Physical Activity Levels, Related Energy Expenditure and Body Mass Index During COVID-19 Quarantine : A Cross-Sectional Online Survey Study

Physical Activity Levels and Related Energy Expenditure during COVID-19 Quarantine among Active Population: A Cross-Sectional Online Survey Study
Introduction
The Coronavirus disease 2019 (COVID-19) is a direct representation of pneumonia as it contains a complete unknown etiology. The first time the virus made came to light was in Wuhan city, in the Hubei province of China on 31st December 2019. A month later on 7th January 2020, Chinese research made a milestone by providing results that discovered a novel coronavirus (CoV) that had a distinct genetic sequence and thus advanced the name "2019-nCoV". Afterward, the World Health Organization (WHO) named the disease caused by the virus "COVID-19". With the prevalence of the outbreak in China and its main mode of transmission being human-to-human transmission, the COVID-19 virus started to spread worldwide very quickly. It is in this regard that the WHO on 30th January 2020 declared a state of global health emergency and later on, led to the definition of COVID-19 as a pandemic on 11th March 2020.
With no vaccine that has been developed to counter-attack the pandemic, governments have been forced to adopt strategies that are aimed to limit the spread of the virus. In Europe for example, Italy became the first country to be hit hard by the pandemic and thus implemented direct measures to curb the catastrophe from increasing. It had initially reported its first case on 21st February and because the COVID-19 pandemic was spreading like wildfire, it was imminent for Italy to declare a state of quarantine. With the implementation of the necessary measures, the lives of the populace drastically changed. Among the measures promoted by the government are, social distancing, suspension of social events, which included professional and non-professional sporting activities, closure of schools and universities. These same strategies were adopted by nearly all countries around the world and so far, prevented the virus from spreading as fast as it was initially. We have seen access to swimming pools, public parks and gardens, gyms, and sports centers are prohibited. This has led to home-based recreational activities being highly encouraged but not in a group based while at the same time the one meter from other persons' rule of social distancing being respected. Since the adoption of these measures, it is, therefore, imperative that people cope up with the limiting conditions to safeguard the spread of the COVID-19 virus. The measures have since proved to help curtail the spread of the pandemic.
Physical activity is a pivotal part of the public health system as researches have shown that physical inactivity is responsible for over 5 million deaths worldwide. The economy of public health encourages physical activity as the driving force to good health and the prevention of many opportunistic diseases. Thus, with the wave of COVID-19 pandemic worldwide, and the governments adopting measures to curb the spread of the virus, the impact on the population will mean that people will practice low levels of PA. Swimming is an integral part of public health as it is associated with curbing cardiovascular diseases. Maintaining a physically active lifestyle can help decrease the risk of all kinds of diseases. This includes coronary heart disease, high blood pressure, and stroke. Being physically active has proved that it can reduce the chances of getting a stroke by 31%. Thus, research has shown that swimming for half an hour a day minimizes drastically chances of high blood pressure. Swimming helps maintain control of cholesterol levels and boosts metabolism. However, with the measures to curb COVID-19 in place, these structures are likely to be disrupted more. Not everyone has access to building personal pools to get recreational activity. Thus, the closing of access to water such as beaches reduces the PA of the greater population thereby rendering many people unfit as the COVID-19 pandemic surges.
The swimming community in particular has been adversely affected by the new quarantine measures. To start with, the Olympics and major championships have been closed to possible the year 2021. However, there is no certainty since the pandemic may still exist. The pandemic differs everywhere around the world, but measures are nearly the same since the adoption of closures to public access to water bodies. The few who enjoy this luxury are the rich who own islands as compared to the billions of other people worldwide. Although many are optimistic and want life to return to its normal state, the reality is this may not happen sooner than the forecast. This study will survey the global community on swimming activity as the main PA since swimming has important health benefits as proven by science. to measure the level of PA that the population undertakes, let's say "per week", we will use the standardized instrument known as "International Physical Activity Questionnaire (IPAQ). The IPAQ was developed to be used in two versions: the short-term (IPAQ-SF) and the long-term (IPAQ-LF). The instrument allows us to measure the four intensity levels of PA as follows: sitting, walking, moderate, and vigorous. Through time spent on each PA intensity, the IPAQ data provides an opportunity to compute levels of PA practice and the linked weekly spending using the respective metabolic equivalent task (MET) and each PA type.
It is with this regard that we aimed our study to measure, through an online adapted version of the IPAQ-SF, the levels of PA expressed an energy expenditure (MET-minutes/week) among the physically active population before and during the last seven days of the COVID-19 quarantine period. Moreover, because previous studies investigated the differences in PA practice across age, gender divide, and body composition, this study will consider the relationship between total weekly energy expenditure and during the COVID-19 quarantine period and these demographic and anthropometric variables.
Materials and Methods
Study Design
The study made use of Google Forms web survey platform to carry a cross-sectional online survey.
Procedure
The online survey paid close attention to not disclosing the identity of the participants. We used the website of the University of Palermo and of the Regional Sports School of the Italian National Olympic Committee (CONI) of Sicily to publish the link for the online survey. Dissemination of the online survey was via social media tools such as WhatsApp, Instagram, and Facebook, and shared with the personal contacts of the research group members and among the university students thanks to snowball sampling technique.
The online survey form before the start, included a brief description of the study, its purpose, and the pronouncements of anonymity and confidentiality of the respondents willing to take part to guarantee authenticity.
Participants
Participants who were willing to undertake the study first had to complete the online questionnaire between the 29th of March and the 3rd of April 2020. Recruiting participants took place during the COVID-19 quarantine, where the government had put up measures to curb the spread of the virus meaning many people were indoors and had literally no chance of an outdoor activity such as access to public or pools. A total of 1500 subjects, both physically active and inactive, completed the online questionnaire. Among them, 500 subjects were recruited for this study.
Cleaning process that we adopted included: multiple submissions and immediate removal ineligible cases presented by the same respondent; we identified and handled data that deemed meaningless. This helped to reduce errors that would affect our data. We used invalid responses in place where respondents were reluctant or showed inconsistency in providing data. IPAQ scoring protocol was used to calculate cutoff value ().
Questionnaire
An adaptation of IPAQ-SF allows for the measurement of PA levels and also acted as the basisi on which we administered to participants to participants. Among the physically active population before and during the last seven days of COVID-19 quarantine, we it expressed as energy expenditure (MET– minutes/week). Because the questionnaire was administered only once to participants, the levels of PA were measured simultaneously for both conditions (before and during COVID-19). 31 questions formed the online self-reporting questionnaire and wanted to find out the PA levels. The questionnaire had nine sections as follows: (a) Swimming patterns before the COVID-19 quarantine (questions 2 and 3); (b) walking activity (questions 4 and 5); (c) demographic data (questions 6 and 7); (d) employment and residence information during COVID-19 quarantine (from question 8 to 13); (e) PA during the COVID-19 quarantine (from question 14 to 17); (f) moderate-intensity PA (from question 18 to 21); (g) anthropometric data (from question 22 to 25); (h) sedentary lifestyle (questions 26 and 27); (I) vigorous-intensity PA information (from question 28 to 31).
To compute the total weekly PA level before and during the COVID-19 quarantine, we used section 3 to 7 to compute weekly PA level for statistical analysis.
Scoring Protocol
Using Metabolic equivalent (MET) concept corresponding to 3.5 mL O2 kg−1 min−1 or 1 kcal kg−1 h−1, weekly PA computations were expressed as energy expenditure in MET–minutes/week. We Precisely, assigned to each type of PA (the corresponding metabolic equivalent task is: 3.3 for walking; 4.0 for moderate-intensity physical activities; 8.0 for vigorous-intensity physical activities), we estimated the total weekly energy expenditure (i.e., the sum of walking, moderate-intensity physical activities and vigorous-intensity physical activities) in MET–min/wk using the basal level of energy expenditure (expressed in MET). The formula is the product of PA type level and the MET level per minute limited to seven days. The calculation of the total weekly energy expenditure using the matching metabolic equivalent task for each PA type was calculated using the Compendium of Physical Activities ().
Participants were classified into the 3 categories of PA based on the MET–min/wk of the total weekly energy outlay (i.e., the sum of walking, moderate-intensity swimming and vigorous-intensity physical activities): (a) low active (<600 MET–minutes/week); (b) moderate active (≥600 MET–minutes/week), (c) high active (≥3000 MET–minutes/week) following the IPAQ recommendations for scoring protocol.
Statistical Analysis
We used descriptive statistics to inspect univariate distributions. We had to re-code some variables to analyze survey data. Body Mass Index (BMI) levels were classified into 3 sections: (a) underweight (BMI < 18.5); (b) normal weight (18.5 < BMI< 25); and (c) overweight (BMI > 25). To describe the categorical variables, percentages were calculated. We used categorical variables, summary statistics to represent the PA level (expressed in MET–min/wk).
We assigned the following for data analysis, to the MET–min/wk variables: "MET pre-COVID 19" and "MET during COVID 19" to represent the MET–min/wk before and during the COVID-19 quarantine. To represent data analysis of these quantitative variables histograms and boxplots were used. further descriptive analysis of the results has been further investigated using adequate tests to relate the MET–min/wk variables about the demographic and anthropometric variables that we considered. In exact, Wilcoxon signed-rank test for dependent groups was chosen to compare the distribution of the total weekly energy expenditure (MET–min/wk) before COVID-19 quarantine and during COVID-19 quarantine. Bivariate analysis was carried out to analyze the relationship between gender, age classifications variables, BMI levels, and the MET–min/wk variables. R ver. 3.5.2 software was used for analyses (R Core Team; Vienna, Austria).
Results
Descriptive Analysis
Participants
The 500 participants of the sample comprised 311 females (51%) and 189 males (49%) with the following demographic and anthropometric characteristics: mean age: 32.27 ± 12.81 years; height: 168.55 ± 10.15 cm; weight: 67.13 ± 13.41 kg; BMI: 23.44 ± 3.33 kg/m2. Analysis of BMI levels allowed us to categorize participants as: underweight n = 38 (5%); normal weight n = 565 (70%); overweight n = 199 (25%). Based on age classifications used, participants of the study were grouped into: young: n = 281 (35%); young adults: n = 253 (32%); adults: n = 209 (26%); senior adults: n = 47 (6%); elderly: n = 12 (1%). All the characteristics of the participants are reported in Table 1.
Table 1. Characteristics of the participants.

Sample
n %

Participants

500

Females

311 51

Males

189 49


Age classifications
n %

Young

281 35

Young adults

253 32

Adults

209 26

Senior adults

47 6

Elderly

12 1


BMI levels
n %

Underweight

38 5

Normal weight

565 70

Overweight

199 25

Note: n—number; %—percentage, BMI—Body mass index.
Energy Expenditure
Figure 1a, b shows the total weekly energy expenditure in MET–min/wk of all the sample before and during the COVID-19 quarantine, respectively. Figure 1c shows the difference between MET–min/wk before and during COVID-19 quarantine conditions for all of the sample.
(a) (b) (c)
Figure 1. (a)Total weekly energy expenditure (MET-minute/week) before COVID-19 quarantine. (b) Total weekly energy expenditure (MET-minute/week) during COVID-19 quarantine. (c) Total weekly energy expenditure (MET-minute/week) difference between before and during COVID-19 quarantine.
The related descriptive analysis carried out, reported in Table 2, showed the prevalent PA level of the participants in the before COVID-19 quarantine condition compared to during the COVID-19 quarantine (median: 3006 vs. 1483.8 MET–min/wk, respectively).
Table 2 shows a decrease of 1168.5 MET–min/wk from before the COVID-19 quarantine to during the COVID-19 quarantine. Using the Wilcoxon signed-rank test, we compared PA levels for these conditions and found a significant difference (p < 0.001).
Table 2. Total weekly energy expenditure in MET–minutes/week.
Variable

Min

1st Q

Median

Mean

3rd Q

Max

MET–min/wk
Before the COVID-19 quarantine

12

1752

3006

3458

4815

9990

MET–min/wk during the COVID-19 quarantine

0

627.7

1483.8

1994.3

2896.1

9639

MET–min/wk difference before and during
the COVID-19 quarantine

−7440

61.75

1168.5

1463.51

2650.5

8934

Note: Min, Minimum; 1st Q, 1st Quartile; 3rd Q, 3rd Quartile; Max, Maximum; MET–min/wk, MET–minutes/week.
Concerning the 3 PA categories suggested by the IPAQ recommendations for scoring protocol (i.e., < 600; ≥ 600; ≥ 3000 MET–min/wk), responses analysis for the before COVID-19 quarantine conditions showed 49 low active participants (6%); 352 moderately active participants (44%); and 401 high active participants (50%); meanwhile, the during COVID-19 quarantine condition results showed an increase of 19% of low active participants (n = 200) and an increase rate of 7% of moderately active participants (n = 409), with a related decrease of 26% of high active participants (n = 193).
A comparison between MET–min/wk before and during the COVID-19 quarantine showed a decrease of the total weekly energy expenditure during the COVID-19 quarantine for the 77% of the sample (n = 615).
Energy Expenditure in Relation to Gender, BMI Levels, and Age Classifications: A Bivariate Analysis
Figures 2–4 show the MET–min/wk comparison before and during the COVID-19 quarantine in relation to the gender, BMI and age groups, respectively. Boxplots show the differences between the distributions of the variables considered, which were analyzed separately, as reported below.
For gender, male participants showed a distribution that shifted to higher values of MET– min/wk compared to female participants before the COVID-19 quarantine (Figure 2a). During the COVID-19 quarantine, although both groups reduced their total weekly energy expenditure, male and female groups showed an opposite trend than the previous one: the male’s distribution was slightly shifted to lower values compared to females (Figure 2b). The absolute variation of MET–min/wk (i.e., the MET–min/wk difference between before and during the COVID-19 quarantine) showed the greatest absolute decrease for males, though the 25% of participants had increased the level of PA, regardless of gender (Figure 2c).
(a) (b) (c)
Figure 2. (a) Box plots (median, interquartile range) describe the total weekly energy expenditure (MET-minutes/week) before COVID-19 quarantine in relation to the gender variable. (b) Box plots (median, interquartile range) describe the total weekly energy expenditure (MET-minutes/week) during COVID-19 quarantine in relation to the gender variable. (c) Box plots (median, interquartile range) describe the total weekly energy expenditure (MET-minutes/week) difference between before and during COVID-19 quarantine in relation to the gender variable.
Regarding the BMI levels, analysis showed that before the COVID-19 quarantine, the normal weight group reported the highest median value, while the underweight and overweight groups showed superimposable values (Figure 3a). We also observed that before the COVID-19 quarantine, the greatest variability of PA level distribution appeared for overweight participants, in terms of interquartile difference and range. During the COVID-19 quarantine, all groups showed a reduction in MET–min/wk, while underweight and normal weight participants maintained the shape of the distributions observed before the COVID-19 quarantine, while the distribution of the overweight participants underwent a significant change, with 75% of cases placed on metric levels below 2000 MET–min/wk. That is, the overweight group showed the lowest median PA level during COVID-19 quarantine compared to the other groups (Figure 3b). For the distribution of the MET–min/wk absolute variation for BMI levels, it was revealed that the values of the first and third quartiles were slightly higher for overweight participants compared to those of the other groups, showing a greater contraction of the PA level. The highest MET–min/wk difference between before and during the COVID-19 quarantine was found for the overweight group (Figure 3c).
(a) (b) (c)
Figure 3. (a) Box plots (median, interquartile range) describe the total weekly energy expenditure (MET-minutes/week) before COVID-19 quarantine in relation to the BMI variable. Legend: u, underweight; n, normal weight; o, overweight. (b) Box plots (median, interquartile range) describe the total weekly energy expenditure (MET-minutes/week) during COVID-19 quarantine in relation to the BMI variable. Legend: u, underweight; n, normal weight; o, overweight. (c) Box plots (median, interquartile range) describe the total weekly energy expenditure (MET-minutes/week) difference between before and during COVID-19 quarantine in relation to the BMI variable. Legend: u, underweight; n, normal weight; o, overweight.
Regarding the age classifications, the comparison between before and during the COVID19 quarantine showed a MET–min/wk distribution characterized by different location and dispersion parameters for each group. Figure 4 a, b shows an inverse relationship between PA level and age. Moreover, as observed by the same figures, a...