Objective:Hispanic/Latinx populations in the U.S. experience higher rates of chronic disease linked to physical inactivity, yet digital health interventions remain largely inaccessible to more than 16 million Hispanic/Latinx adults with limited English proficiency. While large language models (LLMs) offer scalable personalization, their use in non-English behavioral coaching is unexplored. This study introduces MHC-Coach-ES, a Spanish-language LLM fine-tuned on the Transtheoretical Model (TTM) of behavior change. Materials and Methods:We fine-tuned Llama 3-70B-Instruct using a two-stage pipeline. First, the model was adapted to Spanish health and motivational language using a 2.21-million-token corpus. Second, it was instruction-tuned on 3,268 translated human written messages to align the model with the Transtheoretical Model (TTM) of Behavioral Change. We compared MHC-Coach-ES with Llama 3-70B-Instruct and translated human-expert messages using a forced-choice preference survey (N = 77) and blinded expert review (N = 2). Results:Spanish-speaking participants significantly preferred MHC-Coach-ES messages over translated human-expert messages (81% preference, P <0.001). Linguistic analysis showed that MHC-Coach-ES produced more temporally anchored messages than the base model (65% vs. 20%), while maintaining readability. In blinded evaluation, clinical experts rated MHC-Coach-ES higher for alignment with Transtheoretical Model stages than human-expert messages (4.83 vs. 4.38 out of 5). The base model also outperformed translated expert messages across preference and expert ratings. Conclusions:Generative AI can operationalize behavioral science frameworks in Spanish, offering a scalable approach to reducing health disparities. The strong performance of both MHC-Coach-ES and the base model highlights the promise of generative and personalized approaches over translation-based localization for theory-driven behavioral interventions.

BackgroundThe menopausal transition represents a critical period for musculoskeletal health. Progressive strength training is highly effective in mitigating musculoskeletal declines during this transition, yet fewer than 25% of women in midlife meet strength training guidelines. Scalable interventions are needed.MethodsThe Stronger trial is a 2×2+1 factorial randomized clinical trial evaluating a 12-week, remotely-delivered strength training program for women in perimenopause, followed by a 12-week maintenance phase. Assessments will occur at baseline, 6, 12, and 24 weeks. Participants (n=90) will be randomized to one of four conditions or a waitlist control, testing: (1) exercise schedule: distributed "snacks" (1–2 exercises, 3 times/day, two days/week) versus consolidated "sessions" (5 exercises, 1 time/day, two days/week); and (2) tracking simplicity: simplified versus detailed self-monitoring. All participants will receive progressive overload training with sandbag weights, weekly virtual office hours, perimenopause education, and protein intake guidance (goal: 1.6 g/kg/day). The primary outcome is 12-week adherence, measured as total exercise completed out of total prescribed (5 exercises, 3 high effort sets, 2 days per week). Secondary outcomes include maintenance-phase adherence, intervention engagement, and changes in hamstring and quadriceps strength (via handheld dynamometry), sit-to-stands, and step power. Exploratory outcomes include body composition (DXA), functional performance, and psychosocial measures.ResultsRecruitment began in June 2025, with data collection anticipated to end in summer 2026.ConclusionsStronger will identify whether breaking strength sessions into snacks improves adherence, and evaluate the feasibility and acceptable delivery strategies of a scalable digital strength training intervention for perimenopausal women.

Background Cardiovascular disease remains the leading cause of global morbidity and mortality. The original My Heart Counts smartphone application demonstrated the feasibility of large-scale, fully digital recruitment and trial conduct, but was limited by platform exclusivity and the need for human experts to create text-based behavioral interventions. Methods The next-generation My Heart Counts smartphone application is a prospective, observational cohort study with an embedded randomized crossover trial, evaluating personalized text-based coaching prompts, available in both English and Spanish. All study and trial operations will be conducted via the My Heart Counts smartphone application, re-designed using the open-source Stanford Spezi framework to support iOS, with a planned Android release in 2027. The target enrollment is N=15,000 adults across the United States and United Kingdom. The study establishes a comprehensive digital biobank by synthesizing passive mobile health data (steps, flights climbed, heart rate, sleep, workouts), raw sensor data (e.g., accelerometry), longitudinal clinical surveys, active tasks (6-minute walk test and 12-minute Cooper run test), electrocardiograms (ECG), and electronic health record (EHR) data integrated via HL7 FHIR protocols. The embedded trial evaluates the effect of text-based coaching prompts generated by a large language model (LLM) grounded in the Transtheoretical Model of Change on daily physical activity, as compared to generic prompts. Planned Analysis The primary endpoint of the randomized crossover trial is change in daily step count between LLM-driven and generic text-based intervention arms, analyzed using mixed-effects models. Secondary endpoints include change in mean active minutes and calorie burn over each intervention week. Other exploratory analyses include the changes in submaximal (6-minute walk test) and maximal (Cooper 12-minute run test) cardiorespiratory fitness, changes to sensor-derived biomarkers (e.g., sleep quality, resting heart rate, and heart rate variability), and association of sensor-derived biomarkers with EHR-confirmed clinical outcomes. Conclusions By utilizing autonomous, LLM-driven coaching, modular software design, and cross-platform accessibility, our smartphone application-based study will provide a scalable model for inclusive and decentralized preventive care of patients with cardiovascular disease. Trial Status Recruitment commenced in March 2026 and is ongoing.

BACKGROUND:The Women's Health Initiative (WHI) Strong & Healthy (WHISH) randomized trial implemented a 9-year remotely delivered "light touch" multicomponent physical activity intervention (PA-I) aimed at increasing physical activity (PA) and reducing sitting time in a large U.S. cohort of older women. METHODS:Using a randomized consent design, 49 331 women, aged 66-99 years, were assigned to Intervention (N = 24 657), 95.9% (N = 23 653) of whom provided passive consent to receive the WHISH PA-I, or "usual activity" Comparison (N = 24 674). Differences between intention-to-treat arms in self-reported walking, sitting and other PA, averaged over the course of the trial, were derived from a linear regression model with log-transformed outcomes as a function of arm, adjusted for participant selection strata. RESULTS:Pre-randomization mean age (79.7 years), total recreational PA (11.6 MET-hours/week), RAND-36 physical function score, and other characteristics did not differ between trial arms. Over a median 8.7 years, Intervention participants reported higher average walking (3.2%), aerobic (3.4%), and total (3.7%) exercise, as MET-hours/week, and strength training sessions per week (4.4%), and fewer hours/week of sitting (1.5%) than Comparison (p < .001 for all). Intervention versus Comparison differences were greater for aerobic exercise (6.92%) and sitting (-2.48%) in the youngest age group (66 - 76) compared with other age groups (p <.001 and .02 for interaction , respectively). CONCLUSION:A 9-year "light touch" remotely-delivered PA intervention modestly attenuated age-related PA decreases and sitting time increases in older women. Building on the pragmatic WHISH PA intervention to augment behavioral changes, especially in adults aged over 80 years, is an important direction for population health.

BackgroundProlonged sitting and lack of moderate to vigorous physical activity represent 2 independent risk factors for myriad poor health outcomes. The negative effects of prolonged sitting can be ameliorated with as little as 2 minutes of large muscle movement. Further, cardiovascular benefits from moderate to vigorous activity can be accumulated throughout the day in short bouts rather than require continuous long bouts. Taken together, “exercise snacks” provide a way to both interrupt prolonged sitting and accumulate moderate to vigorous physical activity during a sedentary workday. ObjectiveThis protocol describes the feasibility and acceptability pilot of MOV'D (Move Often Every Day)—a fully remote, peer-supported behavioral intervention to interrupt prolonged bouts of sitting at work with exercise snack breaks. MethodsThe MOV’D pilot study aims to recruit approximately 60-80 participants who work full-time in a sedentary occupation. Participants were randomly assigned either to the Fitbit Control or the MOV’D experimental group in a randomized 2-group design. The pilot study had a 4-week active intervention and a 4-week follow-up with assessments at baseline, 4 weeks, and 8 weeks. The Fitbit control group received a Fitbit to self-monitor their physical activity prior to receiving the intervention material at the end of the study. In addition to Fitbit, the MOV’D intervention features included a private social support group chat seeded with daily experimenter prompts, weekly 5-minute behavior change technique videos, daily self-monitoring, and daily exercise snack suggestion videos. ResultsStudy enrollment began in March 2022 and concluded in June 2022. Data collection concluded in October 2022. We enrolled 70 participants, and 68 participants completed all the study assessments. ConclusionsThis protocol integrates findings from education, behavioral sciences, sedentary behavior, and exercise physiology to promote building an exercise snack habit at work: taking short intense exercise breaks to break up prolonged sitting. The results from this pilot study will show the feasibility and acceptability of the MOV’D intervention. Trial RegistrationClinicalTrials.gov CT05360485; https://clinicaltrials.gov/study/NCT05360485

Online peer support groups (OPSGs) have been widely used in health and wellbeing interventions, but their effectiveness varies, influenced by the level of group engagement and the quality of interactions. This study explores the impact of structured elements, such as prompts and tasks, incorporated into OPSG design on peer support quality and group engagement. We analyse chat messages from two OPSGs aimed at reducing workplace sitting through exercise breaks. Despite identical study designs and similar participant demographics, the two OPSGs exhibited significant differences in group engagement, underscoring the importance of understanding group composition, organic interactions, and other factors influencing engagement. The findings emphasise the need for thoughtful OPSG design, including developing structured elements, to foster communication and improve the quality of interactions. Based on our findings, we discuss strategies to promote supportive exchange, focusing on the role of prompts, group cohesion, and social presence to enhance the effectiveness of OPSGs.

Personalized, smartphone-based coaching improves physical activity but relies on static, human-crafted messages. We introduce My Heart Counts (MHC)-Coach, a large language model fine-tuned on the Transtheoretical Model of Change. MHC-Coach generates messages tailored to an individual's psychology (their "stage of change"), providing personalized support to foster long-term physical activity behavior change. To evaluate MHC-Coach's efficacy, 632 participants compared human-expert and MHC-Coach text-based interventions encouraging physical activity. Among messages matched to an individual's stage of change, 68.0% (N=430) preferred MHC-Coach-generated messages (P < 0.001). Blinded behavioral science experts (N=2) rated MHC-Coach messages higher than human-expert messages for perceived effectiveness (4.4 vs. 2.8) and Transtheoretical Model alignment (4.1 vs. 3.5) on a 5-point Likert scale. This work demonstrates how language models can operationalize behavioral science frameworks for personalized health coaching, promoting long-term physical activity and potentially reducing cardiovascular disease risk at scale.

Personalized, smartphone-based coaching improves physical activity but relies on static, human-crafted messages. We introduce My Heart Counts (MHC)-Coach, a large language model fine-tuned on the Transtheoretical Model of Change. MHC-Coach generates messages tailored to an individual's psychology (their "stage of change"), providing personalized support to foster long-term physical activity behavior change. To evaluate MHC-Coach's efficacy, 632 participants compared human-expert and MHC-Coach interventions encouraging physical activity. Among messages matched to an individual's stage of change, 68.0% (N = 430) preferred MHC-Coach-generated messages (P < 0.001). Blinded behavioral science experts (N = 2) rated MHC-Coach messages higher than human-expert messages for perceived effectiveness (4.4 vs. 2.8) and Transtheoretical Model alignment (4.1 vs. 3.5) on a 5-point Likert scale. This work demonstrates how language models can operationalize behavioral science frameworks for personalized health coaching, showing the potential for promoting long-term physical activity and reducing cardiovascular disease risk at scale.

Background The six-minute walk test (6MWT) is a prognostic sub-maximal exercise test used clinically as a measure of functional capacity. With the emergence of advanced sensors, 6MWTs are being performed remotely via smartphones and other devices. The My Heart Counts Cardiovascular Health Study is a smartphone application that serves as a digital platform for studies of human cardiovascular health, and has been used to perform 30,475 6MWTs on 8922 unique participants. Objective As our 30,475 6MWTs represent the largest such collection of data available, we sought to identify associations with measured demographic and clinical variables with 6MWT distance at enrollment and separately determine if use of the My Heart Counts smartphone application led to changes in 6MWT distance. Methods and results We present the public data release of our 30,475 6MWTs and the launch of a webpage-based data viewer of summary-level statistics, to compare the functional capacity of an individual by their age, gender, height, weight, and disease status (https://mhc-6mwts.streamlit.app). Using multivariable regression, we report associations of demographic and clinical variables with baseline 6MWT distance (N = 3606), validating prior associations with age, male gender, height, and baseline physical activity level with 6MWT distance. We also report associations of 6MWT baseline distance with employment status (+12.4 m ±4.9 m, P = 0.011) and feeling depressed (−3.65 m, ±0.79 m, P < 0.001). We separately found that cardiovascular disease status was significantly associated with decreased 6MWT distance for atrial fibrillation (−24.9 m ±7.8 m, P = 0.0013), peripheral artery disease (−41.7 m ±12.5 m, P < 0.001), and pulmonary arterial hypertension (−76.3 m ±24.8 m, P = 0.0022). Heart failure was associated with decreased 6MWT distance but was not statistically significant (−25.5 m ±14.5 m, P = 0.078). In a subset of participants who conducted repeat 6MWTs separated by at least 1 week but no greater than 3 months (N = 1129), we found that use of the My Heart Counts app was associated with a statistically significant increase in 6MWT distance (+17.5 m ±7.85 m, P < 0.001). Conclusions We validate previously identified associations from clinic-performed 6MWTs, demonstrating the utility of a mobile method in collecting 6MWT data for clinicians and researchers. We also demonstrate that use of the My Heart Counts app is associated with small, but significant increases in 6MWT distance. Given the importance of 6MWTs in assessment of functional capacity, our publicly-available data will serve an important purpose as a health and disease-specific reference for investigators worldwide.

Purpose'Exercise snacks'- brief, intermittent sessions of moderate-to-high intensity activity- offer a novel and promising solution for sedentary office workers. However, perceptions and experiences related to engaging in this approach are limited or largely unknown.Approach and ParticipantsThis qualitative study employed semi-structured interviews with 20 office workers who completed a pilot exercise snack intervention called Move Often eVeryDay (MOV'D). The sample consisted of 10 participants from the MOV'D intervention group and 10 participants from the self-monitoring group.SettingInterviews were conducted remotely via Zoom.MethodsInterview transcripts were analyzed using open-coded, thematic analysis, with themes and subthemes identified both a priori and inductively during the analysis.ResultsThe mean (SD) age of participants was 43.8 (11.3) years, 80% identified as women or transgender women, and 35% reported being Non-Hispanic White. Emergent themes included enjoyment of study components, increased motivation and physical activity awareness, and knowledge gained. Common barriers included time constraints and competing obligations. Facilitators of engagement included behavioral adaptations and the minimal space needed for activities. Key suggestions for future iterations included having more reminders throughout the study, enhanced social connection, and adjustments to intervention materials.ConclusionAlthough exercise snacks are brief and accessible, findings highlight the need to refine the intervention to reduce participation barriers and support long-term habit formation. The insights garnered from the present pilot study offer actionable recommendations for enhancing this and other similar health promotion behavioral interventions.Trial registrationThe trial was registered on May 3, 2022 (before recruitment began) at https://clinicaltrials.gov/study/NCT05360485, clinical trial number: NCT05360485.

Physical activity has been shown to improve health outcomes, decrease risk of medical comorbidities, and improve longevity. Despite these known benefits, most individuals do not meet the recommended national exercise guidelines for physical activity. This paper identifies individual and community-level factors impacting engagement in physical activity, thus focusing on solutions to address common barriers to exercise and optimize diversity, equity, and inclusion initiatives for physical activity. Afterward, several cases within one large academic institution and its affiliates illustrate recent endeavors to improve access to physical activity and highlight future opportunities for growth in all communities.

Many modern classrooms impose constraints on student movement, both physical (e.g. stable, upright chairs) and psychological (e.g. norms that "being still" signals paying attention).. We conducted two separate studies on 32 6th grade (11-12 year old) and 43 7th grade (12-13 year old) students, respectively. Study 1 was a within-subject design and compared students' creative ideation on the Alternate Uses Test and a verbal memory task through two conditions: Sit Still (on a regular chair with instructions to not move) or Freedom to Move (on a wiggle stool with ability to move as they wanted). There was a significant effect of condition on novel ideas, as well as number of novel ideas / total ideation (creativity ratio). There was no significant difference by condition on the verbal memory task. Study 2 was a within-subject design (due to the COVID-19 pandemic, each student completed 2 of 3 conditions): Sit Still, Freedom to Move, and Sit as Usual (on a regular chair with instructions to sit as if they were paying attention in class) and compared students' creative ideation and a focused attention task instead of Study 1's memory task. There was a significant effect of condition on novel ideas and creativity ratio, but pairwise comparisons were not statistically significant. There was no condition effect on focused attention. The Freedom to Move effects appear selective only for creative ideation. Implications for future research and small classroom changes to support natural movement are discussed.

Prolonged sitting and lack of moderate to vigorous physical activity represent two independent risk factors for myriad poor health outcomes. The negative effects of prolonged sitting can be ameliorated with as little as 2 minutes of large muscle movement. Further, cardiovascular benefits from moderate to vigorous activity can be accumulated throughout the day in short bouts rather than require continuous long bouts. Taken together, “exercise snacks” provide a way to both interrupt prolonged sitting and accumulate moderate to vigorous physical activity during a sedentary work day. The current protocol describes the feasibility and acceptability pilot of Move Often eVery Day (MOV'D), a fully-remote, peer-supported behavioral intervention to interrupt prolonged bouts of sitting at work with exercise snack breaks. The MOV’D pilot study aimed to recruit approximately 60-80 participants who work full-time in a sedentary occupation. Using a randomized 2-group design, participants were randomly assigned either to the “Fitbit Control'' or the “MOV’D” experimental group. The pilot study had a 4-week active intervention and a 4-week follow-up with assessments at baseline, 4-weeks, and 8-weeks. The Fitbit control group received a Fitbit to self-monitor their physical activity prior to getting intervention material at the end of the study. In addition to the Fitbit, the MOV’D intervention features included: a private social support group chat seeded with daily experimenter prompts; weekly, 5-minute behavior change technique videos; daily self-monitoring, and daily exercise snack suggestion videos. Study enrollment began in March 2022 and concluded in June 2022. Data collection concluded in October 2022. We enrolled 70 participants and 68 participants completed all study assessments. This protocol integrates findings from education, behavioral sciences, sedentary behavior, and exercise physiology to promote building an exercise snack habit at work: taking short, intense exercise breaks to break up prolonged sitting. Results from this pilot study will assess the feasibility and acceptability of the MOV’D intervention. NCT05360485

OBJECTIVES:Tobacco use disproportionately affects Alaska Native people. Physical activity may aid quitting smoking and provides health benefits. We tested telemedicine-delivered heart health interventions in Alaska's Norton Sound region. METHODS:Alaska Native adults (N = 299, 51.5 % male, 60.5 % Inupiaq) with hypertension and/or hypercholesterolemia who smoked daily were randomized to intervention on smoking and physical activity (group 1) or traditional diet and medication adherence (group 2). Intention to change was not required for participation. Stage-tailored mailed workbooks and personalized reports were supported by telehealth counseling at baseline, 3, 6, and 12 months. Study outcomes were assessed at baseline, 3-, 6-, 12-, and 18-months (i.e., 6-months after the final counseling session). Smoking outcomes were self-reported 7-day point prevalence abstinence (7d-PPA),1 bioconfirmed with urine anabasine; 24-hour quit attempts; and 50 % reduction in smoking. Self-reported physical activity outcomes were metabolic equivalent of task (MET) minutes and meeting moderate-to-vigorous physical activity (MVPA) guidelines. RESULTS:At baseline, participants averaged 12.4 (SD = 10.0) cigarettes/day, with 19.4 % prepared to quit smoking, and 81.6 % meeting MVPA guidelines. During the study, most (70.2 % group 1; 63.5 % group 2) reported a 24-hr quit attempt (p = 0.219), and Group 1 (53.6 %) was more likely than Group 2 (28.4 %) to use nicotine replacement therapy (NRT), OR = 2.92, p < 0.001. At 18-months, 40.5 % (group 1) and 32.5 % (group 2) had reduced their smoking by half or more (p = 0.343), and 10.8 % vs. 7.9 % (group 1 vs. 2) reported 7d-PPA with 4 % vs. 6 % (group 1 vs. 2) bioconfirmed. Time and baseline stage of change predicted 7d-PPA (p's≤.015), with no group effect (p = 0.325). Activity levels did not significantly differ by group or time. CONCLUSIONS:Telemedicine counseling supported NRT use but did not significantly affect behavioral outcomes.