Regional policies are not enough to comprehensively combat risk of covid-19 infection, study finds

In a new article published in the The Lancet Regional Health – Europe, Mr Gabriele Guaitoli and Dr Roberto Pancrazi from the Department of Economics at the University of Warwick are investigating whether regional non-pharmaceutical interventions introduced in Italy during the second wave of Covid-19 in the fall of 2020 were effective in eliminating the local risk factors.

Using Italy as a case study, the researchers studied the impact of the country’s regional lockdown interventions on the spread of the virus during the “second wave” which began around November 2020.

their paper Covid-19: Regional policies and local infection risk: Evidence from Italy with a modeling study first identifies a small set of pre-pandemic socio-demographic variables that were most strongly correlated with the incidence of Covid-19 in the Italian provinces. They were:

• Temperature – higher temperatures correlate with a lower incidence of covid-19
• Share of agricultural employment Population – a greater share of agricultural employment correlates with a lower incidence of covid-19
• Number of cases per capita in wave one – the higher the number, the lower the incidence of covid-19 in wave two
• Share of employment in services Population – more people working in the service economy correlate with a higher incidence of covid-19
• Income per capita – higher income, an indicator of more intense economic activity, correlated with a higher incidence of covid-19
• Share of large households in region – regions with more households with more than 5 members had a higher incidence of covid-19
• Travel by public transport – areas with more concentrated public transport use experienced a higher incidence of covid-19

These variables were measured before the pandemic and at the provincial level. This means that there are predetermined local factors that strongly correlate with the spread of the virus. “

Dr Roberto Pancrazi, Department of Economics, University of Warwick

After identifying these important local risk factors, the authors investigate whether the introduction of non-pharmaceutical policies at the regional level eliminated these risk factors.

Dr Pancrazi continues: “The influence of these variables is still visible in the data, even after the implementation of regional policies. It was only for the most stringent regional policies, similar to UK lockdowns, that we could not find a relevant role for these local factors.

“This suggests that regional policies – limited blockages – were not sufficient to comprehensively address the risk of covid-19 infection linked to predetermined characteristics of the province. In other words, regional policies Italians were not “local enough” to fully address covid-19 risk differences. “

Although the study was carried out using Italian data, the authors believe the results will be of interest to policymakers in many countries, as demographic factors such as temperature, use of public transport, size of the family and nature of the job are not -specific.

Dr Pancrazi added: “Our findings are important for policy makers and have implications for public health. They justify the design of interventions that better target areas at high risk of contagion due to local factors, while allowing the relaxation of restrictions in areas with lower risk of infection.

By showing the relevance of local risk factors that have not been taken into account by regional policies, our study provides a rationale and possible orientations for the implementation of very localized interventions, should these again become required. “

Mr. Guaitoli explains: “Our study suggests that Italy could achieve a better balance between public health outcomes and economic damage. Low-risk provinces in high-risk regions were placed under severe restrictions, while high-risk provinces in low-risk regions were initially placed under light restrictions. Only a few regions have tried to target – in addition to national laws – highly localized Covid-19 clusters. “