Perceptions of epidemic risk in Italy and Sweden driven by authorities’ responses to COVID-19
We included data from an anonymous public risk perception survey conducted in Italy and Sweden at two different times of the COVID-19 pandemic. Detailed information about the study has been published elsewhere31. In short, the survey explores the public’s perception of risk for nine threats (epidemics, floods, droughts, earthquakes, wildfires, terrorist attacks, domestic violence, economic crises and climate change). Data was collected over a one-week period in August and November 2020. The samples were independent and drawn from two existing survey panels of 100,000 people in each country, set up by marketing research company Kantar Sifo.32, and should be considered representative of the Swedish and Italian population for gender and age. About 8,000 people from the pool were invited to participate, if they did not respond, up to two reminders were sent. Capital regions were over-represented: with a sampling rate of 1/9 in Italy and a sampling rate of 4/6 in Sweden) (Supplementary Fig. 1). The missing data was quite low,
Individuals who lived in the capital region were over-represented, specific weights were applied in the analysis to take this into account. The present study was approved by the Italian Committee for Research Ethics and Bioethics (Dnr 0043071/2019) and the Swedish Ethical Review Authority (Dnr 2019-03242). The study was carried out in accordance with the ethical standards set by the European Union under Horizon 2020 (EU General Data Protection Regulation and FAIR Data Management). Participants were informed that participation was voluntary and they gave their informed consent to participate in this study when they completed the survey.
Perception of the risk of epidemics
This study examined the public’s perception of the risk of an epidemic by considering seven domains: the probability of epidemics, the impact of the epidemic on the individual and on the population, the preparedness of the individual and authority, knowledge of the individual and authority on epidemics with a Likert-type scale ranging from 1, minimum to 5, maximum.
Predictors of risk perception
Information on direct experience of an epidemic and socio-economic factors such as age, gender, employment (yes versus no), relative income (from 1 to 5), university education ( yes versus no) were collected in the survey and included in the present study as possible predictors of risk perception.
Excess mortality at the regional level in Italy and Sweden during the first wave of the COVID-19 pandemic (from February 15 to May 15 for Italy and from March 1 to May 31 for Sweden) has been taken into account in the study. The regional level has been defined according to the Nomenclature of Territorial Units for Statistics (NUTS) 2 of the European Union33. We retrieved data on excess mortality among Italian regions from Scortichini et al.17. To estimate excess mortality in Sweden, we compared the COVID-19 epidemic to the period before the epidemic. A two-step interrupted time series approach, which relied on a Poisson model with a function that constrains the excess risk to zero at the beginning of March 2020, was used to calculate the excess mortality at the Swedish regional level34. The model was adjusted for time-varying confounders such as (i) seasonality using a natural 3-node spline term, (ii) proxies for the day of the week, (iii) water temperature. air using a term for average daily temperature. Temperature information was extracted from the ERA-5 reanalysis dataset on the Copernicus Climate Data Store35. We performed mixed-effects Poisson regression models with a random term for NUTS3 administrative units to calculate excess mortality at the regional (NUTS2) level taking into account heterogeneity between NUTS3 administrative units.
National policy response
The stringency index18 is a national response index and is used to quantify the measures implemented in response to the COVID-19 pandemic. The Strictness Index is a daily nationwide measure that takes into account nine areas: school closures; workplace closures; cancellation of public events; restrictions on public gatherings; public transport closures; stay-at-home requirements; public information campaigns; internal movement restrictions; and international travel controls. In this article, the level of national policy response was used as a four-level ecological variable (Sweden until August, Italy until August, Sweden until November and Italy until November) and was defined as the area under the stringency index curve for each country, between two successive days until August 5, 2020 (first survey) and November 4, 2020 (second survey). This measure has been standardized on Sweden’s value in August (considered the benchmark).
Possible differences in means and confidence intervals for seven items of risk perception between countries and over time were presented graphically using forest plots and stratified by country and time period. Changing effects by country and time were examined using ordinal logistic regression models with risk perception (independent variables) and country and time as dependent variables. The results were presented as (i) odds ratios (ORs) for each country and period strata, (ii) ORs for the country within the period strata and for the period within the country strata , and (iii) measures of interaction on additive and multiplicative scales.36.
Second, multivariate ordinal logistic regression models were run to assess the association of gender, age, employment, relative income, academic training, and epidemic experience as possible predictors with the seven domains of risk perception (independent variables). The analysis was stratified by country and time period.
Third, we examined whether the perception of risk varied depending on the extent to which an area was affected by the first wave of the COVID-19 pandemic. We compared the means and the confidence intervals for seven risk perception items between the region most affected in terms of excess mortality (Stockholm region in Sweden about 60% excess mortality and Lombardy region in Italy about 100% excess mortality) and comparison with the rest of the country. Next, ordinal logistic regression models were run to examine whether excess mortality at the regional level (dependent variable) was associated with domains of risk perception (independent variables) stratifying by country and adjusting for sex, age and gender. relative income. Finally, the association between the level of measures implemented and risk perception was explored using adjusted ordinal logistic regression models.
The use of ordinal logistic regression models was based on the assumption that the effect was linear on the logarithmic scale and that each independent variable had an identical effect for a one unit increase in the ordinal dependent variable (proportional odds) . Along with this, the goodness of fit of the ordinal logistic models was tested using the deviance goodness of fit test. No multicollinearity between independent variables and correlation between model errors were detected.
As has been suggested in the literature, there are considerable risks in misinterpreting p-values37. Therefore, we chose to interpret the estimates in terms of the possible direction of effects and to use ORs and 95% confidence intervals (CIs), which contain information about significance. Specifically, the width of the confidence interval and the size of the p-value are related: the narrower the interval, the smaller the p-value. In addition, the confidence interval gives additional information related to the magnitude of the studied effect.
Statistical analyzes were performed using Stata version 15.0 (StataCorp, College Station, TX, USA) and R version 188.8.131.52.