Motivational Interviewing with Male Couples to Reduce Substance Use and HIV Risk: Manifestations of Partner Discord and Strategies for Facilitating Dyadic Functioning

Abstract

Background:

The efficacy of Motivational Interviewing (MI) to reduce substance use is well-established; however, its use with couples has met with mixed results. The development of such interventions is particularly relevant for same-sex male couples, as rates of substance use in this population are comparatively high and use is associated with aspects of sexual relationship functioning. One challenge noted in previous research on MI with couples is how to respond to situations where partners disagree with one another or argue against change. Guided by Couples Interdependence Theory, our group conceptualized conflicts within session as failures in the accommodation process. We utilized qualitative analysis to examine manifestations of conflict in session and identify effective provider strategies for responding to conflict.

Methods:

The sample included 14 cis-male couples where at least one partner was aged 18-29; reported substance use; and was HIV-negative. All couples completed three MI sessions lasting 60-75 minutes each.

Results:

Manifestations of conflict included: conflation of thoughts/feelings, vague or indirect communication, and inaccurate assumptions. Effective provider responses included: correcting assumptions, shifting focus, relationship repair, “common ground” reflections, and relationship affirmations.

Conclusions:

Observed conflicts aligned with conceptualizations of destructive resolutions to the accommodation process (i.e., Exit and Neglect). Effective provider responses to conflict facilitated dyadic functioning and catalyzed constructive accommodation. These results provide an initial compendium of provider skills and strategies that may be particularly relevant in work with SMM couples, where achieving accommodation around drug use and sexual health goals is often viewed as a key mechanism of intervention.

Substance use continues to be an important health concern for many sexual minority men (SMM). Gay and bisexual men, in particular, have been found to have high rates of alcohol and drug use relative to their heterosexual counterparts (Medley et al., 2016). The substances most commonly reported by SMM include alcohol, marijuana, cocaine, crystal meth, and other “party drugs” (e.g., ketamine and Gamma-hydroxybutyrate; GHB) (Holloway, 2015; Ristuccia, LoSchiavo, Kapadia, & Halkitis, 2019).

THE INTERSECTION OF SUBSTANCE USE AND SEXUAL BEHAVIOR IN SMM COUPLES

A number of factors coalesce to make the occurrence of substance use among SMM in primary sexual-romantic relationships a matter that warrants the development of tailored intervention strategies incorporating a focus on sex. These include: the epidemiology of HIV infection in the U.S., the co-occurrence of substance use and sexual HIV transmission risk behavior, and an emerging literature linking substance use with relationship functioning among SMM couples. As a result of these convergent factors, mental health providers working with SMM who use substances may encounter the interconnected nature of substance use and sex, particularly when working with partnered SMM.

SMM accounted for 70% of new HIV infections in the U.S. in 2017 (Centers for Disease Control and Prevention, 2018b), with the highest rates of infection observed among men aged 13 to 34. The most common mode of transmission in this population is anal sex occurring in the absence of either condom use or Pre-Exposure Prophylaxis (PrEP) medication to prevent HIV infection (Centers for Disease Control and Prevention, 2018a). The unique vulnerability of SMM in relationships is illustrated in epidemiological studies which estimated that between 35% and 68% of new HIV infections are transmitted between primary – rather than casual – sexual partners (Goodreau et al., 2012; Sullivan, Salazar, Buchbinder, & Sanchez, 2009).

Substance use is associated with increased risk for HIV and other STIs among SMM (e.g., John, Parsons, Rendina, & Grov, 2018; Vosburgh, Mansergh, Sullivan, & Purcell, 2012). The association has been replicated among SMM in relationships (e.g., Mitchell, 2016; Parsons & Starks, 2014) and substance use has also been linked to aspects of sexual relationship functioning. Men in non-monogamous relationships – in which sex with partners outside the relationship is permitted in some way – are more likely to use drugs than those in monogamous relationships (e.g., Brown, Turner, Hern, & Santos, 2017; Mitchell, 2016). In addition, partners in monogamous relationships – in which sex with outside partners is prohibited – tend to report use that is more similar to one another compared to those in non-monogamous relationships (Mitchell, Boyd, McCabe, & Stephenson, 2014; Parsons & Starks, 2014).

MOTIVATIONAL INTERVIEWING (MI): A POTENTIAL VEHICLE FOR TAILORED, BRIEF COUPLES-BASED DRUG USE INTERVENTION WITH SMM

The intersection of substance use and sexual health has led to the emergence of novel dyadic interventions tailored for SMM couples. Couples HIV Testing and Counseling (CHTC) has demonstrated safety, acceptability, and preliminary efficacy among SMM in the U.S. (Stephenson et al., 2011; Sullivan et al., 2014). Subsequently, Starks et al. (2019) found preliminary evidence that the addition of a substance use module (consisting of a review of recent drug use and debriefing questions derived from Motivational Interviewing) and communication skills training videos to CHTC resulted in significant reductions in drug use and related problems. A number of promising interventions which leverage relationship factors to reduce sexual risk and/or drug use among SMM couples have shown promise in recent years (e.g., Newcomb et al., 2017; Whitton, Weitbrecht, Kyryluk, & Hutsell, 2016; Wu et al., 2011),

There are several factors that indicate the value of expanding research related to the application of MI with SMM couples. Of particular relevance, brief MI interventions that integrate foci on sexual behavior and substance use have shown efficacy in reducing substance use and sexual HIV transmission risk behavior among SMM (Chen et al., 2016; Kahler et al., 2018; Parsons, Lelutiu-Weinberger, Botsko, & Golub, 2014; Wang et al., 2018). Although promising, all of these interventions are individually delivered. Secondary analyses of data from one of these studies (Parsons et al., 2014) suggested that participants in relationships at the time they received the MI-based intervention benefitted less than their single counterparts with respect to reducing drug use and CAS under the influence of substances (Starks et al., 2018). Starks et al. (2019)’s pilot research suggested that brief MI can be utilized within the context of CHTC to achieve reductions in drug use; however, that intervention is highly structured and intended for delivery by HIV testers who do not engage in routine substance use intervention. The robust application of MI with couples would be enhanced by the extension and application of existing MI theory and principles.

One of the challenges to implementing MI with couples is that the core concepts and strategies of MI have generally been articulated within the framework of individual delivery. MI is a collaborative process between a provider and a client, aimed at exploring and resolving ambivalence towards change. It is also directive in that it involves a focus on a target behavior (Miller & Rollnick, 2013). Interview content can be organized around four processes. Engagement is characterized by the establishment of a therapeutic alliance in which the client can discuss sensitive material in a non-judgmental atmosphere. Focusing involves the negotiation of a session goal shared by both client and provider. Evoking involves eliciting the client’s perspective on the target behavior – which is the session focus – and the subsequent strategic amplification of client speech in favor of change. Finally, Planning is characterized by the identification of specific behavioral goals and related action steps to achieve them. Across these processes, providers use a set of core skills to establish rapport and draw out change talk: Open-ended questions, Affirmations, Reflections, and Summary statements (OARS).

These processes and skills can be adapted for dyadic delivery, however existing studies that have examined couples-based MI have largely viewed the couple as comprising an “identified patient” and a partner (spouse or significant other) whose involvement serves as an adjunct to treatment (Manuel, Houck, & Moyers, 2012; Monti et al., 2014). There is some outcome data indicating that the involvement of significant others in MI-based substance use interventions may yield improved outcomes relative to individually-delivered MI (Monti et al., 2014). Challenges in this model arise when the significant other focuses the conversation on themselves or argues against change (Manuel et al., 2012). The lack of consensus about how to respond effectively to these situations has led to the view that some couples may be able to be helped by MI whereas for others it is contra-indicated due to differing interests in change between partners (Miller & Rose, 2010).

Our research team (Starks et al., 2018) recently proposed a framework for conducting MI with couples and illustrated its general principles in session transcripts conducted with SMM couples focused on the target behaviors of substance use and HIV prevention. Our approach deviated from most previous efforts to conduct MI with couples in that it positions the couple as the unit of treatment rather than thinking of partner participation as an adjunct to the treatment of one identified patient. In doing so, we drew upon processes defined within Couples Interdependence Theory (CIT; Kelley & Thibaut, 1978; Rusbult, Verette, Whitney, Slovik, & Lipkus, 1991). In relationships characterized by high levels of satisfaction, commitment, and cohesion, partners are more willing to set aside self-interests during moments of conflict or disagreement and consider the impact of their actions on their partner and the relationship as a whole. This transformation of motivation results in the creation of joint goals – termed accommodation (Rusbult et al., 1991; Yovetich & Rusbult, 1994). Building upon these ideas, we acknowledged the potential for “interpersonal ambivalence” towards change (i.e., instances where partners view change differently). We suggested that, in addition to the four processes articulated in existing MI theory, providers need to engage in a process of facilitating dyadic functioning in order to support couples in the resolution of interpersonal ambivalence and the formation of joint goals before proceeding to change planning.

STRATEGIES FOR FACILITATING DYADIC FUNCTIONING

Although Starks et al. (2018) proposed a general framework for approaching MI with couples, we provided relatively little guidance for specific techniques providers might use to facilitate dyadic functioning – and diffuse potential conflicts – in session. In order to utilize the theoretical framework, MI providers need to be able to recognize the emergence of conflict in session – which might signal deficits in dyadic functioning or interpersonal ambivalence towards change – and they need specific linguistic strategies that are likely to be effective in responding to these types of conflict. The purpose of the current study was therefore to provide an initial examination of this technical content and better elucidate the repertoire of MI skills that providers may utilize in session with couples.

We applied the principles of CIT to conceptualize discord – breaches in the relationship between partners in session – as an outcome of the accommodation process. Per CIT, partners’ responses to disagreement can be categorized along two dimensions: (1) influence on the problem at hand (active versus passive) and (2) impact on the relationship (constructive versus destructive). These dimensions yield four potential responses: Voice (active and constructive, e.g., suggesting solutions, discussing the problem, changing one’s own behavior); Loyalty (passive and constructive, e.g., waiting for conditions to improve, supporting one’s partner despite criticism); Exit (active and destructive, e.g., screaming at one’s partner, terminating the relationship); and Neglect (passive and destructive, e.g., ignoring one’s partner, ignoring the problem, letting the relationship fall apart; (Rusbult et al., 1991; Rusbult, Zembrodt, & Gunn, 1982). Successful accommodation occurs when an individual inhibits impulses to respond to dissatisfying responses from a partner by retaliating with destructive behavior (Exit, Neglect) and instead responds constructively (Voice, Loyalty; Rusbult et al., 1991). The following explores how effective provider responses to conflict could highlight couples’ strengths, mitigate the emotional harm inflicted during conflict, and facilitate successful accommodation.

METHODS

Participants

The session content analyzed below was gathered during the formative stage of the Couples Health Project (CHP) between January, 2014 and October, 2015. The overall goal of this stage was to develop a prototype manual for the delivery of MI with SMM couples and to evaluate the feasibility of conducting a subsequent pilot efficacy trial. To this end, the first author developed a prototype manual comprised of three sessions, each lasting 60-75 minutes.

• Session 1 opens with an emphasis on engagement and facilitation. The provider establishes rapport and seeks to draw out the couples’ communication strengths and weaknesses. The provider then introduces the target behavior of substance use and the couple completes a calendar describing their use in the past month. This serves as a starting point to evoke change talk about use. The session ends with the creation of a substance use goal (where possible).

• Session 2 begins with a check-in on goals from the previous week and a review of communication successes and struggles. The couple then completes a card-sort exercise that highlights shared values and joint-goals. Following this activity, providers have the option of focusing on sex within the primary relationship or sex with outside partners. The goal is to examine intersections of substance use and sexual behavior and evoke change talk related to substance use and (where applicable) use of HIV prevention. The session ends with the creation of behavioral goals for the upcoming week.

• Session 3 begins with a check-in on goals from the previous week and a review of communication successes and struggles. The provider has the flexibility to focus on substance use, sex within the primary relationship, or sex with outside partners based upon the couple’s goals. HIV-negative partners are offered the opportunity to complete HIV testing together with a partner. The session ends with a discussion of long-term goals and related planning.

This formative stage of CHP was completed in two steps. First, a group of four couples was recruited meeting the eligibility criteria described below. These couples all completed the prototype CHP intervention. Their participation served to confirm the viability of the prototype manual. Subsequently, a larger sample of 20 couples was recruited. These couples were randomized to receive either the 3-session MI intervention or an attention-matched psychoeducation control condition. Consistent with a focus on identifying technical skills utilized in the delivery of MI to couples, the present analyses utilize data from only those couples who received the MI intervention.

Eligible couples reported a relationship duration of at least three months and recent sexual activity (together or with outside partners). Both partners in each couple were cis-gender male, 18 years of age or older, able to read and speak English, and living in the greater New York City metropolitan area. At least one partner in each couple was between the ages of 18-29, one was HIV-negative, and one reported recent substance use (defined as either one instance of heavy drinking or one instance of illicit drug use in the past 30 days).

The study utilized both online and in-person recruitment strategies. Online recruitment efforts included advertisements on Facebook and other websites frequented by SMM. Study staff also went to bars, nightclubs, and other social events serving SMM. Participants completed a brief online screener that assessed preliminary eligibility and gathered contact information. Study staff then contacted potentially eligible participants and completed a telephone screener with one member of the couple (the index partner).

Eligible couples were asked to schedule a baseline assessment appointment in our research center at a time when both partners could attend. Partners were consented and completed baseline survey instruments separately in separate assessment rooms. Intervention sessions (that produced content used in these analyses) were conducted with both partners together. Session 1 occurred immediately after baseline measures. Sessions 2 and 3 occurred at approximately one-week intervals thereafter. Each partner in the couple received $20 compensation for every intervention session completed. All procedures were approved by the Institutional Review Board of Hunter College of the City University of New York.

For most of the 14 couples (n = 12) who received CHP, the intervention was delivered by a licensed clinical psychologist and member of the Motivational Interviewing Network of Trainers. Two couples were seen by senior clinicians (a licensed Marriage and Family Therapist with more than 10 years of experience delivering MI or a license-eligible clinical psychology post-doctoral fellow) under the supervision of the first author. Intervention fidelity was assessed through review of session recordings by the first author.

Analytic Approach

Among the 14 couples who contributed content to these analyses, one ended their relationship after their second session. The partners in that couple completed Session 3 individually and therefore only contributed two dyadic sessions for analysis. This resulted in a final sample of 41 sessions involving 14 couples available for analysis.

Sessions were transcribed verbatim using an outside transcription service and then checked for quality assurance by study staff. The research team then met to review and discuss transcripts in order to construct a general coding scheme aimed at capturing the ways conflict manifested between partners as well as the ways providers responded to this conflict in ways that promote adaptive dyadic function. Thematic analysis is a technique used identify and analyze patterns of particular phenomenon and involves a deep immersion within the data to examine content and speech patterns contextualized in conversations (Boyatzis, 1998). Thus, this form of qualitative analysis of the data was conducted by a team of coders in Dedoose, an online platform for analyzing qualitative and mixed methods research. Each coding team member conducted one round of coding on the same three intervention sessions before conducting a reliability analysis in order to establish coding protocol familiarity. Thus, thematic analyses required an iterative process of engaging in the emergence of patterns from the narratives provided by the individual within the couple and the therapist (provider), and as such the reported results reflect the themes endorsed by multiple respondents. Codified identifiers for the 14 couples are indicated in the excerpts from the interviews with labels referring to partner one, partner two, and, in some instances the provider. Further, reliability was then measured in Dedoose using an inter-rater reliability test. The coding team established high inter-rater reliability (Cronbach’s α = 0.91).

RESULTS

Descriptive data for the sample are provided in Table 1. The sample was primarily White (60.7%), completed at least a 4-year college degree (67.9%), and self-identified as gay (89.3%). On average, men were 25.9 years old (SD = 4.5 years) and had been in their current relationship for 29.4 months (SD = 22.1). All participants were HIV-negative.

Table 1.

Demographic Characteristics (n= 28)

	n	(%)
Race and Ethnicity
White	17	(60.7)
Non-White	11	(39.3)
Education
Less than a 4 yr degree	9	(32.1)
4 yr degree or more Sexual Identity	19	(67.9)
Gay	25	(89.3)
Bisexual/Queer	3	(10.7)
HIV status
Negative/Unknown	28	(100)
Positive	0	(0)
	M	SD
Age (years)	25.9	4.5
Relationship Length (months)	29.4	22.1

Types of Dyadic Discord

Conflation of thoughts and feelings.

One prominent process that signaled conflict in session was the tendency to verbalize thoughts as emotions. In this scenario, a partner verbalizes his thoughts, beliefs, or hypotheses (about his partner’s behavior or his own) by presenting them as an emotion or feeling. This type of discord is illustrated in the opening statement of the passage below, with emphases added to highlight such instances. Discord becomes further intensified when both partners engage in conflation of thoughts and feelings as the passage continues.

Partner 1: I mean, I feel like you don't really talk about it, your feelings, ever.

Partner 2: I keep mine under control.

Partner 1: Yeah, by not talking about them…

Partner 2: But I speak outside, so if anyone hears me they hear me. He only talks to me about his feelings, and then we try to see how things go from there. But then I guess when it comes to mine, I just keep them to myself. Cause I feel like they're not really important to talk about. They're stupid.

Partner 1: You sit here and talk like that, and it's kind of annoying. It's true. I feel like you should realize that it's kind of dumb that you're saying that… I get why you would feel like your feelings aren't important, but you should acknowledge that they still are.

Partner 2: I acknowledge they are, but just don't feel like they should be.

The conflation of thoughts and feelings inhibits partners’ ability to respond to conflict constructively (i.e., via Voice or Loyalty responses) because it essentially prevents the accurate expression of emotion. Although partners in the above passage use the words “I feel…” several times, neither directly discloses a personal emotion (e.g., frustration, anger, sadness etc.). The inhibition of disclosure potentially restricts the extent to which partners are able to empathize with one another and prevents resolutions to the conflict. The focus on one another’s behavior may also make it particularly likely that a cycle of blaming or defensiveness follows (Christensen, Jacobson, & Babcock, 1995) consistent with an Exit-type resolution (active and destructive) to the conflict (Rusbult et al., 1991; Rusbult et al., 1982).

Vague or indirect communication.

Dyadic discord also arose or was maintained in situations when one partner expressed the belief that the other should be aware of their needs and emotions without them being overtly expressed. In the passage below, Partner 1 begins by describing how he sometimes uses avoidance (e.g., “the silent treatment”) to express his frustration with his partner in moments of conflict. The vague communication here is not embodied in the act of avoidance itself, but rather in Partner 1’s belief that Partner 2 understands the purpose of his avoidance.

Partner 1: Sometimes it’s not even me taking a break. Sometimes it’s me being like, “I'm punishing you right now.” … So it is bad… that’s a bad thing to do.

Partner 2: I don’t think it’s happened often, though…

Partner 1: You haven't noticed?

Partner 2: I haven't, no – for me it’s like, “Okay, whatever. We’re not talking right now. We’re gonna talk later.” For me, it’s like a timeout.

Partner 1: Which is so annoying because I'm like, “I want you to know that I'm doing this on purpose.”

The essential feature of this manifestation of conflict is that one partner inaccurately believes that the other understands the intention or function of communication. It could potentially be thought of as a misalignment of communication norms or expectations between the partners. This source of dyadic discord is similar to the concept of double-bind communication discussed by Bateson, Jackson, Haley, and Weakland (1956). Double-bind is a communication pattern that is composed of vague expressions and conflicting messages that result in the inability to comment or act on the contradictory messages being put forward. It also exemplifies the Neglect approach to accommodation (passive and destructive). In this situation, Partner 1 is indirectly expressing his emotions by ignoring his partner during times of conflict and expressing this injunction through non-verbal means (Bateson et al., 1956). However, his partner is completely unaware of this tactic and therefore is unable to constructively correct the problem at hand.

Inaccurate assumptions of a partner’s intention were the third form of dyadic discord observed, illustrated below when a couple is discussing how often Partner 1 has sex with people outside the relationship. Although the behavior is permitted in their relationship, the topic creates tension for the couple in part because Partner 2’s assumptions about how often Partner 1 has sex with others are inaccurate.

Partner 1: I think it works out. I just think he thinks I'm having more sex than I am. I think his imagination is worse than the reality, but I think actually that's gotten better. I think he realizes I'm not having as much sex as he thinks.

Provider: [to Partner 2] The fun part is, you're here so you can tell us how much sex you think he's having.

Partner 2: I think he has a lot.

Provider: What's a lot of sex in your world? For some people, once a week is a lot, and for some people, a couple times a day is a lot.

Partner 2: What is a lot? I don't know. It depends on the week or the circumstances. I don't know, maybe two or three times a week.

Partner 1: You really think I’m having sex with other people that often?

Partner 2: Uh huh.

Partner 1: When do I have the time to do that?

Partner 2: Any time. You can get out of work any time you want.

Partner 1: I guess so, but with who? That's definitely not happening. Oh, my God.

These inaccurate assumptions lead to the resolution of accommodation through Neglect and inhibit Voice (Rusbult et al., 1991; Rusbult et al., 1982). Partners’ assumptions lead them to believe that discussing the issue will exacerbate conflict; however, avoiding discussion also precludes any opportunity to correct assumptions and formulate a more constructive resolution. This type of conflict also illustrates the power of dyadic intervention. By merely voicing their assumptions within the context of therapy, the couple is able to verify which beliefs about one another are accurate versus inaccurate.

Effective Provider Responses to Conflict

Facilitating the correction of inaccurate assumptions.

When partners express inaccurate assumptions in session, the provider has an opportunity to facilitate their correction in the moment. This potentially allows for greater understanding of each partner’s needs, intentions, and goals, and can thus promote accommodation. In the following passage, Partner 1 describes his behavior while drinking – and discloses his assumptions about how his partner feels about his drinking. The provider utilizes the opportunity to ask Partner 2 whether Partner 1’s assumptions about his feelings are accurate.

Partner 1: I like that he doesn’t take it personal. I hope he doesn’t take it personal. But it’s not like I’m really saying “fuck off,” it’s just that I’m having a good time. And you know, I know that there should always be a limit to how much you’re drinking, but… if I feel happy and I’m having fun, for somebody to come up and be like, “Oh, you shouldn’t be drinking” then I’m just kind of like, “Well, fuck you. You don’t want me to have fun? [Get the] fuck out of my face.”

Provider: You had mentioned hoping that [Partner 2] doesn’t take it personally, which begs the opportunity to ask [Partner 2], “How do you take it when that sort of thing happens?

Partner 2: To be honest, I don’t know. I feel like when you get fucked up, whatever you do is on you because you’ve got to deal with it at the end of the day. So, you can’t tell anybody anything when they’re fucked. So I guess you just let them be and live and learn.

Shifting focus.

Providers also mitigated dyadic discord by shifting the focus of the conversation onto the identification of strategies to constructively handle the problem. This strategy can be understood as the MI provider facilitating engagement in Voice responses (Rusbult et al., 1991; Rusbult et al., 1982) by engaging in direct discussion of partners’ needs, goals and potential solutions to adaptively handling conflict in the future. The following excerpt displays a provider’s use of this strategy in response to discord that has arisen around the amount of sex the couple has and their approach to initiating sex together.

Provider: If you're on the fence about whether you want to have sex, thinking about the fact that you should have sex, is the least successful way to end up having sex.

Partner 1: And I've said that to you before. I don't want you to feel like it [sex] is a chore. When I mention it [having sex] to you… it's not the same, and you feel guilty about it. So it's initiated in a way that's weird. And I don't want there to be any stigma about it.

Provider: What if we flip to … your [Partner 2] initiating, when does it go best? When do you [Partner 2] start [sex] and it works out pleasantly?

Partner 2: I think because I do it less. [laughter]

Partner 1: It’s like, oh wow this is nice.

Provider: So what are the times that you're [Partner 2] most likely to initiate sex between the two of you?

Partner 2: Well, part of the problem is I'm mostly free to do that when both of us have finished with this bucket of things that we have to be doing. Sometimes that time comes never for either or both of us and so that is part of the problem I think.

Provider: So … prioritizing, making time for sexual intimacy in the universe of responsibilities is actually -- it's hard to rank in there.

Relationship repair.

Not all sources of dyadic discord can be remedied through instrumental action. When confronted with situations that required emotion-focused coping (Carver, Scheier, & Weintraub, 1989), providers may make more effective use of strategies which explore how the couple could repair the relationship and move forward. This provider strategy (illustrated in examples below) has the potential to elicit Loyalty (passive and constructive) responses to conflict wherein partners affirm their relationship to one another even in the face of challenges.

Provider: You guys, we have rattled off half a dozen things that could’ve broken a couple up. … And yet somehow you’re still sitting here… What has worked? What is the secret? Because in some ways, you’ve made it past any number of things that other couples wouldn’t.

Provider: My sense is that the two of you have found a way to feel really committed and really connected to one another, despite the fact that sometimes you make one another crazy. [To Partner 1] We’ve talked a lot about the things that [Partner 2] does for you or the ways that he has helped out because this has been kind of a tumultuous year and because finances are hard right now. [To Partner 2] I'm curious, what does [Partner 1] do for you? What keeps you in this relationship? What do you get out of this?

“Common ground” reflections.

Reflections are a core MI skill (Miller & Rollnick, 2013). They are typically used to demonstrate provider attention and empathy. They also encourage a client to expand upon particular content. Analyses revealed that our providers frequently used reflections to emphasize areas of common ground between partners. These reflections served to reframe discord by highlighting areas of partner agreement, shared values, and similarity. This strategy is similar to other techniques in couples therapy that are targeted towards the development of empathy between partners (Christensen et al., 1995). The statements below are examples of reflections:

Provider: What both of you are trying to figure out is: what’s productive. In all of these moments when things get tense, what’s productive? What’s useful? Neither of you wants to hurt the other, and neither of you enjoys a conversation where you feel like you’ve insulted one another, or you have to come back and apologize afterwards.

Provider: The process that both of you need sounds very similar, which is: when you're annoyed, you need a minute to be annoyed and then put it on the table why you're annoyed, and then you can move past that. The challenge is that that requires both of you to give each other that minute to just vent …if one of you moves to fix it too soon, that actually becomes more annoying.

Dyadic affirmations.

Affirmations are another core MI skill (Miller & Rollnick, 2013). In individual MI, affirmations are generally used to acknowledge personal strengths or characteristics of the client that have the potential to facilitate change. In these couples MI sessions, providers made strategic use of affirmations to highlight strengths in dyadic functioning. Similar to the Relationship Repair techniques, this strategy served to mitigate discord by emphasizing aspects of dyadic functioning aspects associated with successful accommodation (Rusbult et al., 1991). The following utterances are examples of dyadic affirmations utilized in session.

Provider: What I’m hearing also from both of you is how – whatever the discussion is, whatever the issue that has to be resolved -- you’re both constantly thinking about how can we talk about this so that I don’t hurt his feelings; or how can we arrive at a conclusion where we’re both okay, and we don’t hurt each other’s feelings any more than we already are, or than we have, without meaning to.

Provider: A lot of couples struggle to balance sharing a lot or having a lot in common, with having some sense of independence and a separate identity as well. And whatever the challenges might be in your relationship, you guys seem to have sorted that out. You both have a degree of independence that you’re very comfortable with, and you also really do a lot for each other and have an identity as a couple together looking forward.

DISCUSSION

These findings provide an initial look at the nature of potential conflicts providers may confront during the delivery of MI to couples, and a starting point for developing a lexicon of technical skills providers may utilize in responding to these conflicts in session. Although these findings are most directly relevant to discussions of substance use and sexual risk with SMM couples, there is at least some potential that the processes and techniques observed here may generalize to MI with other types of couples and target behaviors. Consistent with the general principles outlined by Starks et al. (2018), these findings affirm the potential for CIT to provide a theoretical bridge between the existing MI literature and the delivery of MI to couples.

The formation of joint goals by SMM couples has generally been viewed as supportive of health behavior. Interventions such as CHTC view the development of a sexual agreement and a shared plan to manage HIV risk as facilitating a transformation of motivation wherein partners’ commitment to adhere to the agreement is enhanced by their attention to couple-level consequences (Stephenson et al., 2011; Sullivan et al., 2014). This assumption is also reflected in research on substance use in SMM couples. Lewis, Galdstone, Schmal, and Darbes (2006) found that partnered SMM use a variety of social control strategies to mitigate their partner’s substance use in a manner consistent with CIT, and Starks et al. (2019) found that a discussion of shared drug use goals was associated with decreased use and related problems.

It would therefore be reasonable that providers working with SMM on issues related to sexual behavior or substance use may try to achieve consensus between partners around intervention goals as a way to facilitate behavior change. In fact, Starks et al. (2018) suggested that accommodation around change goals was a necessary prerequisite to successful change planning with a couple. However, CIT would suggest that the mechanisms by which such joint goals influence health behavior are predicated on them arising from constructive resolutions to the accommodation process (Rusbult et al., 1991). Providers therefore need to be able to recognize manifestations of Exit and Neglect in order to avoid inadvertently validating forms of accommodation that are unlikely to support successful behavior change. Destructive accommodation should serve as a signal to providers that they need to utilize MI skills to shift focus away from unproductive blaming, clarify thoughts and feelings, establish effective communication norms between partners, and correct inaccurate assumptions. Recent intervention development efforts with SMM couples have incorporated relationship skill building components (Newcomb et al., 2017; Starks et al., 2019; Whitton et al., 2016; Wu et al., 2011). Future research should examine the potential for the content of these interventions to inform the identification and development of generalized provider strategies for facilitating accommodation.

Notably, most intervention research involving SMM has focused on sexual agreements related to sex with outside partners. CHTC provides a succinct example of this (Stephenson et al., 2011; Sullivan et al., 2014). While this fits with an emphasis on HIV prevention, it potentially overlooks the importance of sex occurring between main partners. There are far fewer models of how providers might engage with couples wherein partners have discrepant views about the quality, amount, or nature of the sex they have together. Improving couples’ sexual functioning has largely been the purview of specialized sex therapies, with a focus on sexual dysfunction in heterosexual couples (Binik & Meana, 2009). The provider strategies illustrated here, and the principles outlined by Starks et al. (2018), may provide a starting point for integrating attention to joint sexual behavior goals within and outside of relationships in a manner consistent with calls for a re-imagining of sex therapy (Binik & Meana, 2009).

The present findings go beyond the previous work of Starks et al. (2018) and provide an initial overview of specific MI skills for responding strategically to conflict or destructive accommodation processes. Several of these strategies can be viewed as dyadic extensions of traditional core MI skills (e.g., OARS). These utterances drew upon information provided by individual partners and make a statement about the couple as a whole. The statement itself was most often directed to both partners rather than to the partners individually. This ability to speak to the “higher-order unit” (i.e. the couple) may require some practice for providers who are more accustomed to individual intervention. Meanwhile, shifting focus, correcting inaccurate assumptions, and relationship repair generally involved multiple provider utterances. Providers utilized a mixture of utterances (e.g., OARS) directed at individual partners and the couple in order to clarify each individual’s perspective and the position of the couple as a whole. As such, these techniques may be thought of as relatively more advanced strategies that required a broader knowledge of relationship functioning and the flexible use of a number of foundational MI skills. While these strategies are grounded in established MI practice, providers wishing to utilize MI with couples initially may benefit from continuing education and focused supervision in order to acquire knowledge about dyadic functioning and hone these skills relevant to couples’ practice.

These results must be viewed in the context of several limitations. First, session content was limited by the brief duration of the intervention (three sessions) and the focus on substance use and sexual behavior. Future research should examine how variability in intervention duration and target behavior influences the manifestation of conflict and associated therapist skills. Second, the requirements of dyadic participation may lead to the inclusion of couples with relatively better dyadic functioning (Hoff & Beougher, 2010; Starks, Millar, & Parsons, 2015; Yucel & Gassanov, 2010). It is plausible that manifestations of partner discord and effective provider skills may vary across the spectrum of dyadic functioning. Third, the sample was limited to SMM who lived in an urban area, all participants were HIV-negative, and at least one participant was 18 to 29 years old. Finally, the providers who delivered the MI intervention were highly experienced and the majority of sessions were delivered by the first author. These results therefore cannot explore provider-level factors associated with intervention effectiveness.

Despite these limitations, these findings provide insight into the technical aspects of conducting MI with couples. They re-affirm the utility of CIT to link existing MI practice to application with couples. Common manifestations of conflict observed in sessions aligned with CIT conceptualizations of destructive accommodation. Observed effective MI strategies generally served to facilitate dyadic functioning and catalyze accommodation in a manner also consistent with the assumptions of CIT. Although the application of MI with couples involves some specialized skills, these skills build upon expertise widely shared by MI providers. Collectively these findings point to the potential that continued research may result in substance use and/or sexual risk intervention strategies which are highly scalable and capitalize on existing intervention capacity to expand services to couples.

Clinical Impact Statement.

Question: This paper utilized qualitative methods to identify manifestations of conflict, or discord, between relationship partners receiving MI and effective MI provider responses.

Findings: Consistent with hypotheses, manifestations of discord could be understood as destructive resolutions to the accommodation process as conceptualized within Couples Interdependence Theory. Effective provider responses facilitated dyadic functioning generally and catalyzed constructive accommodation.

Meaning: Couples-based MI may represent a viable intervention strategy to reduce drug use among sexual minority couples.

Next Steps: Future studies should examine the efficacy of MI with a broader range of couples, target behaviors, and intervention providers in order to further evaluate intervention efficacy and relevant constituent processes.