Provider Behaviors that Predict Motivational Statements in Adolescents and Young Adults with HIV: A Study of Clinical Communication Using the Motivational Interviewing Framework

Abstract

Medical care providers’ use of Motivational Interviewing (MI) is linked to improved medication adherence, viral load, and associated behaviors in adolescents and young adults living with HIV. Mastering MI is difficult for busy providers; however, tailoring MI training to the specific MI communication strategies most relevant for HIV treatment context may be a strategy to increase proficiency. The present study aimed to identify communication strategies likely to elicit motivational statements among adolescent-young adult patients living with HIV. Language used by MI-exposed providers during 80 HIV medical clinic visits was transcribed and coded to characterize patient-provider communication within the MI framework. Sequential analysis, an approach to establish empirical support for the order of behavioral events, found patients were more likely to express motivational statements after provider questions phrased to elicit motivation, reflections of motivational statements, and statements emphasizing patients’ decision-making autonomy. Patients were more likely to express amotivational statements when providers asked questions phrased to elicit amotivational statements or reflected amotivational language. Training providers to strategically phrase their questions and reflections to elicit change language and to emphasize patients’ autonomy may be critical skills for working with adolescents and young adults living with HIV.

Introduction

In 2015, the Centers for Disease Control reported a decline in the overall incidence of HIV in the United States (Prevention, 2016). Despite this promising trend, adolescents and young adults continue to be disproportionately affected by HIV. Specifically, 13 to 24 year olds accounted for nearly a quarter of all new HIV infections in 2015. Adolescents and young adults are at increased risk for contracting and spreading the HIV infection due to their frequent engagement in high-risk behaviors, including unprotected sex and recreational drug use (Centers for Disease Control and Prevention, 2016).

Adolescents and young adults living with HIV are less likely than older adults to be connected to medical care and have a suppressed viral load (Prevention, April 2017). Poor connection to medical care is attributable to many factors, including poor understanding of their diagnosis, stigma surrounding the disease, and a fear of abandonment and discrimination upon disclosure of HIV status (Rao, Kekwaletswe, Hosek, Martinez, & Rodriguez, 2007). Once connected to medical care, adolescents and young adults’ adherence to antiretroviral treatments (ART) is notoriously poor (Pangea Global, 2009), with as many as 49% reporting missed medication doses in the previous week. Low motivation is a common barrier to ART adherence among adolescent and young adult patients who report “forgetting, not feeling like taking medication and not wanting to be reminded of HIV infection” to be the most common barriers to adherence (MacDonell, Naar-King, Huszti, & Belzer, 2013). Poor adherence to ART not only places the health of the individual at significant risk, but also increases the risk of HIV transmission (Cohen et al., 2011). Medical care providers are uniquely positioned to address these barriers with those who are engaged in clinical care.

Motivational interviewing (MI) is a collaborative, goal-oriented method of communication with particular attention paid to language related to behavior change (Miller and Rose, 2009; Rollnick, Miller, & Butler, 2007) and has been adapted for adolescents and young adults (Naar-King and Suarez, 2011). MI provides a highly specified, evidenced-based framework for improving patient-provider communication and promoting behavior change by strengthening a person’s intrinsic motivation in an atmosphere of acceptance and compassion. Providers who demonstrate communication consistent with MI are perceived to be informative, supportive, collaborative, and respectful which, generally, translates to patients who are more satisfied, more committed to treatment regimens, and have better health outcome (Jahng, Martin, Golin, & DiMatteo, 2005; Kaplan, Greenfield, & Ware Jr, 1989; Ong, de Haes, Hoos, & Lammes, 1995; Stewart et al., 2000; Trummer, Mueller, Nowak, Stidl, & Pelikan, 2006).

MI’s efficacy has been demonstrated across multiple behaviors, formats, and disciplines (Lundahl, Kunz, Brownell, Tollefson, & Burke, 2010). MI is associated with improved HIV medication adherence among adults (Thrasher et al., 2006). In adolescents and young adults, MI has been linked to lower viral loads (Naar-King et al., 2009), improved medication adherence (Berwick, 2003), consistent condom use (Chen, Murphy, Naar-King, Parsons, & Adolescent Medicine Trials Network for, 2011), promotion of knowledge of HIV status (Holmes, Swift, Chen, & Hershberger, 2006), reduced substance use (e.g., cannibis, Walker et al., 2011), and reduced alcohol and marijuana use among low to moderate users (Murphy, Chen, Naar-King, Parsons, & Adolescent Trials, 2012). A recent Cochrane review concluded MI is the only evidence-based intervention for youth with HIV (Mbuagbaw, Ye, & Thabane, 2012).

As a result, MI is embedded in HIV clinical care and risk reduction guidelines (U.S. Department of Health and Human Services, Health Resources and Services Administration, & HIV/AIDS Bureau, 2014), however, there are barriers to the integration of MI into these settings. Namely, MI consists of multiple complex behaviors and skills designed to increase intrinsic motivation for enacting behavioral changes, which can be difficult for practicing providers to master (Garrison, Katon, & Richardson, 2005; Leonard, Jang, Savik, Plumbo, & Christensen, 2002; Northam, Lin, Finch, Werther, & Cameron, 2010). Training providers to proficiency with the MI treatment model requires an intensive initial training period and ongoing coaching to achieve and maintain fidelity (Martino et al., 2010; Miller, Yahne, Moyers, Martinez, & Pirritano, 2004; Mitcheson, Bhavsar, & McCambridge, 2009; Theresa B Moyers et al., 2008; Soderlund, Madson, Rubak, & Nilsen, 2011). Barriers reported by healthcare providers include the immense amount of time needed for quality MI training and a lack of confidence in the effectiveness of certain MI components (Schumacher, Madson, & Nilsen, 2014). One approach to increase the feasibility of MI uptake within clinical care settings is to tailor MI training to focus on the specific MI communication strategies most closely linked to change talk [i.e., motivational statements about patients’ own desire, ability, reasons, need for or commitment to behavior change (Miller and Rose, 2009; Rollnick, et al., 2007), language with empirical links to actual behavior change (Apodaca and Longabaugh, 2009; Walker, et al., 2011)] and decrease statements against change [counter change talk or sustain talk (Apodaca et al., 2014)] within providers’ specific treatment context, instead of training providers to full MI competency.

Sequential analysis (R. Bakeman and Quera, 1997; R Bakeman and Quera, 2011) is an analytic approach to examine the sequencing of behavioral events, such as patient-provider communication. It uses a contingency table analysis in which antecedent behaviors form the rows and responding behaviors the columns. Table cells represent the frequency of each antecedent-response behavior sequence and the associated residual describes the extent to which behavior sequences occurred more or less frequently than a chance association. In other words, sequential analysis provides empirical support for establishing the temporal order of patient-provider communication behaviors, such as understand which provider communication behaviors are more likely to elicit patient motivational statements. Early sequential analysis studies of the MI communication process focused on establishing an empirical link between providers’ use of communication strategies consistent with the MI framework [i.e., “MI-consistent communication behaviors” or MICO] and patient motivational language [i.e., change talk and sustain talk]. These studies confirmed that MICO was more likely than other communication strategies to lead to patient change talk (Theresa B. Moyers and Martin, 2006; Theresa B. Moyers et al., 2007; Theresa B. Moyers, Martin, Houck, Christopher, & Tonigan, 2009). The specific MICO communication strategies linked to change talk, however, varied across patient populations. To illustrate, in a study of incarcerated adolescents with high rates of alcohol and marijuana use, change talk was more likely after counselor reflections of change talk; sustain talk, similarly, was more frequent after reflections of sustain talk (Lisa H. Glynn and Moyers, 2010). African American adolescents in weight loss treatment, in contrast, were more likely to express change talk after counselors’ open-ended questions phrased to elicit adolescent change talk and statements emphasizing adolescents’ decision making autonomy (Schoenwald and Hoagwood, 2001). They expressed commitment language [a sub-class of change talk in which patients express their specific intentions, plans for, and action steps taken toward behavior change with a stronger empirical link to behavior change (Apodaca and Longabaugh, 2009)] more often when counselors’ reflected back a previous commitment language statement, asked open-ended questions about commitment, and made autonomy supportive statements. Sustain talk was more likely after reflections about and reflections of sustain talk, reflections of ambivalence, and neutral questions about the target behavior (weight loss).

Understanding the specific provider behaviors likely to elicit motivational language in adolescents and young adults with HIV during clinical interactions will help to tailor and, perhaps, condense MI training for this context. Thus, the goal of the present study was to examine the language used by MI-exposed providers during HIV medical clinic visits to identify the communication strategies most likely to elicit motivational statements from adolescent and young adult patients living with HIV.

Materials and Methods

Participants

Participants were recruited from a multidisciplinary adolescent and young adult HIV clinic located within a large urban teaching hospital providing primary medical care to over 200 adolescents and young adults living with HIV annually. All members of the multidisciplinary care team were eligible to participate, including physicians, nurses, psychologists, social workers, outreach workers/advocates, and case managers. Providers had all been previously exposed to Motivational Interviewing (MI) with varying levels of MI training and proficiency. A total of 80 patient-provider encounters were observed representing the full range of the multidisciplinary clinical team. Half (51%, N=35) were medical providers (n = 15 physician, n= 15 nurses, and n = 5 residents/fellows) and half (49%, N = 34) were psychosocial providers (n = 16 psychologists, n= 16 social workers, and n = 2 health/outreach workers).

Procedures

This study used naturalistic observation of medical encounters. Prior to the initiation of data collection, all providers signed informed consent. Patients treated by consenting providers were approached upon arrival to the HIV clinic for routine appointments. After obtaining patients’ informed consent, audio recorders were placed in patient exam rooms and recorded their clinical encounter. No additional data were collected from patients or providers. A professional transcription company transcribed the audio recordings for qualitative coding. Prior to coding, research staff screened encounters excluding those less than five minutes in duration (M = 14, Range = 5, 64). Interruptions (e.g., other providers entering the room to speak to the provider, patient telephone conversations) were not coded. The university affiliated IRB approved the research.

Minority Youth – Sequential Code for Observing Process Exchanges

The Minority Youth – Sequential Code for Observing Process Exchanges (MY-SCOPE) is a qualitative code scheme to characterize patient-counselor communication during clinical encounters using the MI framework (Idalski Carcone et al., 2013). The MY-SCOPE is an adaptation of the MI-SCOPE (Martin, Moyers, Houck, Christopher, & Miller, 2005) developed for the analysis of dyadic clinical encounters between African American adolescents and a trained MI counselor engaged in weight loss treatment. In the current study, the MY-SCOPE was adapted to study patient-provider communication during HIV medical care visits by providing examples illustrating HIV target behaviors. Target behaviors were HIV medication adherence, HIV clinic attendance, and behaviors that place the patient at risk for transmitting the HIV virus or poor health outcomes (e.g., problematic drinking, substance abuse and sexual risk behaviors). Two coders (research assistants with a bachelor’s degree in psychology) were trained to use the MYSCOPE and released for coding after demonstrating good inter-coder reliability as assessed with Cohen’s kappa (k = .735). Inter-coder reliability across all 16 co-coded sessions was good (k = .688).

Data Analysis Plan

To identify the provider communication behaviors most likely to elicit patient change talk statements, data were analyzed using sequential analysis (R. Bakeman and Quera, 1997; R Bakeman and Quera, 2011), implemented with the Generalized Sequential Querier (GESQ; www2.gsu.edu/;psyrab/gseq/index.html) software package. Sequential analysis is based on a contingency table in which providers’ behavior (given piece of communication) form the rows and patients’ behaviour (target piece of communication) form the columns. The chi square test of association (X2) is used to assess the statistical significance of observed communication sequences. Transition probabilities describe the frequency with which behavior sequences occur within a temporal frame, or lag. In this analysis, we examined transition probabilities at lag 1 which corresponds to the patient statements immediately following a provider statement during a medical encounter. The adjusted residuals and their associated probability values were used to determine which communication sequences occurred more or less frequently than expected (R Bakeman and Quera, 2011). To account for multiple comparisons, we applied a Bonferroni adjustment to our alpha (α = .05 / 100, the number of comparisons, i.e., cells in the matrix, = 0.0005 which corresponds to an adjusted residual >±3.29). To meet the sequential analysis test assumptions, we combined codes into five categories for patients and 21 categories for providers, see Table 1.

Table 1.

Minority-Youth Sequential Coding for Observing Process Exchanges Codes.

Patient (Target) behaviors
Change Talk	Statement conveying underlying motivation, including desire, ability, reason, or need for behavior change. I take my medication because I could give my partner HIV.
Commitment Language	Statement about taking action, including current or future agreement, intention, or obligation related to behavior change I’ll take my medication tomorrow.
Sustain Talk	Statement describing current or future intention to avoid or take action against target behavior change, or statement describing underlying motivation to not change. I didn’t tell my partner about my HIV status.
HIV-Related	Statement about a target behavior that does not meet the definition for change talk, commitment language, or sustain talk. What are the side effects of this medication?
Other Statements	All other patient statements including those that end topic development or develop the topic of the conversation but are about non-target behaviors. When I was a kid, people used to tease me a lot and it really made me mad.
Provider (Given) behaviors
Affirmation	Positive or complimentary statement that expresses appreciation, confidence, or reinforces the patient’s strengths or efforts. So, you did that. I don’t know if you get how huge this is for you, like, and how much we see that. This is a huge step. This is fantastic, and it doesn’t mean that it’s easy.
Emphasize Autonomy	Statement directly acknowledging, honoring, or emphasizing the patient’s freedom of choice, autonomy, or personal responsibility. I really respect the fact that you are not ready to share your diagnosis with your mother.
General Information Positive	Offering educational information, advice, suggestions, or potential solutions/actions using patient-centered communication strategies such as asking for permission, using the third person, giving the option to reject information before providing it, and offering a menu of options. I have some information about drugs, would you like to hear it?
General Information Negative	Offering educational information, advice, suggestions, or potential solutions/actions, or expressing concern in a non-productive way such as not asking for permission, using imperative language (e.g., “I want you to...”), re-emphasizing negative consequences, and disagreeing/arguing with the patient. You will never have an undetectable viral load if you continue to miss doses.
Reflection of Change Talk	An active listening statement that restates a patient’s own statement or behavior regarding his/her underlying motivation for behavior change. You are worried that HIV is going to affect your health.
Reflection of Commitment Language	An active listening statement that restates a patient’s own statement or behavior regarding his/her current or future intention, plan, or action related to behavior change. So, you feel confident that you can take your medication every day.
Reflection of Sustain Talk	An active listening statement that restates a patient’s own statement or behavior regarding his/her current or future intention to avoid or take action against target behavior change, or a statement describing underlying motivation to not change. In the past, side effects were one reason you didn’t take your medication.
Other Reflections	An active listening statement that restates a patient’s own statement or behavior on a topic unrelated to the target behavior. You’re not liking this weather either.
Open Question to Elicit Change Talk	A question designed to allow a wide range of possible answers and phrased to elicit the patient’s underlying motivation for behavior change. Why might you want to talk to your partner about your HIV status?
Open Question to Elicit Commitment Language	A question designed to allow a wide range of possible answers and phrased to elicit the patient’s current or future intention or action related to behavior change. What ideas do you have about talking to your partner?
Open Question to Target Behavior Neutral	A question designed to allow a wide range of possible answers related to the target behavior without having a specific slant towards eliciting change talk and commitment language, or sustained talk. How have you been feeling on this new medication?
Open Question to Elicit Sustain Talk	A question designed to allow a wide range of possible answers and phrased to elicit a patient’s current or future intention to avoid or take action against target behavior change, or a statement describing underlying motivation to not change. Why do you prefer not to wear a condom?
Other Open Questions	A question designed to allow a wide range of possible answers unrelated to the target behavior. So, what do you do in your free time?
Closed Question to Elicit Change Talk	A question that implies a short answer or specifies a restricted range of responses phrased to elicit the patient’s underlying motivation for behavior change. Will you talk to your partner about your HIV status?
Closed Question to Elicit Commitment Language	A question that implies a short answer or specifies a restricted range of responses phrased to elicit the patient’s current or future intention or action related to behavior change. When will you talk to your partner?
Closed Question to Target Behavior Neutral	A question that implies a short answer or specifies a restricted range of responses regarding the target behavior that does not have a specific slant towards eliciting change talk and commitment language, or sustained talk. Do you have any problems with taking your medication?
Closed Question to Elicit Sustain Talk	A question that implies a short answer or specifies a restricted range of responses phrased to elicit a patient’s current or future intention to avoid or take action against target behavior change, or a statement describing underlying motivation to not change. You don’t like coming to clinic?
Other Closed Questions	A question that implies a short answer or specifies a restricted range of responses unrelated to the target behavior. You need some water?
MI Inconsistent Strategies	Counselor strategies that use a negative tone while directly engaging the patient (e.g., shaming, blaming, ridiculing, etc.), warn/threaten of negative consequences of the patients’ behavior, or offers advice, suggestions, or actions in a direct and imperative way (e.g., “You should...”, “Why don’t you...”, etc.,). “I just don’t understand how you continue to forget to take your medication!”
Other Statements	Statements that are not counselor strategies/reflections or are unrelated to target behavior. Okay, Cool.

Results

A total of 12,373 utterances were coded across the 80 medical encounters; of which 6,360 were provider utterances and 5,623 were patient speech. This resulted in 5,396 communication sequences analyzed (Table 2). Less than 10% (9.2%, n = 499) of communication sequences ended with a patient change talk statement due, in part, to the disproportionate number of non-target behavior-related (i.e., “other”) provider utterances (75.0%, n = 4036).

Table 2.

Transition and Marginal Frequencies of Patient-Provider Communication Behaviors during Medical Encounters.

Provider (Given) behaviors	Patient (Target) behaviors
	Change Talk	Commitment Language	Sustain Talk	HIV-Related	Other Statements	Total
Affirmation	13	6	3	4	21	47
Emphasize Autonomy	7	3	5	8	23	46
General Information Positive	33	14	12	82	169	310
General Information Negative	7	7	13	18	46	91
Reflection of Commitment Language	1	15	3	3	30	52
Reflection of Change Talk	16	1	3	4	13	37
Reflection of Sustain Talk	2	3	32	1	9	47
Other Reflections	3	8	4	8	260	283
Open Question to Elicit Change Talk	11	0	3	5	0	19
Open Question to Elicit Commitment Language	1	5	1	1	0	8
Open Question to Target Behavior Neutral	5	6	7	13	7	38
Open Question to Elicit Sustain Talk	1	3	19	1	2	26
Other Open Questions	1	5	9	7	362	384
Closed Question to Elicit Change Talk	52	7	12	2	5	78
Closed Question to Elicit Commitment Language	4	71	17	4	3	99
Closed Question to Target Behavior Neutral	16	38	20	103	26	203
Closed Question to Elicit Sustain Talk	9	25	95	6	11	146
Other Closed Questions	15	9	12	20	1797	1853
MI Inconsistent Strategies	5	11	13	18	66	113
Other Statements	30	30	34	59	1363	1516
Total	232	267	317	367	4213	5396

Table 3 presents the transition probabilities and statistical significance of the observed patient-provider communication sequences. Provider questions phrased to elicit change talk were followed by a patient change talk statement more often than other provider communication strategies. Closed questions phrased to elicit change talk were followed by change talk 67% of the time and open questions were 58% of the time. In addition, reflections of patient change talk (43%), affirmations (28%), statements emphasizing the patient’s decision-making autonomy (15%), and information presented in a patient-centered manner (11%) were also more likely to elicit a patient change talk statement. All p ≤ .0001.

Table 3.

Transition Probabilities of the Joint Frequency of Patient-Provider Communication Behaviors during Medical Encounters.

Provider’s (Given) behaviors	Patient’s (Target) behaviors
	Change Talk (CT)	Commitment Language (CL)	Sustain Talk (ST)	HIV-Related (HIV)	Other Statements (O)
Affirmation	.28*	.13	.06	.09	.45†
Emphasize Autonomy	.15*	.07	.11	.17	.50†
General Information Positive	.11*	.05	.04	.26*	.55†
General Information Negative	.08	.08	.14*	.20*	.51†
Reflection of Change Talk	.43*	.03	.08	.11	.35†
Reflection of Commitment Language	.02	.29*	.06	.06	.58†
Reflection of Sustain Talk	.04	.06	.68*	.02	19†
Other Reflections	.01	.03	.01	.03	.92*
Open Question to Elicit Change Talk	.58*	.00	.16	.26*	.00†
Open Question to Elicit Commitment Language	.13	.63*	.13	.13	.00†
Open Question to Elicit Sustain Talk	.04	.12	.73*	.04	.08†
Open Question to Target Behavior Neutral	.13	.16	.18*	.34*	.18†
Other Open Questions	.00†	.01†	.02	.02†	.94*
Closed Question to Elicit Change Talk	.67*	.09	.15*	.03	.06†
Closed Question to Elicit Commitment Language	.04	.72*	.17*	.04	.03†
Closed Question to Elicit Sustain Talk	.06	.17*	.65*	.04	.08†
Closed Question to Target Behavior Neutral	.08	.19*	.10	.51*	.13†
Other Closed Questions	.01†	.00†	.01†	.01†	.97*
MI Inconsistent Strategies	.04	.10	.12	.16*	.58†
Other Statements	.02†	.02†	.02†	.04†	.90*

^*Greater than chance, p<0.0005

^†Less than chance, p<0.0005.

A similar pattern was found when examining communication patterns associated with patient commitment language. The provider communication behaviors most likely to be followed by patient commitment language were questions phrased such that commitment language was the expected response. Closed questions phrased to elicit commitment language did so 72% of the time and open questions phrased to elicit commitment language elicited commitment language 63% of the time. Patients were likely to respond with a commitment language statement when providers made a statement reflecting back a previous commitment language statement (29%), asked a neutrally worded closed question (19%), and, unexpectedly, asked a closed question phrased to elicit sustain talk (17%).

Lastly, the provider communication behaviors most likely to be followed by patient sustain talk were asking open questions that were phrased to elicit sustain talk (73%), making reflective statements of patients’ sustain talk (68%), and asking closed questions phrased to elicit sustain talk (65%). In addition, provider statements that provided information in a non-patient-centered manner (14%), neutrally worded open questions (18%), closed questions phased to elicit change talk (15%), and closed question to elicit commitment language (17%) were more likely than other provider communication behaviors to be followed by patient sustain talk.

Discussion

The goal of this research was to identify the provider behaviors linked to patient change talk, commitment language, and sustain talk during HIV clinical encounters. The main finding is that young adult patients were more likely to make change talk statements when providers asked open or closed questions that were phrased to elicit change talk or reflected a previous change talk statement. The same was true for commitment language. In contrast, open questions, closed questions, and reflections of sustain talk were the provider behaviors most likely to elicit sustain talk. Neutral questions and reflections of ambivalence were also likely to elicit sustain talk. These findings are consistent with a growing body of MI research suggesting that the content of questions and reflections is important for the elicitation of motivational statements (i.e., change talk and commitment language) and the reduction of amotivational statements (i.e., sustain talk) (Carcone et al., 2013; Lisa H Glynn, Houck, Moyers, Bryan, & Montanaro, 2014; Jacques-Tiura et al., 2017; Theresa B. Moyers, et al., 2009; Rodriguez, Walters, Houck, Ortiz, & Taxman, 2017).

In addition to questions and reflections, using language that emphasized autonomy was instrumental in eliciting change talk statements among young adult patients in the HIV medical care setting. Only one study to date has independently examined the relationship between autonomy supportive statements and the elicitation of change talk. Among African American adolescents and their caregivers in weight loss treatment, statements emphasizing autonomy were more likely to result in change talk or commitment language and less likely to elicit sustain talk (Idalski Carcone, et al., 2013; Jacques-Tiura, et al., 2017). Most MI communication studies have included statements emphasizing autonomy within an index of MI-consistent communication strategies, instead of a specific behavior count, which is sometimes associated with increased change talk and other times not (Gaume, Bertholet, Faouzi, Gmel, & Daeppen, 2010; Theresa B. Moyers, et al., 2009). It is clear that in the context of promoting adolescents and young adults’ motivation through communication, specific statements that emphasize autonomy (e.g., by highlighting personal choice and responsibility, clarifying the provider role as a guide and not an expert) are critical.

These findings have implications for tailoring training for young adult HIV treatment providers. First, training may avoid teaching reflective listening generally, as reflections of sustain talk or even of neutral statements are likely to lead to statements against change. Alternatively, training may focus on the recognition of change talk first, and then on how to reflect and elicit change language. Second, less focus might be paid to closed versus open ended questions as both resulted in the elicitation of change talk and commitment language when focused on this language. It is possible, however, that richer change talk may result from open questions versus closed questions that tend to elicit single word utterances. Future research may address this question by coding the intensity of the change language expressed. Third, statements that emphasize autonomy appear to be a critical skill that deserve special attention and practice when working with adolescents and young adults living with HIV. Finally, a significant proportion of the utterances (75.0%, n = 4,036) addressed topics unrelated to the three behavioral targets – HIV medication adherence, HIV clinic attendance, and risky behavior. Thus, tailored training might emphasize refocusing communication on target behaviors and tying intermediate behavioral outcomes (e.g., employment, mental health) back to HIV-related behaviors to maximize the impact of brief encounters in the clinic setting. We will be testing the effects of such a tailored workshop on provider skill and cascade outcomes in a multi-site protocol within the Adolescent Trials Network for HIV/AIDS Interventions (ATN Protocol 146).(Naar, Carcone, Todd, MacDonell, & Martinez, 2018)

This study is limited by a single observation site with a restricted number of providers and encounters. Thus, replicating this work in a larger more diverse sample is needed and is underway. We will be sequentially analyzing youth-provider interactions across multiple protocols within the ATN Scale It Up center (U19HD089875). This will allow us to test the communication sequences in multiple clinics as well as in settings outside the clinic (e.g., outpatient counseling sessions, telephone interactions). These results, while consistent with the broader literature on the mechanisms of Motivational Interviewing’s efficacy, should be considered in light of the limited focus on three target behaviors – HIV medication adherence, HIV clinic attendance, and behaviors that place the patient risk at for transmitting the HIV virus or poor health outcomes. These behaviors were defined a priori as critical correlates of HIV medication adherence which, in turn, mediates HIV viral load suppression; however, a surprising proportion (75.0%) of the conversations between patients and HIV clinic providers involved non-target behaviors, such as housing stability, mental health, and non-HIV health conditions and concerns. Understanding the extent to which motivational language about these non-target behaviors is associated with HIV outcomes would be an important future direction of this research. Relatedly, we decided to interpret brief responses to closed-ended questions, e.g., “yes” and “no”, as motivational language (change talk, commitment language, and sustain talk). A more conservative approach would have been to restrict motivational language to overt patient statements expressing their desire, ability, reasons, need for, or commitment to behavior change.

Conclusions

This research adds to a growing literature on the Motivational Interviewing communication process by examining medical care providers in a multidisciplinary HIV clinic, a yet to be studied provider population. Findings suggest ways to improving patient care via improved patient-provider communication and may also have implications for improving communication in other behavioral interventions for youth living with HIV. Results suggested that effective communication, that which promotes behavior change, in HIV medical care settings is characterized by specifically phrasing questions and reflections to encourage young adult patients to express their intrinsic motivations (i.e., change talk and commitment language statements) for engaging in health promoting behaviors and statements that emphasize patients’ decision-making autonomy. Testing interventions that incorporate these findings and demonstrating relationships to cascade-related outcomes is a next step.