John D. Piette

BACKGROUND Mental health services in Latin America face shortages of trained professionals, and access to expertise in evidence-based communication skills remains severely limited. In particular, motivational interviewing (MI) "reflections" are a foundational communication skill for empathic listening, but opportunities for individualized practice with feedback are limited. Digital resources supported by artificial intelligence (AI) may expand opportunities for skill development. OBJECTIVE To evaluate an AI-supported online training tool designed to improve the quality of MI reflections among psychologists, psychology students, and community-based providers in Honduras, and to characterize participants’ experiences with the training. METHODS We conducted a longitudinal single-group study among 56 Spanish-speaking adults in Honduras, including 22 psychologists, 11 psychology students, and 23 community leaders working in mental health service roles. Participants completed four rounds of online exercises in which they generated reflective responses to 10 client statements and received AI-generated scores ranging from 0 to 1, along with examples of high-quality responses. The scoring model used the Prompt-Aware Margin Ranking (PAIR) framework, a RoBERTa-based cross-encoder trained using ordinal categories identifying "skilled" responses, "acceptable" responses, and "MI-inconsistent" responses. Participants also received a live booster session with a psychologist between the second and third rounds. Changes in reflection scores were examined longitudinally, and differences in performance among the three participant groups were assessed. Semi-structured interviews were conducted with 12 purposively selected participants to identify perceived benefits, limitations, and recommendations for improving the intervention. RESULTS All participants completed all four rounds, contributing 4480 reflection responses. Mean scores increased across rounds (P<.001). Improvements in scores were observed among psychologists (P=.009), psychology students (P<.001), and community leaders (P<.001). The proportion of responses classified as MI-skilled increased from 26.3% in the first round to 48.2% in the fourth (P<.001), while MI-inconsistent responses decreased from 42.4% to 22.7% (P<.001). Improvements were greatest among community leaders, whose skilled reflections increased from 5.4% to 37.7%. The live booster was associated with significant score improvements over-and-above the linear trend (P<.001), with notable benefit among community leaders (P=.002). Interview participants generally described the tool as engaging, useful, and helpful for translating knowledge about reflective listening into practice. Participants valued both the automated feedback and live booster session, while recommending greater cultural and contextual adaptation of scenarios to Honduran experiences. CONCLUSIONS An AI-supported training platform combined with brief human reinforcement was associated with substantial short-term improvements in the quality of MI reflections among psychologists, psychology students, and community-based providers in Honduras. The largest gains occurred among participants without formal training in clinical psychology, suggesting that the tool could improve the quality of interactions in environments relying on paraprofessionals. Future controlled studies should determine whether improvements are sustained and translate into changes in communication with clients and subsequent clinical outcomes. CLINICALTRIAL Not applicable

Background: Mobile health (mHealth) apps are widely available, and some have proven safe and effective for management of specific chronic conditions. Despite a high degree of interest, the potential of these technologies has yet to be realized. Patient and clinician attitudes are key factors that influence the adoption of mHealth apps but remain poorly understood, particularly in the United States. Objective: This study aimed to identify both patient and clinician attitudes that can influence recommending and adopting mHealth apps. Methods: Using well-established technology adoption and implementation science frameworks, this study included a deductive content analysis using a rapid qualitative analytic method. Semistructured interviews were conducted with patients and clinicians to identify technical and material, social and personal, and policy and organizational factors that can influence the recommendation or adoption of mHealth apps. The interviews and data analysis were performed between September 2023 and August 2024. Results: Participants included 20 clinicians (n=12, 60% general internists) with a mean time in practice of 17 (SD 11.6) years, and 28 patients with a mean age of 59 (SD 12.1) years. A total of 7 categories related to patients’ and clinicians’ attitudes toward mHealth apps emerged: (1) apps as tools to improve health by extending care, (2) the role of apps in enhancing the patient–clinician relationship, (3) the need for simplicity and efficiency in app design, (4) the influence of prior experience with mHealth apps, (5) comfort with technology, (6) recommendations from trusted sources, and (7) education and hands-on experience. Although similar factors were considered by patients and clinicians, their views about older adults’ interest and ability to use mHealth apps differed. Conclusions: Understanding patient and clinician views about mHealth apps provides critical insights for developing approaches to facilitate their use. These findings suggest patients and clinicians share similar views about the benefits of mHealth apps. Nonetheless, clinicians’ perceptions about older patients’ interest and ability to use mHealth apps may negatively impact recommendation of mHealth apps and subsequent adoption by older adults.

Cultural and language factors significantly influence counseling, but Natural Language Processing research has not yet examined whether the findings of conversational analysis for counseling conducted in English apply to other languages. This paper presents a first step towards this direction. We introduce MIDAS (Motivational Interviewing Dataset in Spanish), a counseling dataset created from public video sources that contains expert annotations for counseling reflections and questions. Using this dataset, we explore language-based differences in counselor behavior in English and Spanish and develop classifiers in monolingual and multilingual settings, demonstrating its applications in counselor behavioral coding tasks.

Aylin Ece Gunal, Bowen Yi, John D. Piette, Rada Mihalcea, Veronica Perez-Rosas. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers). 2025.

Background: Cardiometabolic conditions including acute coronary syndrome (ACS) and type 2 diabetes (T2D) require comprehensive care and patient engagement in self-care behaviors, and the drivers of those behaviors at the individual and health system level are still poorly understood. Objective: We aim to gain insights into self-care behaviors of individuals with cardiometabolic conditions. Methods: A convenience sample of 98 adult patients with ACS and T2D was recruited in the United States, Germany, and Taiwan to participate in a mixed methods study using ethnographic methods. All participants completed 7-day web-based diaries tracking their level of engagement, and 48 completed 90-minute web-based semistructured interviews between February 4, 2021, and March 27, 2021, focusing on themes including moments of engagement. Qualitative analysis identified factors influencing self-care practices and a Patient Mind States Model prototype. Results: Patient reports indicate that many patients feel social pressure to adhere to treatment. Patients’ experience can be understood within 5 categories defined in terms of their degree of engagement and adherence (“ignoring,” “struggling,” “juggling,” “controlling,” and “reframing”). Conclusions: For people living with ACS and T2D, the self-care journey is defined by patterns of patient experiences, which can identify areas that tailored digital health care interventions may play a meaningful role.

Background: Hypoglycemia remains a challenge for roughly 25% of people with type 1 diabetes (T1D) despite using advanced technologies such as continuous glucose monitors (CGMs) or automated insulin delivery systems. Factors impacting hypoglycemia self-management behaviors (including reduced ability to detect hypoglycemia symptoms and unhelpful hypoglycemia beliefs) can lead to hypoglycemia development in people with T1D who use advanced diabetes technology. Objective: This study aims to develop a scalable, personalized mobile health (mHealth) behavioral intervention program to improve hypoglycemia self-management and ultimately reduce hypoglycemia in people with T1D who use advanced diabetes technology. Methods: We (a multidisciplinary team, including clinical and health psychologists, diabetes care and education specialists, endocrinologists, mHealth interventionists and computer engineers, qualitative researchers, and patient partners) jointly developed an mHealth text messaging hypoglycemia behavioral intervention program based on user-centered design principles. The following five iterative steps were taken: (1) conceptualization of hypoglycemia self-management processes and relevant interventions; (2) identification of text message themes and message content development; (3) message revision; (4) patient partner assessments for message readability, language acceptability, and trustworthiness; and (5) message finalization and integration with a CGM data–connected mHealth SMS text message delivery platform. An mHealth web-based SMS text message delivery platform that communicates with a CGM glucose information-sharing platform was also developed. Results: The mHealth SMS text messaging hypoglycemia behavioral intervention program HypoPals, directed by patients’ own CGM data, delivers personalized intervention messages to (1) improve hypoglycemia symptom detection and (2) elicit self-reflection, provide fact-based education, and suggest practical health behaviors to address unhelpful hypoglycemia beliefs and promote hypoglycemia self-management. The program is designed to message patients up to 4 times per day over a 10-week period. Conclusions: A rigorous conceptual framework, a multidisciplinary team (including patient partners), and behavior change techniques were incorporated to create a scalable, personalized mHealth SMS text messaging behavioral intervention. This program was systematically developed to improve hypoglycemia self-management in advanced diabetes technology users with T1D. A clinical trial is needed to evaluate the program’s efficacy for future clinical implementation.

Although Amazon Mechanical Turk facilitates the quick surveying of a large sample from various demographic and socioeconomic backgrounds, it may not be an optimal platform for obtaining reliable diabetes-related information from the online type 1 diabetes population.

Background: Some patients prescribed opioid analgesic (OA) medications for pain experience serious side effects, including dependence, sedation, and overdose. As most patients are at low risk for OA-related harms, risk reduction interventions requiring multiple counseling sessions are impractical on a large scale. Objective: This study evaluates whether an intervention based on reinforcement learning (RL), a field of artificial intelligence, learned through experience to personalize interactions with patients with pain discharged from the emergency department (ED) and decreased self-reported OA misuse behaviors while conserving counselors’ time. Methods: We used data representing 2439 weekly interactions between a digital health intervention (“Prescription Opioid Wellness and Engagement Research in the ED” [PowerED]) and 228 patients with pain discharged from 2 EDs who reported recent opioid misuse. During each patient’s 12 weeks of intervention, PowerED used RL to select from 3 treatment options: a brief motivational message delivered via an interactive voice response (IVR) call, a longer motivational IVR call, or a live call from a counselor. The algorithm selected session types for each patient each week, with the goal of minimizing OA risk, defined in terms of a dynamic score reflecting patient reports during IVR monitoring calls. When a live counseling call was predicted to have a similar impact on future risk as an IVR message, the algorithm favored IVR to conserve counselor time. We used logit models to estimate changes in the relative frequency of each session type as PowerED gained experience. Poisson regression was used to examine the changes in self-reported OA risk scores over calendar time, controlling for the ordinal session number (1st to 12th). Results: Participants on average were 40 (SD 12.7) years of age; 66.7% (152/228) were women and 51.3% (117/228) were unemployed. Most participants (175/228, 76.8%) reported chronic pain, and 46.2% (104/225) had moderate to severe depressive symptoms. As PowerED gained experience through interactions over a period of 142 weeks, it delivered fewer live counseling sessions than brief IVR sessions (P=.006) and extended IVR sessions (P<.001). Live counseling sessions were selected 33.5% of the time in the first 5 weeks of interactions (95% CI 27.4%-39.7%) but only for 16.4% of sessions (95% CI 12.7%-20%) after 125 weeks. Controlling for each patient’s changes during the course of treatment, this adaptation of treatment-type allocation led to progressively greater improvements in self-reported OA risk scores (P<.001) over calendar time, as measured by the number of weeks since enrollment began. Improvement in risk behaviors over time was especially pronounced among patients with the highest risk at baseline (P=.02). Conclusions: The RL-supported program learned which treatment modalities worked best to improve self-reported OA risk behaviors while conserving counselors’ time. RL-supported interventions represent a scalable solution for patients with pain receiving OA prescriptions. Trial Registration: Clinicaltrials.gov NCT02990377; https://classic.clinicaltrials.gov/ct2/show/NCT02990377