Abstract
Introduction Treatment interruption (TI), defined as >28 days late for ART appointment, is one of the greatest challenges in controlling southern African HIV epidemics. Negative client-provider interactions remain a major reason for TI and barrier for return to care, especially for men. Motivational interviewing (MI) facilitates client-driven counseling and improves client-provider interactions by facilitating equitable, interactive counseling that helps clients understand and develop solutions for their unique needs. Fidelity of MI counseling in resource-constrained health systems is challenging.
Methods We developed a male-specific MI curriculum for Malawian male TI clients. Four psychosocial counselors (PCs, a high-level Malawian counseling cadre) received a 2.5-day curriculum training and job-aid to guide MI counseling approaches. They participated in monthly phone-based discussions with their manager about MI-based solutions to challenges faced. PCs implemented the MI curriculum with men >15 years who were actively experiencing TI. Clients were found at home (through tracing) or at the facility (for those who returned to care on their own). MI counseling sessions were recorded, transcribed, translated into English, and coded in Atlas.ti v9. MI quality was assessed using a modified version of the validated Motivational Interviewing Treatment Integrity tool. The tool has two measures: 1) counts of key MI behaviors throughout the session (questions, reflections, etc.); and 2) overarching scores (using a five-point scale) that characterize three MI dimensions for an entire counseling session (cultivating change talk, partnership, and empathy).
Results 44 MI sessions were recorded and analyzed between 4/1/22-8/1/22. 64% of counseling sessions focused on work and travel as the main reason for TI. 86% of sessions yielded client-driven, tailored solutions for overcoming TI. PCs implemented multiple MI behaviors very well: asking questions, giving information, simple reflections, and client affirmation. Few PCs used complex reflection, emphasized autonomy, or sought collaboration with clients. Among overarching MI dimensions, HCWs scored high in partnership (promoting client-driven discussions) and cultivating change talk (encouraging client-driven language and behavior change confidence) but scored sub-optimal in empathy. Only 5 sessions had confrontational/negative PC attitudes.
Conclusions PCs implemented MI with fidelity and quality resulting in tailored, actionable plans for male re-engagement in HIV treatment in Malawi.
Clinical Trial Number NCT05137210 and NCT04858243
Introduction
Treatment interruption (TI) from HIV care is one of the most pressing challenges to controlling HIV epidemics in eastern and southern Africa [1–3]. In Malawi, TI is defined as being >28 days late for an ART refill appointment. In the region, it is estimated that 32-46% of clients on ART experience TI within the first five years on treatment [4–6], and 30-40% of clients experience repeat TI [7–8]. Men experience TI at a 15-26% higher rate than women [9] and have worse HIV outcomes like increased morbidity, mortality, and viremia [10–11]. As such, while countries like Malawi are quickly approaching aggregate UNAIDS 95-95-95 targets, young men aged 15-34 have largely fallen behind [12].
Resource limitations have been noted to increase rates of provider compassion fatigue and withdrawal [13], resulting in poor client-provider interactions that both contribute to male TI and diminish male return to care [14]. Providers often describe men as being “ill-informed,” “difficult,” “selfish,” and overall “bad clients [15].” These biases may reflect in inequitable client- provider power dynamics and can result in men avoiding HIV services [14,16]. The hierarchical structure of health systems further entrenches this by positioning providers as the expert on a client’s situation rather than as a collaborator working alongside the client [13]. The result is predominantly didactic, authoritative counseling that fails to address individualized needs of clients [13,17]. Given these issues, the World Health Organization (WHO) has increasingly called for HIV services geared toward person-centered approaches [16].
Person-centered care (PCC) is an umbrella term for services that center client respect, client engagement in informed decision-making, and support for holistic client needs (physical, psychological, and social) [14,18–20]. Specific PCC techniques, like motivational interviewing (MI), promote responsive counseling and client-driven solutions that may help to address barriers to care. MI is a counseling technique aimed at strengthening a clients’ internal motivation, commitment to behavioral change, and development of individualized action plans that promote sustained behavior change [21]. MI has been associated with increased ART adherence and decreased incidence of sexual risk taking [22,23], and qualitative data suggests client-driven counseling techniques such as MI are deeply desired by male ART clients [24,25].
Despite its advantages, MI is infrequently utilized given the extensive resources required that are rarely available in these low-resource settings (e.g., increased personnel time, higher level of personnel training, and increased monitoring and evaluation). Other challenges to regional implementation include poor fidelity (including poor uptake of “the spirit of MI” and its core behaviors, namely empathy and reflective listening) [26,27], and poor session attendance [22]. Furthermore, there is no evidence on MI as a tool for men living with HIV (MLHIV).
To improve client-provider interactions and support men experiencing TI in Malawi, we developed a male-specific MI curriculum and assessed quality and fidelity of its implementation by psychosocial counselors (PC).
Materials & Methods
Setting
We developed a male-specific MI curriculum for male TI clients and implemented it across 8 facilities in Malawi from April 1-August 31 2022. All health facilities were supported by Partners in Hope, a PEPFAR implementing partner that supports HIV service delivery in over 100 facilities serving >200,000 people living with HIV across Malawi. We purposively selected facilities to include large public district hospitals (n=3) and small health centers (n=5), in the central (n=4) and southern (n=4) regions.
Male-Specific MI Curriculum
The MI curriculum was composed of two parts. First, an introductory guide was developed to describe the MI approach and help PC’s develop the needed skills to implement MI. This included guidance on how to (1) build rapport with clients, including discussing their hopes and goals for the future; (2) support clients to identify their personal barriers to care; (3) explore the potential benefits of treatment adherence and connect these benefits with the clients’ goals; (4) encourage clients’ decision making for change, and; (5) help clients set their own short- and long-term goals related to HIV treatment. Our training materials were developed using the REDI Framework and training guide curriculum developed by Engender Health to aid in client contraception choice [28]. It is one of the few examples of MI-based client-centered care in the region. Second, a male- specific MI job aid was developed to be used during one-time counseling sessions. We used literature on barriers to men’s HIV care [29–31] and formative qualitative research [15, 32–37] to identify 11 key domains related to barriers to men’s use of HIV treatment. The job aid provided an overview of each topic to sensitize the PC on common challenges to men’s use of HIV treatment and probes to understand the unique barriers, priorities, and resources of each individual client. Each domain had MI related discussion prompts to facilitate client-led discussions and solutions. These were organized by ‘Think’ and ‘Act’ questions. ‘Think’ prompts were used to help the client identify the barrier to care or to reframe the challenge to gain a better understanding of how it could be overcome. ‘Act’ prompts were meant to support the client’s decisions for change and were used to explore actionable ways of overcoming the challenge. A brief description is provided in Appendix A.
PC Training
Four PC, a high-level Malawian counseling cadre, received a 2.5-day curriculum training comprised of 1) sensitization to the challenges and needs of men as ART clients; 2) education on best practices for male-specific MI, including building rapport, exploring barriers to HIV care, decision making for overcoming challenges, and strategizing how to implement them; and 3) practice of key MI skills through role play and peer feedback. A key focus throughout the training was the co-creation of actionable next steps for improving clients ART adherence. After the training, PC participated in monthly phone-based discussions with their manager for group feedback and discussion about challenges and solutions for implementing the male-specific MI curriculum.
Implementation
Men who were ≥15 years, living in study facility catchment areas, and not currently in HIV care (defined as never initiated ART or experiencing treatment interruption (missed an ART appointment >7 days or defaulted (out of care >28 days)) or having a high viral load (>1,000 copies/ml) were offered the male-specific MI curriculum. Routine medical charts and health registers were reviewed to identify potentially eligible men. Each PC was responsible for MI implementation at two health facilities. Per standard of care, PC actively traced potentially eligible men to offer male-specific MI counseling. Men were recruited either through this community tracing, or at health facilities if they returned for their ART refill and were flagged as recently having a TI or high viral load. All identified men were screened for eligibility and, if eligible, asked for oral consent to record a one-time MI counseling session.
Measuring Fidelity & Quality in Counseling Sessions: MITI
We measured the quality of MI components within counseling sessions using a modified version of the Motivational Interviewing Treatment Integrity tool (MITI), a validated scale to assess MI competencies and provide feedback on the fidelity and quality of MI interactions [38]. We utilized two key measures from the tool: 1) objective counts of key MI behaviors throughout the session; and 2) subjective overarching ‘global’ ratings (using a five-point Likert scale) for domains characterizing the ‘spirit’ of MI (see Appendix B). Some domains from the tool (including Softening Sustain Talk, Persuade with Permission and Persuade) were not included in our analysis due to the cultural context and the fact that some themes from MITI cannot be measured well via transcripts. Behaviors were classified as “simple” versus “complex” in accordance with whether the behavior in question is common to traditional counseling (e.g., asking questions, giving information, simple reflection, affirmation) versus if the behavior is more unique to MI-based counseling (e.g., complex reflection, seeking collaboration, emphasizing autonomy).
In addition to the MITI tool measures, we also assessed PC’s abilities to problem solve and develop next steps to overcome adherence challenges. We rated sessions as ‘effective’ if actionable next steps were created to address the session’s client-identified barriers, or ‘not effective’ if no actionable next steps were created.
Data Analysis
MI counseling sessions were recorded, transcribed, and translated into English for analysis. We completed structured memos based on the MITI scale to measure fidelity and quality, including frequency of MI behaviors and ratings of MI domains. Transcripts were reviewed and key data was extracted and entered into the memo, including simple excerpts of text that reflected strong or weak implementation of MITI domains. Transcripts were simultaneously coded in Atlas.ti v.9 using codes for each of the MITI behaviors to aid in the behavior counts and domains to ensure all content related to themes were captured in their entirety. Client reasons for TI, counts of global rating scores and key MI behaviors per interview were analyzed in Excel.
Ethics Statement
The IDEaL trial is registered with ClinicalTrials.gov (NCT05137210) and was approved by the UCLA and NHSRC of Malawi IRB. Men were recruited either through this community tracing, or at health facilities if they returned for their ART refill and were flagged as recently having a TI or high viral load. All identified men were screened for eligibility and, if eligible, asked for oral consent to record a one-time MI counseling session.
Results
Overall, 44 counseling sessions were conducted. Counseling session duration ranged from 6-60 minutes, with an average of 27 minutes (IQR: 18.75). All sessions were completed by male PCs with a minimum of three years’ counseling experience prior to the study (mean: 5.65 years’ experience).
Client Type and Reason for TI
Client type was recorded for 36/44 counseling sessions. Twenty men (45%) had been out of care for > 28 days, fifteen (34%) had missed an ART appointment by 7 days, one (2%) had high viral load, and 8 (18%) were unrecorded. Across the 44 counseling sessions, participating men most frequently cited “work & travel” (n=28, 64%) as their main reason for TI. Other reasons for TI included difficulty “remembering appointment and medication schedules” (n=5, 11%) and feelings of “loneliness, isolation, and lack of social support” (n=4, 9%).
Application of Key MI Behaviors
We measured frequency of the eight key MI behaviors throughout each counseling session. While no ‘target frequency’ exists for any of the MI behaviors, PC most frequently employed simple MI behaviors (e.g., giving information, asking questions, simple reflection, and affirmation). Advanced behaviors (e.g., complex reflection, seeking collaboration, and emphasizing autonomy) were less frequently utilized. PC overwhelmingly used the seven positive MI behaviors and rarely used the one negative behavior, confrontation, marked as being antithetical to the ‘spirit of MI.’ Only 11% of counseling sessions (n=5) had one or more instances where the PC demonstrated confrontational behavior (Table 1).
Mean number of behavior instances per interview, with variance
Most Frequently Used Behaviors
PC relied heavily on the four simple MI behaviors. Giving information and asking questions were the most frequently utilized. Information was given to resolve knowledge gaps while asking questions worked to identify these gaps and push the conversation deeper. PC utilized simple reflection primarily to help clients feel heard and understood. PC echoed client’s words back to them to indicate that they were paying attention or to give the client opportunity to clarify and reinforce their own thought processes. PC used affirmation to acknowledge a client’s positive attributes and to encourage the positive aspects of a client’s behavior even when the overarching behavioral theme resulted in negative outcomes, like treatment interruption (Table 2).
Summary of most used MI behaviors, with example quotes
Least Frequently Used Behaviors
PC rarely used advanced MI behaviors. When used, complex reflections allowed PC to distill a client’s thought into key points, pushing the conversation forward by both extracting and emphasizing salient emotions and themes that arose throughout the client’s words. Even rarer was the use of seeking collaboration. However, when used, it emphasized power sharing and allowed the client to drive the conversation in a direction most suitable for their personal needs. Emphasizing autonomy was the least utilized positive MI behavior among PC. Despite this, its implementation was critical to fostering client buy-in by dispelling notions of being forced or shamed into regular medication adherence.
While infrequent, confrontation (negative MI behavior) did occur. PC reverted to confrontational language when they felt their clients were not being honest and when they felt they needed to correct behaviors they viewed as negative (Table 3).
Summary of least used MI behaviors, with example quotes
Creating Actionable Next Steps
Across the 44 counseling sessions, ninety-one percent of counseling sessions resulted in actionable next steps for at least one barrier to care and 86% developed solutions for all identified barriers (See Appendix C examples). With the support of PC, men developed plans for overcoming barriers to care including disclosure to close male friends, posting ARV reminders on their bedroom walls, and having a plan in place for unanticipated travel, among others.
Four out of the 44 sessions did not yield actionable steps; the sessions identified as failing to yield next steps were all focused on ‘work and travel’ as the primary barrier to care. Of the four instances where no next steps were given for the topic of “work & travel,” two involved the client not having adequate transport to attend the facility and two involved one-off instances of illness while traveling which prevented return to the facility. Counseling discussions suggested particular difficulty overcoming time and cost associations with work and travel.
Quality of MI by Global Ratings
Global ratings for elements that represented the ‘spirit of MI’ – cultivating change talk, partnership, and empathy – were graded on a 1 (lowest) to 5 (highest) scale for each counseling session (Table 4).
Mean, median, and mode global rating score per interview, with variance
Partnership
PC generally scored best in partnership, or their ability to convey an understanding that expertise and wisdom about change resides mostly within the client, with a mean score of 3.8. PC displayed partnership through statements that highlighted wisdom residing in the client themselves and by using terminology that offered the client an equal position in the discussion. When PC used language to convey partnership, it elicited higher-level engagement from the client; rather than simply listening, the client began to actively participate and offer their critical thoughts on the subject at hand. This helped build rapport and the client’s sense of agency in crafting a plan toward treatment re-initiation. Most commonly, partnership came in the form of questions that evoked solutions from the client, but it occasionally came in the form of explicit statements calling for equality.
Despite scoring well in partnership, several exchanges between clients and PC signaled that clients were resistant to, or initially did not accept, the PC’s bid for partnership. One such example is seen below where the client appears to be either expecting or asking for a more traditional, didactic interaction with the PC:
PC: The medication is not compatible with you. Okay, so do you know how we solve this problem?
Client: You tell me.
Cultivating Change Talk
PC also achieved good scores for cultivating change talk, marked by the mentor actively encouraging the client’s own language in favor of the change goal, with a mean score of 3.16.
Cultivating change talk occurred through extended back and forth dialogue between the PC and client whereby the PC would help identify a cause of treatment interruption and explore the barrier with broad, open-ended, and evocative questions that required the client to consider the issue at hand from multiple angles. The result was often that the client would come to a conclusion that aligns with traditional PC recommendations but on their own accord.
Empathy
PC received the lowest scores for empathy, with an average mean score of 2.75. When employed, PC used empathic language to demonstrate that they saw the client as a human and understood that perfection is not always a reasonable goal. They emphasized that one must give themself grace while on treatment— the client should not wait to start treatment until they can adhere perfectly; everyone forgets at times and persevering is what matters. PC also expressed empathy by apologizing on behalf of the health system for negative experiences the client endured in the past even when the PC was not directly responsible for the negative interaction. Mostly commonly, empathy was expressed in the form of “as men” statements which sought to express to the client that the PC understood, as a fellow man, that there are many conflicting responsibilities that can make adhering to treatment particularly difficult (Table 5).
Summary of global ratings, with example quotes
Discussion
We assessed quality and fidelity of male-specific MI curriculum implementation by PC in routine HIV care settings in Malawi and found that PC were able to achieve both quality and fidelity in certain areas. Following a brief training, PC could (1) effectively use most MI behaviors with rare instances of confrontation, (2) demonstrate successful application of MI domains, and (3) guide counseling sessions that resulted in actionable next steps for clients to overcome TI. However, PC demonstrated infrequent usage of advanced MI behaviors and did not display high rates of empathy with clients suggesting that additional training or altered curricula may be needed to promote advanced MI behaviors and global skills.
When implementing male-specific MI, PC in our study overwhelmingly relied on simple behaviors (asking questions, giving information, simple reflection, and affirmation). Despite their simplicity, these simple MI behaviors seemed to engage male clients by clearing knowledge gaps, making clients feel seen and heard, and affirming clients. Advanced MI behaviors (complex reflection, seeking collaboration, and emphasizing autonomy) were used less frequently but served to deepen conversations, center counseling on the unique needs of the client, and reaffirm that enduring solutions reside within the client, not the PC. Infrequent usage of advanced MI behaviors may reflect ingrained client-provider power dynamics and hierarchies that position providers as experts meant to direct, rather than collaborate, with a client [13]. Infrequent usage of complex behaviors may also reflect the need for additional training to advance the nuanced counseling skills necessary for deeper MI implementation. PCs’ ability to engage in effective MI despite minimal use of complex MI behaviors suggests that simple, paired down MI behaviors may be sufficient to instill the ‘spirit of MI’ into a counseling session in contexts such as Malawi. While infrequent, rare confrontational exchanges worked to entrench typical counseling hierarchies and fortify the unequal power dynamics that MI seeks to dismantle, thus creating a more hostile environment not conducive of open dialogue.
Despite most interviews resulting in actionable next steps and being rated as ‘effective,’ there was wide fluctuation in behavior frequency across interviews, suggesting that there is no ‘right number’ of behavior frequencies required to enact high-quality MI. Frequencies of usage should be expected to be variable, fitting with the spirit of MI, which seeks to provide an individualized approach rather than a ‘one-size-fits-all’ approach. Effective counseling sessions also varied considerably in length, further supporting that there is no one-size-fits-all approach to quality, effective MI.
Forming actionable next step plans was critical to enacting effective MI. Ninety-one percent of counseling sessions resulted in actionable next steps for at least one barrier to care and 86% developed solutions for all identified barriers. With the support of PC, men developed plans for overcoming barriers to care including disclosure to close male friends, posting ARV reminders on their bedroom walls, and having a plan in place for unanticipated travel, among others. While rare, some sessions failed to develop plans for overcoming; this overwhelmingly occurred when the reason for TI was “work & travel,” which was also the main identified reason for TI among study participants. Counseling sessions suggested that time and cost associations with work and travel may be particularly challenging to address due to their logistical nature, particularly in low- resource settings. Other literature has similarly found that many clients experience HIV treatment interruption in part due to insufficient affordable transport solutions and conflicting financial demands which are not easily resolved given current infrastructure and socioeconomic realities [39–41]. Additional novel structural interventions at the health facility level, like home ARV delivery or multi-month dispensing, may be needed to overcome these barriers, if feasible in the local context.
PC were largely proficient in two of the three core elements that represent the ‘spirit of MI,’ partnership and cultivating change talk, after only 2.5 days of training. Despite success in those domains, PC scored worse in empathy. This may reflect several challenges including, but not limited to: (1) cultural practice of expressing empathy in less explicit or non-western manners [42], (2) PC practice of expressing empathy through body language rather than words, thus not flagging in transcripts, or (3) insufficient training in engaging MLHIV with empathetic statements.
The overwhelming proportion of study counseling sessions developed actionable next steps and were ultimately rated as ‘effective,’ suggesting that brief trainings focused on developing MI behaviors and low-cost job aids which remind PC of basic MI principles may be effective tools to aid basic MI implementation in routine settings. As HIV programming transitions back to ministries of health from PEPFAR [43–45], these results bode well for continued implementation; it is possible to provide high-quality, skilled MI counseling with little resources and time with minimal upfront investment.
Conclusion
Our findings are critical for understanding feasibility of providing high-quality MI counselling in resource-limited settings and will only become more relevant as funding for HIV programming across the continent becomes more constrained. Our results suggest that PC can implement high- quality MI with fidelity in a low-resource setting with brief training. While largely successful, additional training or altered curricula may help promote the lesser-seen behaviors and global skills, like complex reflection or empathy. Further, “work & travel” remains a chief barrier to treatment among MLHIV and a particularly challenging barrier to care for PC and clients, alike, to overcome. Future work should focus on taking male-specific MI counseling to scale in the standard-of-care through continually simplified trainings, curricula, and monitoring and evaluation strategies. Additional areas of future focus include exploring challenges to PC engagement with advanced MI skills, like complex reflection and empathy.
Competing Interests
The authors declare that they have no competing interests.
Authors’ Contributions
KD is responsible for funding acquisition. JH, MM, IR, EC and KD developed the male-specific counseling curriculum and study implementation protocol. JH, MM and IR implemented the study and collected all data. KNH, JH and KD contributed to the analysis conception. KNH and JH coded and analyzed the data with support from KD. KNH wrote the first draft and JH, MM, IR, KP, DO, EC, KD and AC edited following drafts. All authors have read and approved the final manuscript.
Funding
The work is supported by the Bill and Melinda Gates Foundation (INV-001423) and by the National Institute of Mental Health (R01-MH122308). KD is supported by the Fogarty International K01-TW011484-01 and UCLA GSTTP. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Disclaimers
None
Data Availability Statement
The data that support the findings of this study are openly available in figshare at https://figshare.com/s/430c414b7401b8e22067 ; DOI: 10.6084/m9.figshare.26240885
List of Abbreviations
- IRB
- International Review Board
- MI
- motivational interviewing
- MITI
- Motivational Interviewing Treatment Integrity Tool
- MLHIV
- men living with HIV
- NHSRC
- National Health Sciences Research Council
- PC
- psychosocial counselors
- PCC
- person-centered care
- TI
- treatment interruption
- UCLA
- University of California, Los Angeles
- WHO
- World Health Organization
Acknowledgements
We are grateful to the male clients and HCWs who participated in the study and our IDI enumerator, McDaphton Bellos, and FGD leads Misheck Mphande and Eric Lungu who conducted the interviews. We would also like to acknowledge Partners in Hope Malawi and thank them for their partnership and collaboration.
Appendix A Topics & Descriptions in the Male Specific MI Curriculum
Appendix B Domains Included in the MITI Scale & Definitions
Appendix C Next Step Action Plans
Footnotes
E-mail addresses of authors: KNH: knholland{at}mednet.ucla.edu, JH: julieannehubbard{at}outlook.com, MM: misheck{at}pihmalawi.com, IR: IRobson{at}mednet.ucla.edu, KP: khumbo{at}pihmalawi.com, DO: donoya{at}heroza.org, EC: elijah{at}pihmalawi.com, KD: KDovel{at}mednet.ucla.edu, AC: augutc{at}gmail.com