This is a cross-sectional study that uses primary data collected in 6 of the 15 districts of Oslo municipality – Alna, Bjerke, Gamle Oslo, Grorud, Sondre Nordstand and Stovner. These districts are located in the eastern part of Oslo and were chosen due to being more socioeconomically deprived and with a higher migrant population (Steinmetz 2022). Oslo is located in the southern part of the country and has a total population of 698,660 individuals. Oslo’s population density is over 1500 inhabitants per square kilometre according to the latest projections of September 2021, making it the most populated area of the country. Oslo was the epicentre of the COVID-19 pandemic in Norway (Mamelund and Dimka 2021) with a rate of over 22,400 reported cases per 100,000 inhabitants (as of 25 January 2022) (NIPH 2022).

The Centre for Research on Pandemics & Society (PANSOC) at Oslo Metropolitan University, in collaboration with the Pandemic Centre at the University of Bergen developed a questionnaire to be conducted as a web-based survey among residents of Oslo. The survey consisted of closed-choice Likert-type items, multiple answer questions and some open-ended questions in nine sections. A private company – Kantar – administered the survey on behalf of the researchers. Kantar has access to a population database with all phone numbers where they sampled 60,000 telephone numbers, 10,000 in each of the six parishes.

Text (SMS) messages, written in English, were sent to almost 60,000 potential participants. Recipients were directed to a web link where they were provided with further instructions and contact information, asked to consent to the survey, and allowed to select their preferred language – English, Arabic, Polish, Somali and Urdu – encourage responses from immigrants who may not have responded to a Norwegian-language survey. The SMS did not include any information on the survey content; therefore, the percent of complete and incomplete responses concern only those who accessed the survey. The final sample includes 5442 participants, for a response rate of around 9%, which might seem low but still allows for computing reliable statistical analysis with scientifically valuable and significant results (Steinmetz 2022).

Participants were eligible for a drawing for three gift cards valued at NOK 1000 (approximately 100 euros) each. The survey was approved by the Regional Committees for Medical and Health Research Ethics (REK Regional Ethics Committee, approval no 250310). Responses were collected between the 16th and the 24th of June 2021.

This study uses a subsample of respondents (n = 4928) for which body mass index can be computed. Participants were asked to report their own height (in cm) and weight (in Kg). Of the 5442 participants in the total data set, 201 did not report height, 121 did not report weight and 175 reported neither. An additional 17 cases were excluded because the height reported was ≤ 100 cm or because the reported weight was ≤ 33 kg; these values are unlikely in adults. For the remaining subsample, BMI was computed, and participants were categorized according to WHO cut-off points into individuals with underweight (BMI < 18.5), healthy weight (BMI between 18.5 and 24.9) and overweight/obesity (BMI ≥ 25) (Ahmed et al. 2018). However, the group of individuals categorized as underweight were not considered in the subsequent statistical analysis, as these only represented 1.5% (n = 79) of the sample.

The other variables included in this study are sex (female, male or no answer), age (collected as numerical data and then recoded into four classes: 18–29 years, 30–44 years, 45–59 years and 60 or more years), highest completed education level (primary, higher general, higher vocational, vocational school/vocational education, university (less than 4 years) or university (4 years or more)), income (up to 399,999kr, 400,000—799,999kr or 800,000kr or more), employment status (full-time job, part-time job, self-employed, retired, unemployed, social security, student or stay-home), district of residence (Alna, Bjerke, Gamle Oslo, Grorud, Sondre Nordstand or Stovner), whether individuals had a confirmed case of COVID-19 any time before filling the questionnaire (yes or no) and whether they were born in Norway (yes or no). Those who were not born in Norway were considered migrants.

Chi-square tests were computed to compare the proportions of individuals with healthy weight and overweight/obesity, and individuals that had COVID-19 or not, by sex, age, schooling level, income, employment status, district of residence and whether they were born in Norway. The percentage of COVID-19 cases was computed for the total sample (number of COVID-19 cases/total sample*100), and separately for the group of individuals with healthy weight and the group of individuals with overweight/obesity, per district of residence.

Nine logistic regression models were computed using as dependent variable having COVID-19 and as independent variable weight status and other sociodemographic factors. In model 1, only weight status (having overweight/obesity) was included. In models 2 to 8, weight status and, respectively sex, age, schooling level, income, employment status, district of residence and born in Norway were included. Model 9 includes all variables. Additional logistic regression models using the Backwards Stepwise (conditional) method were computed to identify the sociodemographic factors associated with having COVID-19 using BMI as categorical and as continuous variables. Results are expressed as odds ratio (OR) and the respective 95% confidence interval (CI).

All statistical analysis were performed using SPSS (v. 27), and significance level was set at p < 0.05.