Methods
Study Design
Diet and alcohol consumption, physical activity, use of tobacco and nicotine products, body weight, height, and educational level were monitored using questionnaire data collected through a mixed-mode approach. Computer-assisted telephone interviews (CATI) served as the primary method, while computer-assisted web interviews (CAWI) were offered as a secondary option. The online questionnaire was introduced in the 2024 data collection to improve response rates (Greene et al. 2008). To maintain comparability with the 2014 data collection, the distribution of data collection methods was planned to be 75% via telephone interviews and 25% via online questionnaires.
In each Nordic country, a simple random sample of adults and children was drawn from national population registers. The sample included adults aged 18–65 and children aged 7–12 (from registers as of 1 October 2023); in Sweden, parents of 7–12-year-olds were sampled, and when more than one child qualified, the child with the most recent birthday was selected. The target sample size was 1,000 adults and 500 children from each of the five Nordic countries.
Study Design Questionannaire Statistical Methods References |
Study Design
Diet and alcohol consumption, physical activity, use of tobacco and nicotine products, body weight, height, and educational level were monitored using questionnaire data collected through a mixed-mode approach. Computer-assisted telephone interviews (CATI) served as the primary method, while computer-assisted web interviews (CAWI) were offered as a secondary option. The online questionnaire was introduced in the 2024 data collection to improve response rates (Greene et al. 2008). To maintain comparability with the 2014 data collection, the distribution of data collection methods was planned to be 75% via telephone interviews and 25% via online questionnaires.
In each Nordic country, a simple random sample of adults and children was drawn from national population registers. The sample included adults aged 18–65 and children aged 7–12 (from registers as of 1 October 2023); in Sweden, parents of 7–12-year-olds were sampled, and when more than one child qualified, the child with the most recent birthday was selected. The target sample size was 1,000 adults and 500 children from each of the five Nordic countries.
In Denmark, sampling was provided by the Danish Health Data Authority (Sundhedsdatastyrelsen) and phone-number enrichment by Bisnode–Dun & Bradstreet. In Finland, sampling was provided by the Digital and Population Data Services Agency (Väestörekisterikeskus) and enrichment by Profindr. In Iceland, sampling was provided by Registers Iceland (Þjóðskrá) and enrichment by Gallup Iceland. In Norway, sampling was provided by the Norwegian Tax Administration (Skatteetaten) and enrichment by Bisnode–Dun & Bradstreet. In Sweden, sampling was provided by the State Personal Address Register (SPAR; Statens personadressregister) and enrichment by ILR Media.
Norstat Denmark coordinated the overall data collection and collaboration across countries. Local Norstat offices conducted data collection in Denmark, Finland, Norway, and Sweden, while Gallup Iceland was responsible for data collection in Iceland.
Only individuals enriched with a phone number were included in the invitation process. Adult participants and parents or guardians of child participants received an invitation letter via SMS one week before interview attempts began. Interviewers made up to six contact attempts by phone at varying times. After three attempts, a follow-up SMS with a link to the online questionnaire was sent. If unanswered within 48 hours, calling resumed. After six total attempts without contact, the case was closed.
Questionnaire
The questionnaire in the recent survey (see Appendix 1 and Appendix 2) was a modified version of the validated NORMO 2009 questionnaire (Fagt et al. 2012). It consisted of 17 questions for adults and 13 for children. The child version excluded questions related to alcohol consumption and use of tobacco and nicotine products.
The dietary section of the questionnaire used in the 2024 data collection consisted of a Food Frequency Questionnaire (FFQ), in which participants were asked to reflect on their food and beverage intake the last 12 months and report the average frequency of their intake per day, week or month. Alcohol consumption was assessed through questions on frequency of intake over the past 12 months, the number of alcoholic drinks consumed per day during the last seven days, and occasions of binge drinking (defined as ≥5 units) during the past month. The physical activity section captured hours and minutes spent on both “moderate or harder” and “vigorous” leisure-time physical activity during the past week. It also included active transportation (walking and cycling), as well as time spent on leisure screen activities, which was used as an indicator of sedentary behaviour. Following this, participants were asked about their use of tobacco and nicotine products, with separate items addressing smoking, snuff, nicotine pouches, and e-cigarette use. The final sections of the questionnaire included items on educational attainment and self-reported height and weight.
An English master version of the questionnaire was translated into the five Nordic languages by the respective national working groups. Translations were then compared by the working group. A feasibility and pilot study was carried out in Norway and Sweden prior to the main data collection.
Changes from 2014 to 2024
The questionnaire for adults and children used in 2024 underwent susbstantial revisions compared to the questionnaire used in 2014. Question structures and wording were simplified to improve clarity and reduce response time. Control filters were introduced to enhance data quality. Redundant items were removed, while new, more relevant questions were added — including items on sustainability, such as meat and pulse consumption, dairy intake, and active transportation. Country-specific examples were incorporated throughout, particularly in questions about fats, dairy products, bread types, and energy drinks. Screen time questions were updated to reflect current technology use. The adult questionnaire included a new item on weekly alcohol consumption and expanded the tobacco section to cover e-cigarettes and nicotine pouches. Sensitive questions on weight and height were moved to the end of the questionnaire. Finally, the education section was revised using ISCED-inspired categories, supplemented with country-specific examples. See Appendix 3 for a detailed overview of the revisions.
Statistical Methods
Survey weights
Sample weights were developed to improve the representativeness of each country's sample based on age, gender, and education level (parent or guardian education for the child survey). Full details on the weighting procedure are described in Appendix 4 .
Survey measures
Dietary intake
Dietary outcomes were based on self-reported consumption frequencies. Composite variables were created where appropriate. Outcomes included total fruit and vegetable intake (excluding juice), fruit, vegetables (including pulses), pulses (not assessed in 2014), whole grain bread (hard, rye, and whole wheat), fish, meat (including subtypes), dairy (cheese and milk products), discretionary foods (cake, pies, biscuits, chocolates, candy), discretionary drinks (sugary beverages, energy drinks, and sugar-free beverages), and juice.
Alcohol use
Alcohol consumption frequency was assessed over the past 12 months. Binge drinking was defined as consuming ≥5 drinks on one occasion at least once in the past month. The number of alcoholic drinks consumed per day was assessed over the last seven days (2024 only). Individuals reporting ≥30 drinks in any one day were excluded as outliers.
Physical activity and sedentary behaviour
Physical inactivity was defined based on WHO 2020 guidelines. Vigorous activity was weighted double compared to moderate activity (i.e., 75 minutes of vigorous activity was equivalent to 150 minutes of moderate activity). Additional outcomes included screen time, total moderate-to-vigorous physical activity, and time spent walking or cycling. Implausible values were excluded: MVPA > 35 h/wk (adults, children), VPA > 21 h/wk (adults), and leisure screen time > 18 h/day (adults, children); out-of-range values were flagged only for children. Screen time measures from 2014 and 2024 were not directly comparable, so only 2024 data were reported.
Tobacco and nicotine use
Tobacco-related outcomes included daily smoking, smokeless nicotine use (snuff and nicotine pouches), and e-cigarette use. Composite indicators were created for any daily use. Smokeless nicotine was not assessed in Denmark in 2014; e-cigarettes were not assessed in any countries in 2014.
Anthropometrics
Self-reported height and weight were used to calculate BMI. Values outside expected age-specific ranges were flagged as extreme and excluded. Due to inconsistencies in units, height values outside the expected range in centimeters were checked against meters, millimeters, and inches; if within range after conversion, they were retained. Adults were classified as having overweight if BMI was 25–30, and having obesity if BMI >30. Children were classified using age-specific cutoffs.
Educational level
For adults, education was categorized as low (primary education, ≤10 years), medium (upper secondary education), or high (higher education, including short-cycle tertiary education or above). For children, education was defined as the highest educational attainment of the parent or guardian completing the interview and was categorized as low/medium (primary education, ≤10 years, or upper secondary education) or high (higher education, including short-cycle tertiary education or above).
Statistical analysis
Survey weights were applied within each country and then rescaled so that each country contributed equally to the pooled Nordic average, regardless of population size. All analyses were conducted using R version 4.4.2 (2024-10-31) and relevant packages, including survey, srvyr, emmeans, dplyr, and ggplot2. Survey-weighted generalized linear models (svyglm) were used to estimate overall means and group differences in continuous and categorical outcome variables. Models were run separately within each country and for the pooled Nordic Region. Nordic means were only presented for outcome variables measured with equivalent survey questions across all countries.
Comparisons were conducted for adults across gender (male, female), age group (18–29, 30–44, 45–65), education level (low, medium, high), and NORMO survey year (2014, 2024). For children, comparisons were made across gender (male, female), parent/guardian education level (low/medium, high), and survey year (2014, 2024). For children, the low-education group was too small for separate analysis and was combined with the medium-education group. Global p-values were calculated using regTermTest() from the survey package to test the overall significance of each grouping variable. Estimated marginal means and pairwise comparisons were obtained using emmeans(), with 95% confidence intervals computed using design-based standard errors that accounted for the complex survey design. For each analysis, all available observations with non-missing values for the specific variable being analyzed and the relevant grouping variable(s) were included (i.e., a complete case approach per outcome variable).
However, the analyses were descriptive and did not include statistical adjustments for potential confounding by age or gender, which may influence some observed differences between countries or subgroups. As such, results should be interpreted as population-level patterns rather than causal relationships
Pairwise comparisons were only performed after a statistically significant global test. Due to the large number of dietary questions, p < 0.001 was used as the cut-off for statistical significance. For all other data questions, p < 0.05 was used.
Because data collection seasons differed between 2014 and 2024 in all countries except Finland (see chapter on Study participation and composition), we limited cross-year comparisons accordingly. Measures with longer recall periods—diet (12 months), alcohol frequency (12 months), binge drinking (30 days), and tobacco/nicotine (current/ever use)—were compared across years. Physical activity and sedentary behaviour was assessed over the prior seven days, so cross-year analyses were conducted only for Finland, and no Nordic averages were presented for physical activity. The number of alcoholic drinks per day over the past seven days was not assessed in 2014.
For dietary data visualizations, country-level mean intake frequency of selected dietary variables in 2024 were expressed as a percentage relative to the pooled Nordic average (set to 100%). This normalization approach facilitates visual comparisons across countries by highlighting deviations from the regional average.
References
Fagt, S., Andersen, L. F., Anderssen, S. A., Becker, W., Borodulin, K., Fogelholm, M., Groth, M. V., Gunnarsdottir, I., Helakorpi, S., Kolle, E., Matthiessen, J., Rosenlund-Sørensen, M., Simonen, R., Sveinsson, T., Tammelin, T., Thorgeirsdottir, H., Valsta, L., & Trolle, E. (2012). Nordic monitoring of diet, physical activity and overweight: Validation of indicators (TemaNord 2011:556). Nordic Council of Ministers. https://norden.diva-portal.org/smash/get/diva2:702049/FULLTEXT01.pdf
Greene, J., Speizer, H., & Wiitala, W. (2008). Telephone and web: Mixed-mode challenge. Health Services Research, 43, 230–248. https://doi.org/10.1111/j.1475-6773.2007.00802.x