Setting and design
The Impact of Covid-19 on Physical Activity, sedentary and dietary behaviour of School-going Adolescents (ICPASA) study will be carried out in schools of Mohali (SAS Nagar) city of Punjab state in the northern region of India and adopts a ‘natural experiment’ approach. The UK Medical Research Council (MRC) defines natural experiment as ‘events, interventions or policies which are not under the control of researchers, but which are amenable to research which uses the variation in exposure that they generate to analyze their impact’(22). In this study, adolescents are exposed to COVID-19 related governmental measures of school closures, social distancing and home confinement which act as a natural experiment. Thus, the current study is a before-after experimental study design in which subjects are measured before the pandemic during baseline and exposed to COVID-19 as a natural experiment starting March 22, 2020. The baseline measurements were conducted in September 2018- February 2019, while follow-up measurements will be conducted in 2021. The study uses quantitative methods to measure the change in proportion of adolescents performing physical activity as per WHO guidelines, and their sedentary and dietary behaviours, before and during COVID-19 pandemic and associated school closures. The study also employs qualitative measures to understand the determinants for changes in lifestyle behaviours.
Recruitment
Following permission by the Secretary, School Education, Department of Education, Punjab in June 2018, a list of public and private schools in Mohali (SAS Nagar) district was obtained from them. All schools falling under the area of SAS Nagar Municipal Corporation were identified as eligible for inclusion. A total of 20 schools (10 government and 10 private schools) were randomly selected from the school list. The Department of Education, Punjab, sent a letter to the selected schools informing about the study following which the school heads were approached to discuss the study and give their consent. Since school support is necessary, participation of the schools was on voluntary basis. One of the ten private schools which were not willing to participate was randomly replaced with another school from the list. All school-going adolescents studying in schools of SAS Nagar (Mohali) city between the ages of 10–19 years were eligible for inclusion. The course of the study is illustrated in Fig. 1.
Students who were studying in sixth to eighth standard at the time of baseline data collection (September 2018 – February 2019) were selected from each participating school. Consent forms and information sheet were distributed to 1308 students who were asked to discuss with their families about participation in the study. Only those students who brought back signed consent forms from their parents/guardians were eligible for inclusion in the study and were asked to fill assent forms. About 83% students (n = 1086) submitted both consent and assent forms and participated in the baseline measurements. In middle schools which enroll students up to eighth standard, students from sixth standard were purposively selected in order to be able to repeat the measurements with the same subjects at the end of the study.
Power Calculation
The baseline proportion of students reaching PA levels of ≥ 60 minutes/day in the present sample is 35.4%. Therefore, 0.354 is taken as the true proportion of the population. Although studies have shown a steep decline in physical activity levels of children and adolescents during school closures as a result of the COVID-19 pandemic, the baseline PA levels in the current sample is quite low. Thereby, expecting a smaller decrease of 6.5% (14) in proportion of physically active adolescents, the null hypothesis proportion is assumed as 0.289. Due to COVID-19 situation, resultant school closures, and limited access to mobile and internet, we are assuming a high drop-out of 45% in follow-up measurements as compared to baseline. Therefore, power calculation is done for 55% of 1086 adolescents, i.e. approximately 600 adolescents. If pre and post measurements are recorded for 600 adolescents, we have a power of about 91%, which is a good power to observe the differences over time that occurred as a result of COVID-19.
Study Outcome
Primary Outcome
The change in proportion of adolescents reaching the recommended level of physical activity using self-reported measures as per WHO guidelines during COVID-19 will be the primary outcome of the study.
Secondary Outcome
Secondary outcomes of the study include lifestyle changes in adolescents in terms of physical activity and sedentary behaviours, and dietary modifications, using self-report measures; objective measures of physical activity (assessed in a subsample); changes in health parameters, particularly BMI and waist circumference will be assessed as indicators of undernutrition and overweight/obesity. The determinants of change in lifestyle behaviours of adolescents as described during the qualitative interviews and focus group discussions (FGDs) are also a secondary outcome.
Data Collection Tools For Baseline Measurements
Measurement of physical activity, sedentary and dietary behaviours
The study adopts a ‘natural experiment’ approach in which baseline measurements were done for the study planned earlier, while follow-up measurements will be done while the governmental measures to prevent the spread of COVID-19, such as social distancing and online schooling are still in place. Baseline measurements included quantitative assessment of physical activity and dietary behaviour using a modified version of GSHS-2006 questionnaire. The Physical Activity Module of the Global School based Student Health Survey (GSHS) Questionnaire 2006 is used as a subjective measure for the physical activity and sedentary behaviour in school-going adolescents. The physical activity module was modified to meet the objectives of the study. Questions for leisure time physical activity were included from the BAECKE questionnaire (23). Questions for recess time PA were added from PAQ-C questionnaire (24). Dietary behaviours module of the GSHS 2006 questionnaire is used as it is. The questionnaire was pre-tested in a similar population by self-administration method. It was translated and back translated into vernacular language (Punjabi) and pilot tested.
Objective Measure Of Physical Activity By Accelerometry
The ActiGraph wGT3X-BT, which is a triaxial accelerometer, is used to objectively measure the physical activity of adolescents. Schools were chosen based on convenience sampling as only seven schools who agreed that adolescents wear accelerometers around their waist for seven days were given the ActiGraphs. The selection of students wearing the ActiGraph was also on voluntary basis. A sub-group of students (n = 233) were given verbal instructions to wear ActiGraphs for 7 days mounted on the right side of the waist in front of the right hip at all times except during sleeping and performing water activities such as bathing, swimming etc. Efforts were made to equally distribute the ActiGraph among government and private schools. The Actigraphs were distributed to the adolescents one school after another, by initializing for date and time using the ActiLife6 software. The raw acceleration data will be converted into objective activity with the help of ActiLife6. Outcome measures of interest are duration of sedentary time, light physical activity, moderate to vigorous physical activity (MVPA), metabolic equivalents (METs), average step count, and counts per minute (CPM).
Anthropometry Measurements
Objective anthropometric measurements were performed on study participants. A flat surface inside the classroom or infirmary of the school was chosen for weight and height measurements. Weight of the adolescents was measured in kilograms up to one decimal point using Seca 813 digital flat scale. Students were asked to be in light clothing i.e. remove sweaters/blazers, belts, and shoes before taking measurements. Height was measured in centimeters using Seca 213 portable stadiometer. Light clothing and being barefoot will be followed for height measurement as well. Students were instructed on the correct posture for height measurement; to touch the heels to the back of the stadiometer wall, legs positioned straight, shoulders relaxed and head looking straight ahead. The research staff was trained to note the height of the pupil avoiding any parallax. Waist circumference was measured in centimeters using the Seca 203 measuring tape. Research staff explained the procedure to the adolescents and asked them if they are comfortable in lowering their pants/skirts for the measurement. Upon their consent, waist circumference was measured individually with each adolescent standing in a relaxed position and breathing normally. Measurement was done on the right side of the pupil after they exhale. Any parallax was avoided. In case of uncertainty in any of the measurements, the result was recorded three times and the average will be taken as the final measurement. Body Mass Index (BMI) of adolescents will be calculated using individual measures of weight and height and manually applying the formula for calculation.
Interviews
Qualitative data was collected in the form of focus group discussions (FGDs) and semi-structured interviews with the participants of the study, that is, the adolescents, school staff, and parents. In this study, FGDs were held with the adolescents, their parents and school staff after completion of baseline measurements to aid in better understanding of the determinants of physical activity and sedentary behaviour among adolescents.
Data Collection During Covid-19
Follow-up measurements will be conducted with the same population as baseline in the form of online questionnaires, and anthropometric measurements will follow as schools start to reopen with the recommended COVID-19 restrictions. Schools that participated during baseline will be invited for follow-up measurements. All measurements conducted during baseline, will be repeated following the same protocol, with some adjustments. The questionnaire will be web-based and schools will be asked to share the web-link of questionnaire on their preferred platform of online information sharing. Schools can choose whether they prefer web-based or paper-based questionnaires. The GSHS-2006 questionnaire is modified to include questions on lifestyle changes in adolescents due to COVID-19. It is created on Google Forms in both English and Punjabi language and has been pre-tested on a similar population before roll-out in the study participants. Students can choose their preferred language of the survey. Physical activity will be objectively measured using ActiGraph over a 7-day period in the same sub-sample. Sanitization of the accelerometers will be done after each use. All anthropometric measurements (weight, height and waist circumference) will be repeated in follow-up using the same procedure as in baseline measurements. The latest guidelines issued by the government to prevent the spread of COVID-19 will be followed. Focus group discussions and interviews will be conducted with students, parents and school staff from the sample population, but not the same as that during baseline, to supplement quantitative data and understand the reason for changes in lifestyle behaviour of adolescents that may have arisen, but are not limited to socioeconomic status, school type and curriculum. It will also aim to discover innovative approaches of staying active during the COVID-19 era of physical distancing.
Data Analysis
The causal inferences of a natural experiment of COVID-19 related governmental measures of school closures, social distancing and home confinement will be derived related to physical activity and lifestyle changes in adolescents. Data will be analyzed using SPSS software version 26. Data will be tested for normality using the Shapiro-Wilk test. Before-after changes in physical activity levels, sedentary and dietary behaviours of adolescents will be analyzed using paired t-tests for continuous data, and the McNemar tests for categorical data or proportions. Qualitative data will be analyzed using thematic analysis.