Short-term effects of mitigation measures for the containment of the COVID-19 outbreak: an experience from Northern Italy

Publication date: 29/03/2020 – E&P Code: repo.epiprev.it/257
Authors: Giovenale Moirano,¹ Maurizio Schmid,² Francesco Barone-Adesi³

Abstract: In the last weeks several countries implemented mitigation measures in the attempt to curb the COVID-19 epidemic. While these interventions showed to be effective in Wuhan (China), it is not clear whether these results are directly applicable to other countries (1). We conducted a first evaluation of the effectiveness of such measures in a small Italian area, taking advantage of the fact that it introduced the mitigation measures two weeks ahead the rest of the country.

Following the detection of the first cases of COVID-19 in Lombardy, the Italian Government enforced different policies to contain the local outbreaks. On February 23^rd, a total lockdown was issued for 10 municipalities in the Lodi Province (Lombardy Region), and the measures included:  a) strict home confinement to the entire population; b) closure of all the non-essential commercial activities; c) mobility restrictions related to the involved municipalities (2).

These policies were later extended to the whole of Lombardy (March 8^th) and to the entire Country (March 9^th). The early implementation of such regulations in these municipalities (hereafter defined as the “red zone”), allows a sufficient observation time to conduct a thorough evaluation of their effect.  Specifically, we investigated changes in the time-varying reproductive number R_t namely the estimate of the average number of secondary cases that each infected individual would infect if the conditions remained as they were at time t (3).
A decrease of R_t over time provides insights on the effectiveness of interventions, given that the goal of control efforts is to reduce it below the threshold value of 1. Data on incident cases of COVID-19 recorded between February 28^th and March 27^th were obtained from the website of the Italian Civil Protection (4).
As data at the municipality level are not available, we used data from the whole Lodi province and  two neighboring provinces  to indirectly estimate the daily number of cases in the red zone. Observed rates in the whole Lodi province are, indeed, a weighted mean between the rates in the red zone (accounting for 22% of the population of the Province) and the rates in the other municipalities (the remaining 78% of the provincial population). We assumed that the municipalities of the Lodi Province out of the red zone had incidence rates similar to the neighboring provinces (Milan and Pavia). Applying incidence rates of Milan and Pavia provinces to the 78% of Lodi Province population, we could calculate the expected number of new cases occurring in the red zone. We then estimated the time-varying reproductive number, R_t, using the R package EpiEstim for the red zone (5).
Figure 1 shows that incidence in the red zone increased until March 5^th, when a steady reduction started. Coherently, there was a drop in the values of R_t which decreased from 2 to 0.9 and remained stable afterwards. As the incidence rates in the provinces of Milan and Pavia, which we used to approximate the rates in the municipalities of the Lodi province out of the red zone, are among the lowest in the Lombardy Region, our results should be regarded as conservative. Had we applied the rates of other provinces, the estimated reduction in R_t would have been even larger.
These results are consistent with what observed in Wuhan Province, China (5) and suggest that the effects of a lockdown can be appreciated in about two weeks, a time range that is consistent with the COVID-19 incubation period combined with the delay of the diagnosis after symptoms onset (1).
Should this be the case, it is plausible to expect a similar reduction in the countries that recently introduced similar measures, if population compliance to the enforced containment measures is similar to that seen in the red zone. These results offer the hope that mitigation measures similar to those issued by the Italian Government can influence the progression of local transmission of COVID-19. They also provide support to the implementation of analogous policies in other countries.

Topic: COVID-19

Key words: COVID-19,

AVVERTENZA. GLI ARTICOLI PRESENTI NEL REPOSITORY NON SONO SOTTOPOSTI A PEER REVIEW.

Cite as: Giovenale Moirano, Maurizio Schmid, Francesco Barone-Adesi (2020). Short-term effects of mitigation measures for the containment of the COVID-19 outbreak: an experience from Northern Italy. E&P Repository https://repo.epiprev.it/257

References

1. (No Title) [Internet]. [cited 2020 Mar 29]. Available from: www.who.int/docs/default-source/coronaviruse/who-china-joint-mission-on-covid-19-final-report.pdf
2. Italians struggle with “surreal” lockdown as coronavirus cases rise | World news | The Guardian [Internet]. [cited 2020 Mar 29]. Available from: www.theguardian.com/world/2020/feb/24/italians-struggle-with-surreal-lockdown-as-coronavirus-cases-rise
3. Thompson RN, Stockwin JE, van Gaalen RD, Polonsky JA, Kamvar ZN, Demarsh PA, et al. Improved inference of time-varying reproduction numbers during infectious disease outbreaks. Epidemics. Elsevier B.V.; 2019 Dec 1;29.
4. COVID-19 ITALIA – Desktop [Internet]. [cited 2020 Mar 23]. Available from: opendatadpc.maps.arcgis.com/apps/opsdashboard/index.html#/b0c68bce2cce478eaac82fe38d4138b1
5. Cori MA. Package “EpiEstim” Title Estimate Time Varying Reproduction Numbers from Epidemic Curves. 2019 [cited 2020 Mar 23]; Available from: orcid.org/0000-0001-5797-6803
6. Nishiura H, Linton NM, Akhmetzhanov AR. Serial interval of novel coronavirus (2019-nCoV) infections. medRxiv [Internet]. 2020 [cited 2020 Mar 23];2020.02.03.20019497. Available from: medrxiv.org/content/early/2020/02/17/2020.02.03.20019497.abstract

Info

Affiliations:
1 Cancer Epidemiology Unit, Department of Medical Sciences, University of Turin, 10128, Turin, Italy
2 BioLab³ – Engineering Dept. – Roma Tre University, 00146 Rome, Italy
³ Department of Translational Medicine, University of Eastern Piedmont, 28100 Novara, Italy

Authors’ contributions: FBA, GM and MS conceived the study; FBA and GM carried out the statistical analysis; FBA and MS and GM drafted the final version of the manuscript

Conflict of Interest: none declared

Role of funding source: None.

Ethics committee approval: The study was based on publicly available aggregate data. No Ethics committee approval was necessary.

Copyright: The copyright holder for this preprint is the author/funder, who has granted E&P Repository a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license

---

Figure 1. Upper Panel: Incidence rates (per 10,000 inhabitants ) of COVID-19 cases in the red zone from February 28th to March 27th. Lower Panel: Time-Varying reproduction number 𝑅𝑡 in the red zone. 𝑅𝑡 estimates are based on a 7-day sliding window and assuming a serial interval with a mean of 4.7 days and a standard deviation of 2.9 days (6)

---

References & Citations
Google Scholar