A statistical model to identify excess mortality in Italy in the period 2011-2022 (preprint)

Excess mortality is defined as an increase in the number of deaths above what is expected based on historical trends, hereafter called baseline. In a previous paper, we introduced a statistical method that allows an unbiased and robust determination of the baseline to be used for the computation of excesses. A good determination of the baseline allows us to efficiently evaluate the excess of casualties that occurred in Italy in the last 12 years and in particular in the last 3 years due to the Coronavirus Disease 2019 (COVID-19) epidemic. To this extent, we have analyzed the data on mortality in Italy in the period January 1st 2011 to December 31th 2022, provided by the Italian National Institute of Statistics (ISTAT). The dataset contains information on deaths for all possible causes, without specific reference to any particular one. The data exhibit strikingly evident periodicity in the number of deaths with pronounced maxima in the winter and minima in the summer, repeating itself in amplitude along the whole twelve-year sample. Superimposed on this wave-like structure are often present excesses of casualties, most likely due to occasional causes of death such as the flu epidemics (in winter) and heat waves (in summer). The very accurate periodicity along the seasons (the "baseline"), allows us to determine with great accuracy and confidence the number of expected deaths for each day of the year in the absence of occasional contributions. Each of the latter can be modeled with an additional function that parameterizes the deviation from the baseline.

Increasing situational awareness through nowcasting of the reproduction number (preprint)

The time varying reproduction number R is a critical variable for situational awareness during infectious disease outbreaks, but delays between infection and reporting hinder its accurate estimation in real time. We propose a nowcasting method for improving the timeliness and accuracy of R estimates, based on comparisons of successive versions of surveillance databases. The method was validated against COVID-19 surveillance data collected in Italy over an 18-month period. Compared to traditional methods, the nowcasted reproduction number reduced the estimation delay from 13 to 8 days, while maintaining a better accuracy. Moreover, it allowed anticipating the detection of periods of epidemic growth by between 6 and 23 days. The method offers a simple and generally applicable tool to improve situational awareness during an epidemic outbreak, allowing for informed public health response planning.

Effect of an enhanced public health contact tracing intervention on the secondary transmission of SARS-CoV-2 in educational settings: the four-way decomposition analysis (preprint)

Background: We assessed the impact of testing contacts immediately instead of at the end of quarantine on the transmission of SARS-CoV-2 in schools in Reggio Emilia Province. Methods: We analysed surveillance data on notification of COVID-19 cases in schools between 1 September 2020 and 4 April 2021. Results: Median tracing delay decreased from 7 to 3.1 days and the percentage of the known infection source increased from 34% to 54.8% (IRR 1.61 1.40-1.86). Implementation of prompt contact tracing was associated with a 10% decrease in the number of secondary cases (excess relative risk, EER -0.1 95%CI -0.35 to 0.15). Knowing the source of infection of the index case led to a decrease in secondary transmission (IRR 0.75 95% CI 0.63-0.91) while the decrease in tracing delay was associated with decreased risk of secondary cases (1/IRR 0.97 95%CI 0.94-1.01 per one day of delay). The direct effect of the intervention accounted for the 29% decrease in the number of secondary cases (EER -0.29 95% -0.61 to 0.03). Conclusions: Prompt contact testing in the community seems to reduce the time of contact tracing and increases the ability to identify the source of infection in school outbreaks. Yet, observed differences can be also due to differences in the force of infection and to other control measures put in place.Funding: This project was carried out with the technical and financial support of the Italian46 Ministry of Health - CCM 2020 and Ricerca Corrente Annual Program 2023

Heat wave, COVID-19, and mortality excess in the 2022 summer: a cohort study on data from Italian surveillances. (preprint)

We aimed to assess differences in the summer excess of mortality by COVID-19 history using data from the mortality and COVID-19 surveillances. We found 4% excess risk in 2022 summer, compared to 2015-2019. A mortality rate ratio of 1.59 (95%CI 1.39-1.82) for COVID-19 survivors compared to naive, was found. Both were higher in people aged >74 years. During the July heat wave, the excess for COVID-19 survivors decreased and disappeared when excluding people living in nursing homes.

The decline of COVID-19 severity and lethality over two years of pandemic (preprint)

Undernotification of SARS-CoV-2 infections has been a major obstacle to the tracking of critical quantities such as infection attack rates and the probability of severe and lethal outcomes. We use a model of SARS-CoV-2 transmission and vaccination informed by epidemiological and genomic surveillance data to estimate the number of daily infections occurred in Italy in the first two years of pandemic. We estimate that the attack rate of ancestral lineages, Alpha, and Delta were in a similar range (10-17%, range of 95% CI: 7-23%), while that of Omicron until February 20, 2022, was remarkably higher (51%, 95%CI: 33-70%). The combined effect of vaccination, immunity from natural infection, change in variant features, and improved patient management massively reduced the probabilities of hospitalization, admission to intensive care, and death given infection, with 20 to 40-fold reductions during the period of dominance of Omicron compared to the initial acute phase.

The decline of COVID-19 severity and lethality over two years of pandemic (preprint)

Undernotification of SARS-CoV-2 infections has been a major obstacle to the tracking of critical quantities such as infection attack rates and the probability of severe and lethal outcomes. We use a model of SARS-CoV-2 transmission and vaccination informed by epidemiological and genomic surveillance data to estimate the number of daily infections occurred in Italy in the first two years of pandemic. We estimate that the attack rate of ancestral lineages, Alpha, and Delta were in a similar range (10–17%, range of 95% CI: 7–23%), while that of Omicron until February 20, 2022, was remarkably higher (51%, 95%CI: 33–70%). The combined effect of vaccination, immunity from natural infection, change in variant features, and improved patient management massively reduced the probabilities of hospitalization, admission to intensive care, and death given infection, with 20 to 40-fold reductions during the period of dominance of Omicron compared to the initial acute phase.

Estimation of the incubation period and generation time of SARS-CoV-2 Alpha and Delta variants from contact tracing data (preprint)

Background. During 2021, the COVID-19 pandemic was characterized by the emergence of lineages with increased fitness. For most of these variants, quantitative information is scarce on epidemiological quantities such as the incubation period and generation time, which are critical for both public health decisions and scientific research. Method. We analyzed a dataset collected during contact tracing activities in the province of Reggio Emilia, Italy, throughout 2021. We determined the distributions of the incubation period using information on negative PCR tests and the date of last exposure from 282 symptomatic cases. We estimated the distributions of the intrinsic generation time (the time between the infection dates of an infector and its secondary cases under a fully susceptible population) using a Bayesian inference approach applied to 4,435 SARS-CoV-2 cases clustered in 1,430 households where at least one secondary case was recorded. Results. We estimated a mean incubation period of 4.9 days (95% credible intervals, CrI, 4.4-5.4; 95 percentile of the mean distribution: 1-12) for Alpha and 4.5 days (95%CrI 4.0-5.0; 95 percentile: 1-10) for Delta. The intrinsic generation time was estimated to have a mean of 6.0 days (95% CrI 5.6-6.4; 95 percentile: 1-15) for Alpha and of 6.6 days (95%CrI 6.0-7.3; 95 percentile: 1-18) for Delta. The household serial interval was 2.6 days (95%CrI 2.4-2.7) for Alpha and 2.4 days (95%CrI 2.2-2.6) for Delta, and the estimated proportion of pre-symptomatic transmission was 54-55% for both variants. Conclusions. These results indicate limited differences in the incubation period and intrinsic generation time of SARS-CoV-2 variants Alpha and Delta compared to ancestral lineages.

Tracking the progressive spread of the SARS-CoV-2 Omicron variant in Italy, December 2021 - January 2022 (preprint)

The SARS-CoV-2 variant of concern Omicron was first detected in Italy in November 2021. Data from three genomic surveys conducted in Italy between December 2021 and January 2022 suggest that Omicron became dominant in less than one month (prevalence on January 3: 78.6%-83.8%) with a doubling time of 2.7-3.1 days. The mean net reproduction number rose from about 1.15 in absence of Omicron to a peak of 1.83 for symptomatic cases and 1.33 for hospitalized cases, while it remained stable for critical cases.

Secondary transmission of SARS-CoV-2 in educational settings in Northern Italy from September 2020 to April 2021: a population-based study (preprint)

BackgroundWe aimed to quantify the risk of transmission of SARS-CoV-2 in the school setting by type of school, characteristics of the index case and calendar period in the Reggio Emilia province (RE), Italy. The secondary aim was to estimate the promptness of contact tracing. MethodsA population-based analysis of surveillance data of all COVID-19 cases occurring in RE, Italy, from September 1, 2020, to April 4th, 2021, for which a school contact and/or exposure was suspected. Indicator of the delay in contact tracing was computed as the time elapsed since positivity of the index case and the date on which the swab for classmates was scheduled (or most were scheduled). ResultsOverall, 30,184 and 13,608 contacts among classmates and teachers/staff, respectively, were identified and received recommendation for testing; 43,214 (98.7%) performed the test. Secondary transmission occurred in about 40% of the investigated classes, and the overall secondary case attack rate was 4%, slightly higher when the index case was a teacher, but with almost no differences by type of school and stable during the study period. Promptness of contact tracing increased during the study period, reducing the time from index case identification and testing of contacts from 7 to 3 days, as well the ability to identify possible source of infection in the index case. ConclusionsDespite the spread of the Alpha variant during the study period in RE, the secondary case attack rate remained stable from school reopening in September 2020 until the beginning of April 2021.

Co-circulation of SARS-CoV-2 variants B.1.1.7 and P.1 (preprint)

SARS-CoV-2 variants of concern (B.1.1.7, P.1 and B.1.351) have emerged in different continents of the world. To date, little information is available on their ecological interactions. Based on two genomic surveillance surveys conducted on February 18 and March 18, 2021 across the whole Italian territory and covering over 3,000 clinical samples, we found significant co-circulation of B.1.1.7 and P.1. We showed that B.1.1.7 was already dominant on February 18 in a majority of regions/autonomous provinces (national prevalence 54%) and almost completely replaced historical lineages by March 18 (dominant in all regions/autonomous provinces, national prevalence 86%). At the same time, we found a substantial proportion of cases of the P.1 lineage on February 18, almost exclusively in Central Italy (with an overall prevalence in the macro-area of 18%), which remained at similar values on March 18, suggesting the inability by this lineage to outcompete B.1.1.7. Only 9 cases from variant B.1.351 were identified in the two surveys. At the national level, we estimated a mean relative transmissibility of B.1.1.7 (compared to historical lineages) ranging between 1.55 and 1.57 (with confidence intervals between 1.45 and 1.66). The relative transmissibility of P.1 estimated at the national level varied according to the assumed degree of cross-protection granted by infection with other lineages and ranged from 1.12 (95%CI 1.03-1.23) in the case of complete immune evasion by P.1 to 1.39 (95%CI 1.26-1.56) in the case of complete cross-protection. These observations may have important consequences on the assessment of future pandemic scenarios.

Return to normal: COVID-19 vaccination under mitigation measures (preprint)

Being unable to suppress SARS-CoV-2 transmission, the majority of countries worldwide have resorted to a mitigation approach towards COVID-19, allowing some degree of viral circulation in the population. Here, we investigate the expected outcomes of the interplay between vaccination rollout and adaptive mitigation measures constantly altering the epidemic trajectory and keeping the reproduction number around the unit. Using a novel mathematical modeling framework, we estimate that, for vaccination capacities of at least 4 daily doses administered per 1,000 inhabitants, a complete release of mitigation measures can be expected within 7 to 13 months since the start of vaccination, with a two-year cumulative incidence of deaths between 0.18 and 0.46 per 1,000 population. A heavier burden of deaths and a delayed <> is expected for lower vaccine capacities, if viral transmissibility exceeds by >60% the one estimated at the beginning of the pandemic, or if vaccine protection is short-lived. Failure to prioritize the elderly or a premature release of mitigation measures after vaccination of the most fragile will conspicuously increase the expected mortality. Finally, strategies oriented to prioritize the suppression of SARS-CoV-2 by maintaining strict restrictions will take a similar time as a mitigation approach, possibly resulting in acceptability issues. Persisting unknowns about the evolving epidemiology of SARS-CoV-2 variants and on the effectiveness of available and upcoming vaccines may warrant a future reassessment of these conclusions.

Socio-Economic Patterns of COVID-19 During the Pandemic in Italy (preprint)

Objective: To investigate the association between deprivation and COVID-19 outcomes in Italy during pre-lockdown, lockdown and post-lockdown periods.Design: Retrospective cohort study.Setting: All municipalities in Italy with less than 50,000 population.Participants: 38,534,169 citizens and 222,875 COVID-19 cases reported to the Italian epidemiological surveillance were assigned to quintiles based on the deprivation index of their municipality of residence.Interventions: The COVID-19 pandemic during pre-lockdown, lockdown and post-lockdown from the 20th of February to the 15th of October of 2020.Main outcome measures: Multilevel negative binomial regression models, adjusting for age, sex, population-density and region of residence were conducted to evaluate the association between deprivation and COVID-19 incidence, case-hospitalisation rate and case-fatality. The association measure was the rate ratio.Results: During pre-lockdown, lockdown and post-lockdown, the incidence rate ratios (IRR) with 95% confidence interval (CI) in the most deprived quintile with respect to the least deprived quintile were 1.17 (95% CI 0.98 to 1.41), 1.14 (1.03 to 1.27) and 1.47 (1.32 to 1.63), respectively. In those three periods, the case-hospitalization IRR were 0.68 (0.51 to 0.92), 0.89 (0.72 to 1.11) and 0.99 (0.81 to 1.22) and the case-fatality IRR were 0.92 (0.75 to 1.13), 0.95 (0.85 to 1.07) and 1.02 (0.73 to 1.41), respectively.Conclusions: During lockdown and post-lockdown, but not during pre-lockdown, a higher incidence of cases was observed in the most deprived municipalities compared with the least deprived ones. No differences in case-hospitalisation and case-fatality according to deprivation were observed in any period under study.

Effectiveness of regional restrictions in reducing SARS-CoV-2 transmission during the second wave of COVID-19, Italy. (preprint)

To counter the second COVID-19 wave, the Italian government has adopted a scheme of three sets of restrictions (coded as yellow, orange, and red) imposed on a regional basis. We estimate that milder restrictions in regions at lower risk (yellow) resulted in a transmissibility reduction of about 18%, leading to a reproduction number Rt of about 0.99. Stricter measures (orange and red) led to reductions of 34% and 45% and Rt values of about 0.89 and 0.77 respectively.

Secondary transmission of COVID-19 in preschool and school settings after their reopening in northern Italy: a population-based study (preprint)

BackgroundSchool closures was one of the main measures undertaken to reduce the number of social contacts during the first wave of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic. We aimed to describe the data on secondary transmission of SARS-CoV-2 among students and teachers/personnel after the reopening of preschools and schools in Reggio Emilia, Italy. MethodsThis prospective population-based study included all consecutive cases leading to an investigation in 41 classes of 36 educational institutions (8 infant-toddler centres and preschools, 10 primary and 18 secondary schools) in the period September 1 - October 15, 2020, in Reggio Emilia province, Italy. We report the characteristics of the school, of the index case, including the possible source of infection, the number of contacts (students and teachers/personnel) that were identified and tested and the characteristics of secondary cases. ResultsIn the study period, 994 students and 204 teachers were tested during related investigations due to notification of 43 primary cases (38 among students and 5 among teachers). Of these, 10 students and two teachers created 39 secondary cases, resulting in an attack rate of 3.9%. There were no secondary cases among teachers/stuff. Secondary transmission occurred in one primary school and 8 secondary schools. Except for two students and one teacher, the possible source of infection for all index cases was identified as they had all had previous contact with a positive case; the majority of secondary cases did not report any previous close contact with a positive case. The clusters ranged from one to 22 secondary cases. ConclusionsTransmission at school occurred in a non-negligible number of cases, particularly in secondary schools. Prompt testing and isolation of classmates could probably reduce the risk of transmission in school settings.

Epidemiological characteristics of COVID-19 cases in non-Italian nationals notified to the Italian surveillance system (preprint)

BackgroundInternational literature suggests that disadvantaged groups are at higher risk of morbidity and mortality from SARS-CoV-2 infection due to poorer living/working conditions and barriers to healthcare access. Yet, to date, there is no evidence of this disproportionate impact on non-national individuals, including economic migrants, short-term travellers, and refugees. MethodsWe analysed data from the Italian surveillance system of all COVID-19 laboratory-confirmed cases tested positive from the beginning of the outbreak (20th of February) to the 19th of July 2020. We used multilevel negative-binomial regression models to compare the case-fatality rate and the rate of admission to hospital and intensive care unit (ICU) between Italian and non-Italian nationals. The analysis was adjusted for differences in demographic characteristics, pre-existing comorbidities, and period of diagnosis. ResultsWe analysed 213,180 COVID-19 cases, including 15,974 (7.5%) non-Italian nationals. We found that, compared to Italian cases, non-Italian cases were diagnosed at a later date and were more likely to be hospitalised [(adjusted relative risk (ARR)=1.39, 95% confidence interval (CI): 1.33-1.44)] and admitted to ICU (ARR=1.19, 95% CI: 1.07-1.32), with differences being more pronounced in those coming from countries with lower HDI. We also observed an increased risk of death in non-Italian cases from low-HDI countries (ARR=1.32, 95% CI: 1.01-1.75). ConclusionsA delayed diagnosis in non-Italian cases could explain their worse outcomes compared to Italian cases. Ensuring early access to diagnosis and treatment to non-Italians could facilitate the control of SARS-CoV-2 transmission and improve health outcomes in all people living in Italy, regardless of nationality.

The relationship between human mobility and viral transmissibility during the COVID-19 epidemics in Italy (preprint)

In 2020, countries affected by the COVID-19 pandemic implemented various non-pharmaceutical interventions to contrast the spread of the virus and its impact on their healthcare systems and economies. Using Italian data at different geographic scales, we investigate the relationship between human mobility, which subsumes many facets of the population's response to the changing situation, and the spread of COVID-19. Leveraging mobile phone data from February through September 2020, we find a striking relationship between the decrease in mobility flows and the net reproduction number. We find that the time needed to switch off mobility and bring the net reproduction number below the critical threshold of 1 is about one week. Moreover, we observe a strong relationship between the number of days spent above such threshold before the lockdown-induced drop in mobility flows and the total number of infections per 100k inhabitants. Estimating the statistical effect of mobility flows on the net reproduction number over time, we document a 2-week lag positive association, strong in March and April, and weaker but still significant in June. Our study demonstrates the value of big mobility data to monitor the epidemic and inform control interventions during its unfolding.

The impact of a nation-wide lockdown on COVID-19 transmissibility in Italy (preprint)

On March 10, 2020, Italy imposed a national lockdown to curtail the spread of COVID-19. Here we estimate that, fourteen days after the implementation of the strategy, the net reproduction number has dropped below the epidemic threshold - estimated range 0.4-0.7. Our findings provide a timeline of the effectiveness of the implemented lockdown, which is relevant for a large number of countries that followed Italy in enforcing similar measures.

Epidemiological characteristics of COVID-19 cases in Italy and estimates of the reproductive numbers one month into the epidemic (preprint)

Background In February 2020, a locally-acquired COVID-19 case was detected in Lombardia, Italy. This was the first signal of ongoing transmission of SARS-CoV-2 in the country. The outbreak rapidly escalated to a national level epidemic, amid the WHO declaration of a pandemic. Methods We analysed data from the national case-based integrated surveillance system of all RT-PCR confirmed COVID-19 infections as of March 24th 2020, collected from all Italian regions and autonomous provinces. Here we provide a descriptive epidemiological summary on the first 62,843 COVID-19 cases in Italy as well as estimates of the basic and net reproductive numbers by region. Findings Of the 62,843 cases of COVID-19 analysed, 71.6% were reported from three Regions (Lombardia, Veneto and Emilia-Romagna). All cases reported after February 20th were locally acquired. Estimates of R0 varied between 2.5 (95%CI: 2.18-2.83) in Toscana and 3 (95%CI: 2.68-3.33) in Lazio, with epidemic doubling time of 3.2 days (95%CI: 2.3-5.2) and 2.9 days (95%CI: 2.2-4.3), respectively. The net reproduction number showed a decreasing trend starting around February 20-25, 2020 in northern regions. Notably, 5,760 cases were reported among health care workers. Of the 5,541 reported COVID-19 associated deaths, 49% occurred in people aged 80 years or above with an overall crude CFR of 8.8%. Male sex and age were independent risk factors for COVID-19 death. Interpretation The COVID-19 infection in Italy emerged with a clustering onset similar to the one described in Wuhan, China and likewise showed worse outcomes in older males with comorbidities. Initial R0 at 2.96 in Lombardia, explains the high case-load and rapid geographical spread observed. Overall Rt in Italian regions is currently decreasing albeit with large diversities across the country, supporting the importance of combined non-pharmacological control measures.