The impact of control and mitigation strategies during the second wave of coronavirus infections in Spain and Italy

European countries struggled to fight against the second and the third waves of the COVID-19 pandemic, as the Test-Trace-Isolate (TTI) strategy widely adopted over the summer and early fall 2020 failed to contain the spread of the disease effectively. This paper sheds light on the effectiveness of such a strategy in two European countries (Spain and Italy) by analysing data from June to December 2020, collected via a large-scale online citizen survey with 95,251 and 43,393 answers in Spain and Italy, respectively. Our analysis describes several weaknesses in each of the three pillars of the TTI strategy: Test, Trace, and Isolate. We find that 40% of respondents had to wait more than 48 hours to obtain coronavirus tests results, while literature has shown that a delay of more than one day might make tracing all cases inefficient. We also identify limitations in the manual contact tracing capabilities in both countries, as only 29% of respondents in close contact with a confirmed infected individual reported having been contact traced. Moreover, our analysis shows that more than 45% of respondents report being unable to self-isolate if needed. We also analyse the mitigation strategies deployed to contain the second wave of coronavirus. We find that these interventions were particularly effective in Italy, where close contacts were reduced by more than 20% in the general population. Finally, we analyse the participants’ perceptions about the coronavirus risk associated with different daily activities. We observe that they are often gender- and age-dependent, and not aligned with the actual risk identified by the literature. This finding emphasises the importance of deploying public-health communication campaigns to debunk misconceptions about SARS-CoV-2. Overall, our work illustrates the value of online citizen surveys to quickly and efficiently collect large-scale population data to support and evaluate policy decisions to combat the spread of infectious diseases, such as coronavirus.


S1. SURVEY QUESTIONS
depicts the questions whose answers have been analysed in this paper. The survey has a total of 24 questions. However, there are a few conditional questions such that the number of the questions is that a person might answer could be less than 24. There are gaps in the numbering since during the lifetime of the survey questions regarding the lockdown behaviour were removed as they were no longer relevant.

Question
Possible answers

Community engagement
The cooperation of the population by complying with social distancing and personal protection measures is essential to control the transmission of coronavirus and maintain a low reproduction number [1]. Thus, we analyse the participants' perception of the risk of getting a coronavirus infection associated with performing different activities and/or going to various places (Q15 in SI Table S1). We also analyse the personal protection measures they report adopting to prevent a SARS-CoV-2 infection (Q20 in SI Table S1) and whether they telework regularly (Q17, Q17 1 and Q18 in SI Table S1).
We find some differences in the behaviour of participants in the two countries. Hence, we report the results for each country separately. Figure S1A   protection measures adopted in Spain and Italy. C) Percentage of workers that report teleworking per sector, in Spain and Italy. In this figure, we report the 95% confidence intervals.
of infection. We refer to the SI Figure S14 and Figure S15 for additional details. Figure S1B shows the personal protection measures adopted to prevent a coronavirus infection.
Again, we observe that Spanish respondents are more cautious about COVID-19 and adopt more countermeasures than Italians in all areas but the digital contact tracing app installation, where more Italian respondents report having the app installed than Spanish respondents (Italy: 26%, Spain: 17%). The most notable difference is found in the adoption of ventilation: 48% of Spanish vs 29% of Italian participants report ensuring good ventilation indoors.
We report more information in SI Figure S16 and Figure S17.
Finally, most countries deployed public communication campaigns and programs to promote telework and hence reduce the risk of infection at work. Figure S1C shows the percentage of participants who report remote working in the new normality phase in Spain and Italy. The categories that have more remote workers are Communications, Science/Tech/Professional and Finance, which are also amongst the most digital sectors [2]. Although the average is similar in both countries (around 15%) and considerably larger than the values before the COVID-19 Spanish respondents and women in both countries tend to be more compliant in adopting personal protection measures -such as wearing facial masks-and more conservative in their estimation of COVID-19 risk associated with different activities/places. This identified gender difference is aligned with previous studies on attitudes and behaviours during the COVID-19 pandemic [4] as well as with the literature showing that women tend to be more risk-averse and more open to government interventions than men [5,6]. Overall, these results highlight the challenges posed by behavioural changes in response to new risks and the need for targeted risk communication strategies [4].

Perception of the Government Measures
As the number of confirmed COVID-19 cases rose again after July 2020, governments were   We also observe that the public opinion about a possible second lockdown changed over time in Italy and Spain (Q12 in SI Table S1). Both in Italy and Spain, we find an increase in the percentage of participants asking for a second lockdown as soon as the second wave hit the two countries in October 2020 (see Figure S2B and Figure S2C). Notably, in Italy, the percentage of respondents asking for a second lockdown increased by around 20% from the first week of October to the first week of November 2020; in Spain, the increase was over 20% between the first week of September until the end of the year. We note that our analysis starts on July 31st and June 21st in Italy and Spain, respectively. In A) we report the 95% confidence intervals.

S2. DESCRIPTION OF THE ITALIAN RESTRICTIONS
With the rise of confirmed COVID-19 cases in Italy in early October, the Italian Prime Minister imposed on October 7th 2020, the compulsory use of face masks in public spaces [8]. Between the 13th and the 26th of October, stricter rules were implemented, such as in the opening hours and maximum allowed capacity of restaurants and bars [9][10][11]. Museums, theatres, cinemas, gyms and swimming pools were closed. One month after the rise of the confirmed number of COVID-19 cases, the government defined a three-tiered system on November 6th 2020 [12]. Each region was assigned a tier depending on several quantitative indicators: i) the level of transmission (e.g. the effective reproduction number, the number of positive cases in the past two weeks, the number of novel outbreaks), ii) the resilience and effectiveness of the control strategies (e.g. the rate of positive tests, the number of positive cases that were contact traced, the temporal delay between the symptoms onset and the COVID-19 test outcome), and iii) the burden on the healthcare system (the hospital occupancy rate for COVID-19 cases, the number of intensive care units). High schools were closed on November 6th, and students were taught remotely; universities were encouraged to teach remotely.

S3. DESCRIPTION OF THE SPANISH RESTRICTIONS
Spain did not have a centralised COVID-19 mitigation strategy after the central government lifted the state of alarm on June 21st 2020 [13]. Hence, during the time period between June 21st and October 25th 2020, each of the 17 autonomous regions made their own decisions related to the pandemic, imposing restrictions as needed. However, in October 2020, all the autonomous regions in Spain (except for Madrid and the Basque Country) decided to adopt a common system to determine the pandemic risk via a set of indicators and implemented the same mitigation measures to slow the spread of COVID-19. They agreed to adopt a 5-level risk system ranging between new normality (no risk) to extreme risk based on their healthcare occupancy levels, the cumulative incidence, the positivity rate and the percentage of traced cases. Moreover, on October 25th, 2020, the government declared again a state of alarm [14], which enabled establishing a national curfew and gave power to local authorities to ban travel across regions, provinces and municipalities if needed.
In terms of the measures, while every Autonomous Region decided to adopt their own, they were similar across regions and included limiting the mobility in the evening and at night, implementing regional/local confinements, reducing the maximum number of people in social meetings to 4-10 individuals, limiting the maximum capacity in locales to 25-50% and partially or totally closing shops in non-essential sectors, restaurants and bars.

S4. LOESS ROBUSTNESS
In the main paper, we use the LOESS method to estimate the Italian trend of the capacity to trace positive people. The method we use is not robust to the effects of outliers (like other least squares methods) but allows to estimate the confidence intervals of the trend.
We here verify whether the iterative robust LOESS regression [15] has different outcomes. Figure S3 shows that the two methods produce very similar trends and have a Spearman's r correlation of 0.95 pvalue<0.0001. Thus, our regression is not greatly influenced by outliers.       FIG. S14. Perception of safety per place, country, and demographic group during the Phase I -new normality.         Figure) with a Theil-Sen regression estimator. Figure S21 shows the trends of the weekly close contacts from outside the household fitted through a Theil-Sen regression estimator [16]. More formally, we interpret the number of weekly