"Effectiveness of Supervised and Unsupervised mHealth-based Exercise Interventions on Clinical Outcomes in Adults With Type 2 Diabetes and Obesity Within a Primary Care Context"

Overview

Type 2 diabetes (T2D) and obesity are two of the most common chronic conditions worldwide, and they frequently occur together. Both conditions are closely linked to physical inactivity, which is currently recognized as a major risk factor for a wide range of diseases. Although structured exercise programs have shown benefits for people living with these conditions, maintaining long-term participation remains a significant challenge. The rise of mobile health (mHealth) technologies - such as smartphone apps and wearable devices - offers a promising new way to support people in becoming and staying more physically active. This study aims to compare the effects of two different 20-week exercise intervention programs versus standard medical care in adults aged 30 to 65 who have been diagnosed with both T2D and obesity. Participants will be randomly assigned to one of three groups: Group 1 (Supervised Training): Participants will attend two in-person, supervised exercise sessions per week at a local sports facility, each lasting 60 minutes. Sessions will be led by a qualified trainer and will include strength, flexibility, and functional exercises. Participants will also receive weekly motivational text messages and group health coaching. Group 2 (Unsupervised mHealth Training): Participants will follow the same exercise plan independently using a mobile application (ActiVital), which is part of the Andalusian Physical Activity Prescription Plan (PAPEF). They will receive their weekly training schedule through the app and will also receive text messages and group health coaching. Group 3 (Control Group): Participants will continue with their usual medical care without any structured exercise program. All participants across the three groups will be provided with a Fitbit Charge 6 wearable device to monitor daily physical activity and health indicators. The study's central hypothesis is that both structured supervised exercise and app-guided unsupervised exercise will lead to meaningful improvements in clinical indicators - such as blood glucose levels, glycated hemoglobin (HbA1c), blood lipids, and blood pressure - as well as in body composition, physical fitness, sleep quality, and overall quality of life, when compared to standard care alone. It is also hypothesized that the mHealth approach will show particular benefits in terms of adherence, self-management, and scalability. A total of 126 participants will be recruited through the Primary Care Center of Osuna (Seville, Spain), with measurements taken before and after the 20-week intervention period. The study will help determine which type of exercise program is most effective and feasible for people managing both type 2 diabetes and obesity in a primary care setting.

Background and Rationale Type 2 diabetes (T2D) and obesity represent a growing global health burden, both individually and as comorbid conditions. Physical inactivity is recognized as the fourth leading cause of death worldwide and is directly implicated in the development and progression of both pathologies. While lifestyle interventions combining exercise and behavioral change have demonstrated efficacy in improving glycemic control and metabolic profile, their long-term effectiveness remains limited due to poor adherence and lack of scalability. Mobile health (mHealth) interventions - delivered via smartphones, wearable devices, and digital platforms - have emerged as a promising complementary strategy to extend the reach of structured exercise programs beyond traditional clinical settings. Despite a growing body of evidence supporting mHealth for T2D and obesity independently, no study to date has rigorously evaluated the comparative effectiveness of supervised versus app-guided unsupervised exercise interventions in patients presenting with both conditions concurrently. Study Design This is a 20-week, three-arm, parallel-group randomized controlled trial (RCT) conducted at the Primary Care Center of Osuna, Seville, Spain. Exercise sessions will take place at the covered indoor facilities of the Municipal Sports Center of Osuna. The trial will compare two active intervention arms against a usual care control group. Randomization will be performed using computer-generated block randomization (variable block sizes of 4-10) via Research Randomizer (https://www.randomizer.org), stratified by sex and baseline glycated hemoglobin (HbA1c) level. Allocation concealment will be ensured by maintaining the randomization list with a researcher not involved in outcome assessments. The allocation sequence will remain concealed until participants are formally enrolled and assigned. External evaluators, blinded to group allocation, will conduct all pre- and post-intervention assessments. Intervention Description All participants across the three groups will receive a Fitbit Charge 6 wearable device for continuous monitoring of physical activity indicators, and will attend two group health coaching sessions (weeks 4 and 12) focused on lifestyle habits relevant to T2D and obesity management (diet, physical activity, sleep). All groups will also receive two weekly motivational text messages encouraging daily use of the wearable device and reminding participants of scheduled assessment appointments. Experimental Group 1 (EG-1) - Supervised In-Person Training Participants will complete 40 supervised exercise sessions over 20 weeks (2 sessions/week, Tuesdays and Thursdays), each lasting 60 minutes, at the Municipal Sports Center of Osuna. The exercise prescription follows the guidelines of the American Diabetes Association (ADA), the American College of Sports Medicine (ACSM), and the World Health Organization (WHO) for individuals with T2D and obesity. Each session is structured as follows: Warm-up (10 min): joint mobility, marching in place, dynamic stretching Main block (40 min): resistance training targeting major muscle groups Cool-down (10 min): static stretching, breathing exercises, relaxation The 20-week program is organized into four progressive phases: Phase 1 - Adaptation (Sessions 1-8): Muscular activation and technical learning. 2-3 sets × 10-12 repetitions. Exercises: chair squat, chest press with resistance band, seated row with band, lateral raises, biceps curl, triceps extensions, knee plank. Phase 2 - Initial Progression (Sessions 9-20): Strength-endurance development. 3 sets × 10-15 repetitions. Exercises: free squat, deadlift with resistance band, wall push-ups, glute bridge, military press, inclined row, lateral plank with knee. Phase 3 - Intermediate Progression (Sessions 21-32): Increased load and postural control. 3-4 sets × 8-12 repetitions. Exercises: assisted lunges, double deadlift, bench chest press, kneeling row, weighted squat, calf raises, core exercises. Phase 4 - Consolidation (Sessions 33-40): Functional circuits. 2-3 rounds per session. Circuit 1: squat + military press + plank. Circuit 2: lunges + band row + crunch. Every Minute on a Minute (EMOM) formats and progressive repetition ladders are incorporated. During each session, heart rate (Fitbit Charge 6), Rating of Perceived Exertion (RPE), and blood oxygen saturation (pulse oximeter) will be recorded. A standard leaflet with recommendations for autonomous aerobic activity on non-training days will be provided at baseline. Experimental Group 2 (EG-2) - Unsupervised mHealth Training Participants will follow the identical exercise program as EG-1, delivered autonomously through the ActiVital mobile application - the official digital platform of the Andalusian Physical Activity and Exercise Prescription Plan (PAPEF, 2023-2030). Participants will access their individualized weekly training plan via the app and will confirm session completion directly within the platform after each workout. Basic training and onboarding on app usage will be provided at study initiation. Control Group (CG) Participants will continue their usual medical care as prescribed by their primary care physician, without any structured exercise intervention. They will receive the wearable device and text messages consistent with the other groups but will not receive an exercise plan. Outcome Assessment Protocol All participants will attend two assessment visits (pre-intervention: weeks 0-1; post-intervention: weeks 20-21), each consisting of two separate appointments spaced 9 days apart. Visit Day 1: Delivery of informed consent documentation and health questionnaires; distribution of ActiGraph GT9X Link+ accelerometers (worn on the wrist for 7 consecutive days); distribution of Fitbit Charge 6; app installation and onboarding for EG-2. Visit Day 2 (9 days later): Blood analysis, physical fitness assessments, body composition measurements; questionnaire completion; accelerometer retrieval. Measurement Instruments and Procedures Clinical Variables: Fasting blood glucose will be measured using an Accu-Chek Inform II glucometer. HbA1c, full lipid panel (total cholesterol, Low Density Lipoprotein (LDL), High Density Lipoprotein (HDL) and triglycerides), C-reactive protein (CRP), creatinine, estimated glomerular filtration rate (eGFR), and hepatic transaminase levels (GOT/GPT) will be obtained via venous blood draw following a 9-12 hour fast. Blood pressure (systolic and diastolic) will be assessed using an OMRON sphygmomanometer and Littmann stethoscope. Medication use (initiation, modification, or discontinuation) will be tracked throughout the entire 20-week period. Body Composition: Body mass (kg), BMI (kg/m2)(calculated from height and weight), body fat percentage, and muscle mass will be assessed via bioelectrical impedance analysis (BIA) using an InBody 770 device (Biospace Co., Ltd., Seoul, South Korea). Waist and hip circumferences will be measured using a standard anthropometric tape measure. Physical Fitness: Aerobic capacity: 6-Minute Walk Test (6MWT), measuring total distance covered on a regulation athletics track. Lower limb strength: 30-second Sit-to-Stand Test (30-STS), counting full stands completed in 30 seconds with arms crossed over the chest. Explosive lower limb power: Countermovement Jump (CMJ), measured using an optical measurement system (Optojump, Microgate S.R.L., Bolzano, Italy), recording jump height (cm), power output (W), and take-off velocity (m/s). Functional mobility and fall risk: Timed Up and Go test (TUG), measuring time (seconds) to stand from a chair, walk 3 meters, turn, and return to seated position; and 6-Meter Walk Test, measuring gait speed (m/s) using photoelectric cells. Resting and active heart rate: continuously monitored via the Fitbit Charge 6 wearable device. Sleep Quality and Daily Physical Activity: Both outcomes will be assessed using an ActiGraph GT9X Link+ accelerometer worn on the wrist for 7 consecutive days at each measurement point. Variables recorded include total sleep time, sleep efficiency, sleep latency, wake after sleep onset, sedentary time, and time spent in light, moderate, and vigorous physical activity, estimated energy expenditure (kcal), and daily step count. Psychological Well-being and Health-Related Quality of Life: Assessed using the SF-36 Health Survey (Short Form-36; McHorney et al., 1994), a validated generic instrument measuring eight domains: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. Each domain is scored 0-100, with higher scores indicating better health status. Adherence: In EG-1, adherence will be tracked via manual attendance records. In EG-2, session completion will be logged directly within the ActiVital app. Wearable device usage will be monitored continuously across all groups throughout the 20-week period. Statistical Analysis A descriptive analysis of all variables will be performed using means and standard deviations (or medians and interquartile ranges for non-parametric data), and frequencies and percentages for categorical variables. Data normality will be assessed using Kolmogorov-Smirnov or Shapiro-Wilk tests. Between-group comparisons will be conducted using ANOVA or ANCOVA with repeated measures, adjusting for relevant covariates (age, sex, baseline fitness level). Non-parametric alternatives (Kruskal-Wallis, Friedman test) will be applied where normality assumptions are not met. Intra-group pre-to-post differences will be assessed using paired t-tests or Wilcoxon signed-rank tests as appropriate. Post-hoc analyses will use the Scheffé test. Effect size will be calculated using Cohen's d or partial eta-squared (η²), interpreted as: small (\<0.25), moderate (0.25-0.40), or large (\>0.40) per Cohen (1973). Pearson or Spearman correlation analyses will examine associations between improvements in clinical variables and physical activity adherence. Multiple regression models will be used to identify predictors of response to the exercise programs. Statistical significance will be set at p \< 0.05. All analyses will be disaggregated by sex where possible. Statistical analyses will be performed using SPSS v.25 (IBM, Portsmouth, UK). Data will be stored and managed through REDCap (Research Electronic Data Capture), a secure, web-based platform for research data collection, with access restricted to authorized study investigators. Sample Size Justification Sample size was calculated using G\*Power 3.1, based on an a priori ANCOVA with fixed effects. Using a large effect size (f = 0.80) for HbA1c reported in a comparable study (Yin et al., 2022), with α = 0.05, power = 95%, 3 groups, 1 covariate, and 10 numerator degrees of freedom, a minimum of 42 participants per group was determined. Accounting for an estimated 10% dropout rate, the target sample is 47 participants per group, yielding a total of 141 participants. After excluding non-eligible individuals, 126 participants will be enrolled (42 per group). Data Management and Ethical Considerations Data will be collected and managed in accordance with the EU General Data Protection Regulation (GDPR) and applicable Spanish legislation. Personal data will be pseudonymized using a dissociation-pseudonymization process (reversible coding). Anonymized final datasets will be stored in idUS, the University of Seville's open-access repository, with unique Digital Object Identifiers (DOI). Data will be retained for 10 years following study completion, after which it will be securely destroyed. This study has been approved by the Ethics Committee for Research of the University Hospitals Virgen Macarena-Virgen del Rocío (approval code: SICEIA-2026-000343; favourable opinion issued on 26 May 2026), in accordance with the principles of the Declaration of Helsinki and applicable Spanish biomedical research legislation (Ley 14/2007, of 3 July). All participants will provide written informed consent prior to enrolment (informed consent version V1, dated 29 January 2026).