The study will be a randomized controlled single-blind parallel group study, conducted at a single center, and registered in the Clinical Trials Registry (NCT06048757). The protocol includes all elements from the Clinicaltrials.gov registry platform. This protocol is developed in accordance with the SPIRIT guidelines for randomized controlled trials (RCTs) (26).

The Pediatric Endocrinology Unit at the University Hospital of Navarra, in collaboration with Navarrabiomed, located in Pamplona, Spain, is currently conducting this pragmatic trial. For additional information, please refer to the Clinical Trials Registry: NCT06048757.

Children and adolescents of both sexes, aged 8-18 years diagnosed with T1DM, will be recruited as participants from the Pediatric Endocrinology Unit at University Hospital of Navarra (Pamplona, Spain). Participants will be eligible to be part of the study if they meet the following inclusion criteria: a willingness to participate in the intervention, proficiency in the Spanish language, and a minimum of six months having passed since the diagnosis. The exclusion criteria include any comorbidity that limits the capacity to participate in physical activity or an inadequate understanding of the Spanish language. Additionally, participants will be excluded if they lack an internet connection, do not own a smartphone or tablet, or do not have the ability to use the application.

The Diactive-1 Study is a 24 weeks smartphone intervention with the aim of improving daily insulin dose requirements, glucose control management, adherence to resistance training, and compliance with PA guidelines recommendations (10, 27, 28) for children and adolescents with T1DM. The Diactive-1 App has been developed to be compatible with both IOS and Android smartphones. A screenshot of the Diactive-1 App is displayed in Figure 1 .

The Diactive-1 App was developed by a team of researchers with expertise in PA and T1DM. Intervention includes evidence-based recommendations at management of glucose control and physical exercise for children and adolescent with T1DM (10, 27, 28).

The Diactive-1 App consists of an automated program designed to offer evidence-based guidance for creating exercise training sessions (10, 27, 28). The sessions are tailored based on the individual’s physical fitness level (assessed beforehand), glucose levels, and glucose trend arrow at the moment. Furthermore, the Diactive-1 App can be integrated with CGM Freestyle 2 devices to display glucose levels and trend arrows prior to the commencement and at the end of the session. If patients use a different device for CGM (e.g., MiniMedTM 780G, Medtronic), the application will ask them to manually enter their glucose levels and trends before and after the exercise session.

Educational glucose monitoring is a feature included in the Diactive-1 App. This feature presents messages after users input their glucose levels and trend arrow. Depending on these two parameters and, in special cases, patients may also be prompted to input their ketone levels. In accordance with the position statement from the European Association for the Study of Diabetes (EASD) and the International Society for Pediatric and Adolescent Diabetes (ISPAD) (27), the Diactive-1 app sends advisory messages to the patient based on their present condition. For instance, if their blood glucose exceeds 330 mg/dL regardless of ketone levels, the app recommends refraining from exercise, suggests correcting glucose levels with insulin, and proposes attempting exercise again after a 30-minute interval.

The Diactive-1 App incorporates a gamification concept. Each patient starts at player level 1, which increases based on the number of training sessions completed. As previously mentioned, each patient must complete three sessions per week (mandatory) with a maximum of seven sessions (including the three mandatory ones). Completing each of the mandatory sessions’ rewards participants with 20 experience points. Additional sessions beyond the mandatory three yield 30 experience points each. Upon accumulating 100 experience points, they advance to the next level. To continue leveling up, they must accumulate another 100 experience points. Furthermore, the Diactive-1 App includes a ranking system that positions patients based on their player levels. This feature aims to encourage healthy competition and promote patient adherence to using the Diactive-1 App.

The interventions offered by the Diactive-1 App can take place at locations chosen by the participants, such as their homes, parks, or schools. A face-to-face session will be conducted before commencing the intervention to ensure that participants are familiar with the fundamental movements, thus reducing the risk of potential muscle injuries. Each training session is designed to last between 13 and 33 minutes. The level of fitness—low, medium, or high—determines the number of exercises within the sessions: four exercises for those with low fitness and five for those with medium and high fitness levels. Reference values for handgrip strength in European children and adolescents aged 6 to 18 will be employed (29). Using the median percentile of the handgrip strength, individuals falling below the 20th percentile will be classified as having low fitness levels, those between the 21st and 79th percentiles will be considered to have moderate fitness levels, and those at or above the 80th percentile will be categorized as having high fitness levels, with consideration for their sex and age. The training sessions consist of three types: equipment-based training (utilizing resistance bands and an aqua ball, which will be provided to the participants before the intervention), equipment-free training (bodyweight exercises), and partner-assisted training. Participants will be guided through their workouts by a 3D Avatar displayed on their smartphone screen. This Avatar will show the exercises for the training session, offering visual cues for correct movements ( Figure 2 ). Additionally, a background voice (narrator) will provide verbal instructions, including the start and end of each set, transitions to the other side (in the case of unilateral exercises), and designated rest periods.

The Diactive-1 App’s training program will include performing 3 to 4 sets, each comprising 6 to 12 repetitions. When training with equipment, the weight level of the aquaball and the resistance color band will vary based on the participant’s fitness level and age group: 8-12 for children and ≥13 for adolescents. However, training without equipment will follow the same prescribed regimen for both children and adolescents. Progression in weight level (aquaball) is calculated based on the average body weight of the longitudinal Diactive-1 project participants (approximately 60 kg). This progression is tailored to specific body segments (upper limb, lower limb, and core) and fitness levels. The exercises are categorized into three main muscle groups: upper body, lower body, and core. Examples of exercises include bench press, triceps press, squat, leg extensions, plank, and sit-ups. In exceptional cases, if the Diactive-1 App determines that performing resistance training would be counterproductive (for instance, when glucose levels range from 271 to 330 mg/dL with a trend arrow indicating an increase, diagonal increase, or to the right, and ketone levels are ≤ 1.5 mM before the exercise session), the session will exclusively consist of aerobic exercise. In such situations, the patient can return to the app and proceed with their strength training session later, provided their blood glucose levels allow it. Recovery periods between sets will range from 30 to 60 seconds, while recovery between exercises will last from 60 to 75 seconds. After completing five training sessions, the training load will increase. This can involve adding more repetitions, sets, kilograms of weight (if using an aquaball), or using a different color band (if using an elastic band). Users can access these options within the Diactive-1 App. Various progression examples are included in Supplementary Tables 1 - 10 . After participating in the intervention for four and a half months, participants will transition from their current fitness level to the next tier (e.g., from low to medium or medium to high fitness), receiving new intensities and exercises accordingly. The overall program will span 24-weeks, equivalent to a minimum of 72 sessions.

The waiting-list control group will receive standard hospital care, and after 24 weeks of the Diactive-1 App intervention, they will be given access to the App and training material.

An overview of the general procedure is provided in Figure 3 .

An electronic monitoring system will be used to oversee participants’ adherence to the training sessions. This will involve checking our database to verify the sessions completed by the participants. Additionally, we will employ the Polar Varity Sense device (Polar Electro, Kempele, Finland), which displays heart rate (HR) during training sessions, to facilitate verification. Throughout the intervention, the research team will maintain regular communication with participants through phone calls and messages to provide encouragement and support, with the aim of promoting participant retention. Moreover, participants will receive a monthly report detailing goal achievement summaries and their progress relative to other participants.

A summary of the variables analyzed in the Diactive-1 Study is presented in Table 1 .

The primary outcome measure is the daily insulin dosage requirement. Participants will maintain a diary for nine days, recording their carbohydrate intake and insulin doses. This diary will record information about insulin injections. Children and adolescents who use an insulin pump will be assessed by obtaining objective information through downloading their data. The collected information will be used to calculate the insulin units per day per kilogram of body weight. A comparison will be made by providing participants with the same diary again, both at 12 weeks into the intervention and 9 days before the intervention concludes at week 24.

Venous blood samples will be collected from the antecubital vein between 7:00 and 9:00 AM after a 10–12 hour overnight fast. These samples will encompass measurements of fasting glucose, glycated hemoglobin, triglycerides, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, apolipoproteins, alanine aminotransferase (ALT), and aspartate aminotransferase (AST). All assessments will be conducted both before and after the intervention at the central laboratory of the University Hospital of Navarra in Pamplona, Spain.

Standing height will be measured in bare feet using a SECA 213 stadiometer (Hamburg, Germany). Participants will be instructed to stand with their heels together and touching the base of the vertical measuring column, with their back straight and their head positioned in the Frankfurt horizontal plane (31). The standing height will be recorded to the nearest 0.1 cm.

Sitting height will be measured using the SECA 213 stadiometer and a wooden box.

Body weight will also be measured in bare feet and light clothing, using a SECA electronic scale (Scale 869), and recorded to the nearest 0.1 kg. Body mass index (BMI) will be calculated by dividing the weight in kilograms by the square of height in meters.

Sexual maturation will be assessed by the Pediatric Endocrinology Unit at the University Hospital of Navarra (Pamplona, Spain), reporting the pubertal status on a scale of 1 to 5 in relation to secondary sexual characteristics. The assessment will be conducted using the Tanner and Whitehouse criteria (32). For girls, assessment will be based on the stage of breast development (Tanner A) and the distribution of pubic hair (Tanner B), while for boys, assessment will be based on the stage of genital development (penis size and testicular volume - Tanner A) and pubic hair distribution (Tanner B).

To obtain the peak height velocity (PHV), a common indicator of growth and development in children and adolescents (33), we will use anthropometric measures (weight, height, and seated height) as per Moore’s equations (34). To calculate the years after PHV, we will subtract the age at PHV from the actual age. The difference in years between these values will be referred to as the maturity offset.

Total body fat, lean mass, subcutaneous and visceral adiposity, bone mineral content and density will be measured using dual-energy x-ray absorptiometry (DXA Lunar iDXA, GE Healthcare). Participants will be positioned in a supine position, with their arms slightly separated from the body and their feet and legs hip-width apart. This assessment will take place in weeks 12 and 24.

Heart rate variability (HRV), sympathetic and parasympathetic nervous system indices, as well as low and high frequencies, will be measured using the Polar V800, among other variability data. These measurements are related to autonomic function at the cardiac level and serve as indicators of autonomic dysfunction at this level. The HRV data will be analyzed using Kubios software (Kubios HRV Premium, ver. 3.5, Kubios Oy, Kuopio, Finland) (35).

Vascular function will be measured in the four extremities, the cardio-ankle vascular index, the brachial-ankle pulse wave velocity and the ankle-brachial index at rest using the Vasera VS-2000 Vascular Screening System (Fukuda Denshi, Japan).

Cardiorespiratory fitness will be assessed through a graded stress test using ergospirometry (Cosmed, K5 b2, Italy) on a cycle ergometer (Excalibur Sport 925909, Lode, The Netherlands). A standardized protocol for children and adolescents will be followed, including a warm-up phase, a systematic increase in resistance (10 or 20W per minute) until reaching maximum exertion, and then transitioning into the recovery phase. Peak oxygen consumption (VO_2peak) and metabolic equivalents (METs) will be determined.

Muscular fitness will be measured by handgrip strength using the Takei III Smedley Type Digital Dynamometer, which provides an estimate of an individual’s overall strength. Then, EGYM Smart Strength machines (developed by eGym® GmbH in Munich, Germany) will be used to measure both maximal strength and muscular power in upper (chest and arms) and lower (legs and hip) extremity muscles. Handgrip strength assessment will take place in weeks 12 and 24.

The volume and intensity of physical activity will be measured using a GENEActive triaxial accelerometer (ActivInsights) worn on the wrist of the nondominant hand. The accelerometers will be programmed to measure at a frequency of 87.5 Hz over a period of nine consecutive days (36). The research team will determine that sampling 86 times per second is sufficient to capture the majority of movements performed by patients. The accelerometer data will be extracted using GENEActiv PC Software (version 3.3) and processed and analyzed using the R package GGIR (37). Waking wear time for valid cases represented children and adolescents with at least seven days and at least 10 hours of waking wear time in a 24-hour period, including one weekend day, will be considered for analysis. Validated cut points will be used to determine different physical activity variables (38, 39): sedentary activity (for children: 0–56.3 mg; for adolescents: 0–50 mg), light physical activity (for children: 56.3–191.6 mg; for adolescents: 50–150 mg), moderate physical activity (for children: 191.6–695.8 mg; for adolescents: 150–500 mg), and vigorous physical activity (for children: >695.8 mg; for adolescents: >500 mg). Moderate-to-vigorous physical activity will be defined as activities for which at least 80% of 1 minute of time satisfies the moderate physical activity threshold criteria (i.e., 191.6 mg for children and 150 mg for adolescents), in order to remove signals related to random wrist movement (40). The duration of sleep will also be determined. According to van Hees et al. (41) a sleep algorithm will be used to detect sleep and wake between bedtime and get uptime.

The Pittsburgh Sleep Quality Index (PSQI) will be used. It evaluates seven established aspects of sleep quality: subjective sleep quality, time taken to fall asleep, duration of sleep, sleep efficiency, sleep disturbances (such as nightmares, pain, or feeling too hot or cold), use of sleep medication, and daytime dysfunction (42).

Sedentary behaviors will be assessed using the Youth Leisure-Time Sedentary Behavior Questionnaire (YLSBQ). Participants will report the time spent on TV, video games, computers, and mobile phones on both weekdays and weekends. To calculate the weighted average daily sedentary screen time for each behavior, we will use a 5:2 ratio. For instance, this calculation involves multiplying the daily TV viewing time on weekdays by five, the daily TV viewing time on weekend days by two, and then dividing the sum by seven (43). The total daily sedentary screen time will be determined by summing the durations of various daily screen time activities. Additionally, total screen time for both weekdays and weekends will be calculated.

The measurement of physical activity will be based on the following question: “Typically, how many days will you engage in physical activity for a total of at least 60 minutes?”. Response options will range from 0 to 7 days per week, in 1-day increments. Physical activity will be defined as less than 60 minutes of physical activity per day on at least 7 days per week (44). The measurement of muscle-strengthening activities will be based on the following questions: “In the past 7 days, how many days did you perform exercises to enhance or tone your muscles, such as pushups, sit-ups, or weightlifting?” The response choices ranged from 0 to 7 days.

Participants who engaged in at least 60 minutes of moderate to vigorous physical activity per day and at least three days of muscle-strengthening activities, had less than two hours of recreational screen time per day, and achieved uninterrupted sleep for 9 to 11 hours per day (for children) or 8 to 10 hours per day (for adolescents) will be categorized as meeting the comprehensive 24-hour movement guidelines (45).

The International Fitness Scale (IFIS) is designed for assessing self-reported physical fitness. This scale consists of five elements that employ a 5-point Likert scale to inquire about the children’s general perception of their physical fitness, as well as their perception of their cardiorespiratory fitness, muscular fitness, speed-agility, and flexibility relative to their peers. The Likert scale provides choices ranging from very poor to poor, average, good, and very good physical fitness (46, 47).

The perception of hypoglycemia will be measured using the Clarke test, which consists of eight questions with different possible answers, A score greater than three reflects impaired awareness of hypoglycemia (48).

Self-reported variables will be collected via a questionnaire administered to the participants’ parent(s) or guardian(s). The questionnaire will cover participant details such as school, sex, age, birthplace, race/ethnicity, and language spoken at home. Additionally, it will include information about the parent(s) or guardian(s), including birthplace, age, education level, professional qualifications, employment status, job role, monthly income, household location, neighborhood, and birth weight.

To evaluate the HRQL in context of a chronic illness, the Spanish version of the “Questionnaire for Young people with Diabetes” (DISABKIDS) will be used (49). This questionnaire comprises 12 questions about how a patient has felt in the last four weeks that require answers on a 5-point Likert scale from 1 (never) to 5 (always).

We will also assess the HRQL using the Screening for and Promotion of HRQL in Children and Adolescents (KIDSCREEN-10) (50). This is a generic 10-item unidimensional instrument that focuses on the functional, mental, and social aspects of well-being in children and adolescents aged 8-18 years. The instrument will consist of the following items, starting with “thinking of last week, have you: 1) felt physically fit and well, 2) felt full of energy, 3) felt sad, 4) felt lonely, 5) had enough time for yourself, 6) been able to do the things you want in your free time, 7) felt treated fairly by your parent(s), 8) had fun with your friends, 9) got along well at school, 10) been able to pay attention at school?”. For each item, participants will provide their responses on a five-point scale, ranging from “never” to “always” or from “not at all” to “extremely”.

We will also assess subjective well-being using the “Cuestionario Unico de Bienestar Escolar” (CUBE) (51). The CUBE questionnaire consists of 5 items that assess various aspects of life satisfaction. All these variables will be measured using a 10-point Likert scale ranging from 0 to 10 (0 = totally disagree, 10 = totally agree).

In terms of positive affect, the scale includes five items assessing emotions such as happiness, joy, cheerfulness, contentment, and fun. Additionally, there are five items evaluating negative affect, which encompasses feelings of humiliation, annoyance, irritation, bitterness, and sadness. The scale follows a bifactorial structure with five items per factor.

We will use the Mediterranean Diet Quality Index for Children and Teenagers (KIDMED) index to assess adherence to the Mediterranean diet (52). The KIDMED index ranges from 0 to 12 and is based on a 16-question test. Unhealthy characteristics associated with the Mediterranean diet are assigned a score of -1 point, while healthy characteristics receive a score of +1 point.

The sum of all scores obtained from the KIDMED test will be utilized to classify individuals into three different levels: (a) optimal Mediterranean diet (>8 points), (b) improvement needed to align with Mediterranean dietary patterns (scores ranging from 4 to 7), and (c) very low diet quality (≤3 points).

Food frequency consumption will be assessed some food frequency questionnaires (FFQs) for the Spanish young population (53, 54). These FFQs includes several items groped into 17 food groups and were previously validated for its use among children (54) and adolescents (53) in a self-reported way. Subsequently, macronutrients, micronutrients, total energy consumption and other diet-related variables will be estimated. On the other hand, adherence to the healthy and sustainable dietary recommendations (e.g., fruits, vegetables, nuts, etc.) of the Ministry of Consumer Affairs of the Government of Spain will be determined (55).

To evaluate disordered eating, the modified SCOFF (mSCOFF) questionnaire will be used. This screening tool consists of five straightforward questions and can be conveniently incorporated into a routine check-up. It has proven to be reliable and valid in its assessment of eating disorders among children and adolescents with T1DM (56).

We will use the Spanish Version of the User Version of the Mobile Application Rating Scale (uMARS) (57), that will serve as a comprehensive and objective measure of app usability, consisting of 20 items. Each item is assessed using a 5-point scale, ranging from 1 (inadequate) to 5 (excellent). The uMARS will be structured into four subscales: engagement, functionality, aesthetics, and information quality. Subscale scores will be calculated as the average of their respective items, and the mean of these subscale scores will provide an overall app quality score.

Eligible and consenting participants will complete a baseline assessment. The intervention group will then be allocated to a 24-weeks exercise training program using the Diactive-1 App. The control group will continue standard care. Outcomes will be assessed at baseline (T0), 12-week (T1) and after the 24-week intervention period (T2) ( Table 2 ).

At least 52 children and adolescents of both sexes, aged between 8 and 18 years old, will be recruited from the Pediatric Endocrinology Unit at University Hospital of Navarra (Pamplona, Spain). This sample size was determined based on the results of a previous meta-analysis (12), which suggested an expected effect size on daily insulin dose requirements of 0.81. Using the G*Power software (58) and considering a power of 0.80, a significance level of 0.05 and accounting for a dropout of 15% (59, 60), a minimum of 26 children and adolescents per group is required.

Participants meeting eligibility criteria will be randomly assigned to one of two groups in a 1:1 ratio using block randomization with a computer-generated schedule (Research Randomizer V.4). This will ensure equal group sizes. Randomization will continue until a predetermined number of participants are assigned. The allocation code will be kept confidential at Navarrabiomed until final analysis. To ensure blinding, each study group will be assigned an alpha-numeric code. Researchers will receive the code for their assigned group. Data analysts will not have access to the code until completing the coded intervention analysis. Due to the nature of the study, patients in the Diactive-1 App group will not be blinded.

This study is an unblinded, practice-level intervention.

Researchers will be responsible for finalizing the study protocol and maintaining regular communication through phone calls, emails, and meetings. Weekly meetings will also be held for investigators and pediatric staff to discuss progress and updates.

The provided information will be documented in the database using individualized study codes assigned to each participant. This data will be securely stored on a computer that requires a password for access. Only the data manager, who operates independently and without conflicting interests, will have permission to access and retrieve the data. Due to the low level of risk associated with the study, there is no requirement for a Data Monitoring Committee. However, any significant modifications to the study protocol will be promptly communicated and updated on both the Clinical Trial Registry and the publication journal. Within a maximum timeframe of three years from the collection of the end-line assessment at the 24-week mark, a fully anonymized dataset will be submitted to an appropriate data archive for sharing purposes.

The quantitative variables will have their mean (M) and standard deviation (SD) provided, whereas the qualitative variables will include frequencies (n) and percentages (%). To assess data normality visual inspection of Q-Q plots and Shapiro-Wilk test will be used. For the homogeneity of variances, Levene’s test will be used. Subsequently, for two-group comparisons, either Student’s t-test or Mann–Whitney U test will be used based on adherence to the normality assumption.

Associations among qualitative variables will be examined using Pearson’s chi-square (χ²) test. For quantitative variables, the association will be tested using Spearman’s rank correlation coefficient (ρ) or Pearson’s correlation coefficient (r), depending on the assumption of normality. Initial analyses will establish frequency, range, variability, and distribution patterns of each variable, guiding the choice of the most appropriate statistical test for comparisons.

Given the experimental design of this RCT involving two data collections—baseline (t0 = 0 weeks) and post-intervention (t1, t2 = 12 and 24 weeks)—in both intervention and control groups, a comparative analysis will be conducted to identify intergroup differences. Multilevel mixed-effects regression models with repeated measures will be used to evaluate the intervention effect for each dependent variable. Multivariate analyses will account for autocorrelation between repeated measures.

For data analysis, both intention-to-treat (ITT) and per-protocol (PP) approaches will be employed. ITT measures the impact of intervention assignment, while PP analysis gauges the effect of intervention receipt. Statistical analyses will be performed using Stata software (version 17.0) (StataCorp, College Station, TX, USA), the statistical software R (Version 4.1.1) (R Core Team, Vienna, Austria), and RStudio (Version 2021.09.2) (Posit, Boston, MA, USA). Statistical significance will be determined by a p-value ≤ 0.05.

In situations where exercise sessions are missing, the analysis of outcomes will incorporate a dose-response approach. This approach will consider potential variations in outcome measurements based on the degree of exposure to the intervention. Additionally, depending on the nature of the missing data for both primary and secondary outcome measures, a range of techniques like multiple imputation, listwise deletion, or specific analytical methods will be employed. These measures aim to mitigate the impact of missing data, thereby minimizing any potential reduction in the generalizability of the collected data (61).

Subgroup analyses will be carried out to ascertain whether exercise is more or less effective in reducing secondary outcomes compared to the waiting-list control group. These subgroup analyses will adhere to the same methodology as the primary analysis, including both the primary analysis variables and their interaction with the experimental condition.

The team will oversee and document any unfavorable incidents, ensuring that severe adverse events are promptly reported to the designated committee, following their recommendations. There won’t be a requirement for a formal data monitoring committee in this particular RCT. The need for a data monitoring committee was not deemed necessary given the low-risk nature of this intervention.

Serious adverse events associated with the intervention will be reported to the Pediatric Endocrinology Unit at the ethics committees from the University Hospital of Navarra (Pamplona, Navarra).

As detailed in this protocol, the team will conduct weekly monitoring of all RCT aspects. This encompasses ensuring adherence to the protocol, maintaining ethical and governance standards, overseeing database management, assessing outcomes, conducting research staff training, and regularly reporting on informed consent.