findings should be interpreted through the lens of the study limitations, particularly the small sample and the restrictions in the analyses we were able to run due to the small sample. However, our findings suggest that paraprofessionals are able to deliver HMT with positive outcomes for those who participate in their groups.

In the future, it could be useful to extend training opportunities to more paraprofessionals. Many of those who facilitated their groups were motivated to help their communities and were able to run their groups with limited resources, aside from the support and encouragement of the research team. Our study suggests that this approach is likely to be effective in other similar contexts.

4.6. Limitations

There are a number of limitations associated with this study. First, even though a total of 61 group members completed some HMT assessments, only 35 people completed both pre‐ and posttest surveys. As a result, only 35 group members were included in the t‐tests to evaluate pre‐ and posttest changes. While this was not ideal, it did provide us with preliminary evidence of the initial efficacy of HMT in this population. Second, because our study included linked couples and related couple outcomes, it would have been ideal to conduct dyadic analyses at the couple level. Unfortunately, the sample size did not allow for these types of analyses, and this is a study weakness. Third, a limitation of the study is the missing data from pre‐ to posttest. Unfortunately, it is difficult to collect longitudinal data in under‐resourced settings, so findings from this pilot study should be interpreted cautiously. Finally, as with any intervention and assessment in communities where different languages are spoken, measurement issues can occur. While we translated our assessments into Afrikaans, we acknowledge that some meaning could be lost in translation.

5. Conclusion

In this study, we aimed to explore if HMT could be delivered efficaciously by paraprofessionals after a relatively short training period. HMT delivered by paraprofessionals was efficacious for group participants based on quantitative and qualitative findings. The program was delivered at a low cost in under‐resourced communities with positive outcomes on relationship and psychological distress variables. Future research on the cultural adaptation of HMT is needed.