Comparison of Home- vs. Gym-based Exercise Delivery Modes

A total of 40 participants from the project SPORTS4ALL by Ginásio Clube Português (GCP) were recruited to participate in the study. Study advertising took take place through direct invitation by the research team and rehabilitation technicians. The time frame for recruitment of participants take approximately 1 month. Eligible volunteers were invited to an initial meeting with their families/legal guardians, exercise physiologists, and rehabilitation technicians. During this meeting, the procedures of the trial, testing procedures, benefits, risks, and the time required for the study will be explained. In addition, an information sheet about the research and an informed consent form for participants and parent(s)/ legal guardian(s), in accordance with the approval by the Ethical Committee of Faculdade de Motricidade Humana - Universidade de Lisboa, was distributed. Informed written consent were signed by participants and parent(s) legal guardian(s) before allocation. After signing the informed consent, participants or parents/legal guardians were asked to answer the Physical Activity Readiness Questionnaire for Everyone (PAR-Q+) questionnaire, to assess their readiness to increase their levels of physical activity, and a sociodemographic data questionnaire. During a second visit, all volunteers participated in familiarization sessions before baseline testing at the laboratory. The number of familiarization sessions was repeated until the researchers observed that the participants are confident and able to perform all assessments correctly (1-2 sessions). Of the 36 participants recruited, by the time the mandatory COVID-19 home confinement was declared by the Portuguese government, only 17 participants had completed laboratorial exercise testing. Only these participants were part of the present study.

This study was a randomized controlled trial, where a convenience group of 17 who had completed laboratory exercise tests, were randomized in a 1:1 ratio to an SIT or CAET regime by using a random-block randomization scheme. Participants in both aerobic regimes received a 16-week supervised exercise program performed 3 times a week for 60 min each session. The intervention started with 8 weeks of online training via Google Meets at their homes followed by 1 month of detraining with no exercise sessions besides those made available by the SPORTS4ALL program, and another 8 weeks of on-site training at the gym. During the intervention, adherence to the protocols were assessed with an exercise diary. Assessments were conducted on 4 occasions over a 20-week period: before (M1; 0 weeks) and after (M2; 8 weeks) the home-based intervention; and before (M3; 12 weeks) and after (M4; 20 weeks) the gym-based intervention. During M1, assessments and personal information was obtained. All laboratory tests were conducted by the same researchers during the morning. Participants were asked to refrain from eating for at least 3 h, alcohol for 24 h, caffeine for 8 h, and vigorous exercise for 48 h before data collection. All laboratorial evaluations were performed on a single day of testing at each measurement round. On this day, the participants had their (1) body composition measured and then quietly rest for at least 15 min in the supine position before data collection in the following order: (2) brachial arterial pressure, (3) carotid intima media thickness, (4) stiffness indices by carotid vascular ultrasound, (5) regional arterial stiffness by applanation tonometry and (6) an incremental test to exhaustion.

The exercise regimes were delivered at participants' homes for 8 weeks via Google Meets and at the gym during an identical period. All sessions were designed by experienced exercise physiologists together with rehabilitation technicians, following the guidelines of the American College of Sports Medicine (ACSM) and the National Strength and Conditioning Association (NSCA). Exercise physiologists in charge of delivering the interventions underwent standardized training to ensure that the sessions offered were comparable and went through a 2-month familiarization period where they observed the exercise sessions from the project SPORTS4ALL by GCP, met the participants, and led exercise activities, while under the supervision of the responsible Rehabilitation Technicians of the project SPORTS4ALL. During this period, the rating of perceived exertion was reinforced using the OMNI scale. The multi-component exercise programs included aerobic, resistance, balance, and flexibility training. Both SIT and CAET regimes were divided into 2 groups: 8-9 participants/group and the participant-Exercise Physiologist ratio of 3:1, partnered with at least 1 Rehabilitation Technician in each exercise session. Both programs were divided into 3 phases: initial phase (week 1-3); improvement phase (week 4-6); and maintenance phase (week 7-8). Sessions for both regimes were also matched for total duration. All sessions included 5 min of standardized warm-up and cool-down exercises. The warm-up was implemented to gradually increase the HR, body temperature, joint mobility and prepare the body of the participants to the exercises of the conditioning phase. After the conditioning phase, cool-down exercises were performed followed by stretching exercises. During the washout period (4 weeks), 14 participants performed 3 sessions representing each phase of the home-based intervention wearing heart rate monitors to obtain a proxy of the exercise intensity in each phase of the home-based intervention. Resistance training included a total of six exercises: 3 for the upper limbs (seated biceps curl, triceps extension, and frontal shoulder raise); 2 for the torso (1 for the back: seated row and 1 for the abdominals: side bend); and 1 for the lower limbs (chair squats). All exercises were performed with 2-4 kg dumbbells, or the participants' body weight. The intensity of the strength training was monitored using the 0-10 points Rate of Perceived Exertion OMNI- Resistance Exercise Scale (RPE OMNI-RES) validated for resistance training. The intensity of resistance training was kept constant at 2 x 12 rep at RPE OMNI-RES 8, throughout the study period. Balance training was designed to challenge the central and peripheral mechanisms involved in the control of movement and stability (vestibular; visual and somatosensory). Flexibility exercises were performed after the cool-down period of each session. Sustained stretches were used for each major muscle group targeted during the session. Active and passive static stretching exercises were used, holding the position for 10-30 s at a point of tightness or slight discomfort.

Statistical Analyses An intention-to-treat analysis was performed using all randomized participants. The data are presented as mean and standard deviation. The Shapiro-Wilk and Levene tests, as well as plot inspection, were used to verify normality and homoscedasticity assumptions, respectively. To compare baseline individual characteristics independent-samples t-tests were used.

The changes in the cardiorespiratory fitness and vascular outcomes were examined using linear mixed models fitted with restricted maximum likelihood and applying Satterthwaite's method for approximating degrees of freedom for the F test from the R lmerTest package. The fixed effects were time and aerobic regime, whereas the random intercept was defined for each participant. Using the R sjstats package, partial eta squares (η2) were calculated for each main effect and interaction and interpreted as small (η2 < 0.05), medium (η2 < 0.25), and large (η2 > 0.25) effect sizes. In the presence of significant differences in main effects and interactions, post-hoc comparisons using Tukey's HSD test were performed using the R emmeans package. All statistical analyzes were performed with a significant level (α) of < 0.05 using the R software, version 4.1.2.