This was a retrospective study analyzing data from a three-month period (23/02/2020 to 23/05/2020). The study was designed according to the guidelines of the Helsinki Declaration for conducting human experimentation and was approved by the Farhat HACHED ethical committee, Sousse, Tunisia (FH/16102020). Each participant was informed of the purpose, procedure and study details. Prior to the lockdown, participants were all continuously practicing regular PA, with an average of 12 h/week for more than the past 10 years.

The sample size was estimated using the following formula (Buysse et al., 1989[10]): N = [(Z_α)² × P × (1 - P) × D]/E²; where «E» was the margin of error, «Z_α» was the normal deviate for one-tailed alternative hypothesis at a level of significance, «D» was the design (= 1 for simple random sampling), and «P» was the proportion of the main event of interest (i.e., percentage of decrease in PA pre- and during- the home confinement). Given the pioneer character of our study, the proportion of the main event of interest was derived from a previous Spanish study (Sañudo et al., 2020[41]) aiming to evaluate the impact of COVID-19 home confinement on PA, sedentary behavior, smartphone use, and sleep patterns (n=20 young adults). In that study, the number of steps/day (objectively measured PA) significantly decreased by 68 % (p=0.68) during home confinement [from 8525±3597 (pre-lockdown) to 2754±1724 (during-lockdown)] (Sañudo et al., 2020[41]). Assuming a confidence interval of 95 % (Z_α = 1.64) and an «E» of 0.20, the total required sample size was 15 participants.

At the beginning of the study, age (year), height (cm) and weight (kg) were determined, and body mass index (kg/m²) was calculated. PA, physiological, and sleep data were collected over the following four consecutive months (Table 1(Tab. 1)):

Month 1: pre-confinement (up-to 23/02/2020-22/03/2020),

Ramadan month occurred during the 2^nd month of confinement (from 23/04/2020 to 22/05/2020).

Data were collected using a smartwatch (Garmin Fenix 6 pro, USA). This wearable technology device is useful, reliable and valid in the assessment of sleep and PA levels (Saganowski et al., 2020[39]). Participants continuously tracked their physical and physiological activities for four consecutive months. The watch was only removed for charging purposes (~90 to 120 minutes/week). No missing data was reported other than during the charging time. Smartwatch data were downloaded using the software package (Fenix 6 Pro/6 Pro Solar software version 13.10) and were exported to a customized Microsoft Excel spreadsheet. Several investigations have also demonstrated high validity and reliability evidences of smart watch including the Garmin Fenix 6 pro or earlier models in accessing PA, step count HR, RR, calories (Navalta et al., 2020[33]) and sleep parameters (Chinoy et al., 2021[11]). A detailed description of PA (i.e.; PA, step count), physiological (i.e.; HR, RR, calories) and sleep (i.e.; sleep time, sleep latency, REMS time) parameters selected to be analyzed in the current study, is presented in Table 2(Tab. 2). HR was expressed in bpm (Souissi et al., 2021[44]), and as a percentage of the predicted maximal HR (PMHR (bpm) =220 - age (year)) (Robergs and Landwehr, 2002[36]).

The normality of the data was verified by the Kolmogorov-Smirnov test. All data were expressed as means ± standard deviation (SD). One-way analysis of variance (ANOVA) with repeated measures was used to explore differences between the four consecutive months in the averaged data of PA, physiological and sleep pattern. Effect sizes were calculated as partial eta-squared (ղp²) to estimate the meaningfulness of significant findings. Tukey's post-hoc test was used to verify pairwise comparisons. Statistical significance was set at p<0.05. All statistical analyses were conducted using the statistical package for the social sciences (SPSS, Version 18.0, SPSS Inc., Chicago, IL, USA).

Our results demonstrated a marked decrease in step count during the 1^st and the 2^nd month of confinement. Additionally, a decrease in PA time was observed from pre- to during-confinement. These results were expected, as all participants were banned or forbidden to leave their homes, thus preventing them from engaging in their regular PA. Moreover, the decrease in step count was more significant during the 2^nd month of confinement, likely due to participants training less during the lockdown. Changes in routine, increased stress, and increased anxiety are factors that decrease engagement in PA (Andreato et al., 2020[7]). Furthermore, our results reported calories spent/day decreased during the 1^st month of confinement and continued to decrease during the 2^nd month of confinement. It should be acknowledged that RIF month (i.e., 4^th pillar of Islam, during which Muslim adults are not allowed to eat and drink from dawn to sunset during a consecutive 29 or 30 days) occurred during the 2^nd month of confinement. Lessan et al. (2018[27]) reported that RIF results in profound disruptions of daily activity patterns, marked by a decrease in the total number of steps per day. It appears that RIF could have exacerbated the decrease of the total number of steps during the 2^nd month of confinement.

The marked decrement in PA, step counts, and calories/day in response to the increase of detraining period during home confinement may also reduce cardiovascular fitness. The reduced PA reached levels well below the daily recommendation of 7500-10,000 steps per day and may further exacerbate inactivity-related health problems (Booth et al., 2017[8]). In the same context, our findings show resting HR increased during the 1^st and the 2^nd months of confinement. Narici et al. (2021[32]), reported that the spent sitting time has been linked with a reduced cardiorespiratory and cardiovascular performance. Moreover, the reduction of PA to <5000 steps/day for only a few days can impair nitric oxide mediated vasodilation (Porcelli et al., 2020[34]). Additionally, the present results showed a decrease in the average and highest RR during home confinement. The decrease in average and highest RR could be explained by the reduction in PA time and step counts and/or exercise intensity accompanied by increased behavioral stress.

Low-intensity exercise training was recommended to prevent COVID-19 (Ghram et al., 2021[21]; Jiménez-Pavón et al., 2020[25]). Acute responses to higher intensities and volumes of exercise can involve a greater risk of illness and impaired immune function (Moreira et al., 2009[28]). In this context, Toresdahl and Asif (2020[46]) advised athletes to follow a conservative approach, limiting training sessions to <60 minutes and to <80 % of maximum ability during this time to prevent contracting COVID-19.

Our results showed a modification of sleep parameters during home confinement. Total sleep time increased by 8 % during the 1^st month of confinement. In addition, total sleep time increased further during the 2^nd month of confinement (i.e., Ramadan month). The previous finding was unexpected as the results of previous meta-analyses (Faris et al., 2020[20]; Trabelsi et al., 2020[48]) concluded a decrease in total sleep time in athletes during RIF. The decrease in training load/PA levels could have blinded the effect of RIF on total sleep time. Additionally, sleep latency, and REMS increased during home confinement. The increase in the total sleep time, sleep latency, and REMS during home confinement may be due to people staying home and not waking up to go to work. However, deep sleep time decreased during home confinement, despite the increase of total sleep time. The decrease of deep sleep time is associated with an increase in sleep latency, which was not previously observed during RIF in physically active individuals (Hsouna et al., 2019[24]), indicating an impairment of sleep quality during home confinement (Akerstedt et al., 1994[2]). Again, changes in routine and increased stress and anxiety may decrease the quality of sleep and impede engagement in PA.

Interestingly, the results showed all sleep parameters were restored during post-confinement. Furthermore, we observed an increase in PA time associated with a marked increase in step count during the 1^st month of post-confinement. PA time and step count increased during post-confinement reaching the same values as pre-confinement. Calories/day was also restored during home confinement; however, unlike parameters of sleep and PA, physiological parameters (Resting HR, HRpeak and highest RR) were not fully restored. Our finding highlighted higher values of HRpeak and highest RR during post-confinement vs. pre-confinement. This may be related to the adaptation difficulties of HR response with the same training load performed during post-confinement compared to pre-confinement. We hypothesized that fitness coaches had a decline in physical fitness during post-confinement due to significantly reduced training. Although they succeeded to enhance their quality of sleep and their daily PA, those parameters were still lower compared to the pre-confinement; participants also become more fatigable during post-confinement vs. pre-confinement. Our findings confirm a modified lifestyle due to home confinement can lead to serious deficits in the quality and quantity of training (Andreato et al., 2020[7]). Our work is in agreement with Narici et al. (2021[32]), who reported that living in home-confinement for several weeks represents a physiological challenge with subsequent significant health risks. It has been showed that in response to 20-day bed confinement, young healthy participants lose on average 11 % of heart volume and 28 % of maximum oxygen uptake (Saltin et al., 1968[40]). Furthermore, a period of 2-4 weeks of detraining causes a decrease in blood volume together with a decrease in haemoglobin content. These physiological changes cause a decrease in muscle capillarization and a loss of efficiency of the body temperature regulation mechanisms (Costill et al., 1985[14]; Souissi et al., 2022[43]). Furthermore, after a detraining period longer than one month, a decrease in skeletal muscle oxidative enzyme activity is observed (Costill et al., 1985[14]). This important mismatch of the aerobic system, both from the central and the peripheral perspectives, requires attention during the retraining phase.

Based on these profound physiological changes of the aerobic system, the training period during home confinement should be based on a careful choice of intensity volume, and type of training. Herrera-Valenzuela et al. (2020[23]) recommended high-intensity interval training at home to maintain cardiorespiratory endurance and physical fitness. Similarly, Jukic et al. (2020[26]) warned that detraining is one of the biggest negative consequences of the current “stay at home” confinement. Therefore, the authors recommended that the athlete's house should be equipped with cardiovascular and resistance training equipment (e.g., portable cycle or rowing ergometer). It is also recommended that athletes incorporate some type of endurance exercise into their daily routine to avoid the effects of detraining during the forced quarantine. However, to avoid increased chances for upper respiratory tract infections, the athlete should train at relatively high intensities, but should not fully exhaust themselves during a training session (Moreira et al., 2009[28]; Puta et al., 2017[35]). This study also provided insight regarding the use of smart wearable devices; digital technology has become popular in the general population and has already been adopted for health monitoring. To date, smart devices have been generally used to investigate PA in daily life, showing to be effective and applicable for a large part of the healthy population and patients. Here we recommend the use of smartwatch technologies by patients, athletes, and scientific researchers based on the amount of data that they can provide and cost effectiveness. The authors, therefore, hope that, like seasonal influenza, COVID-19 will become endemic in the human population, and that COVID-19 vaccines will be incorporated as an add-on to seasonal influenza vaccinations, provided every winter for at least the next few decades (Dergaa et al., 2021[15]; Musa et al., 2021[29]). With proper immunization, the globe will return to pre-COVID levels of normalcy, and with the hope that COVID-19 lockdown scenario will not recur.

The present study has some limitations that should be acknowledged. Our findings pertain only to physically active males. Future studies on female individuals are warranted. In addition, a control non-Muslim group (not fasting during RIF) is lacking; however, the recruitment of such group is difficult in Muslim countries.

The authors are particularly grateful to all the participants for their enthusiasm and their performance.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The data that support the findings of this study are openly available upon request from the corresponding author.

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.