More than 5 months after the World Health Organization (WHO) declared the coronavirus infectious disease 2019 (COVID-19) outbreak a pandemic, and almost 900,000 deaths around the world, we are still far from fully understanding COVID-19 epidemiology and natural history. As we fear the consequences of second epidemic waves, as stated in a recent commentary (Middleton et al. 2020), it is not clear why the clinical burden of COVID-19 has decreased (Johns Hopkins University 2020). Experts claim this to be possibly related to infection transmission in younger age groups, as compared to the past (current average age of COVID-19 cases in Italy is 31 years versus 70 years in the first 2 months of the outbreak) (Italian National Institute of Health 2020), to decreased disease severity or to better therapeutic approaches.

National-level COVID-19 surveillance system data from Italy, the first European country to be hit by the COVID-19 outbreak and one of the countries with the highest number of reported cases and deaths in Europe (together with France, Spain and the UK) (Odone et al. 2020), are to be explored to better understand age-specific COVID-19 case-fatality rate over time.

Table 1 reports the number of confirmed COVID-19 cases and deaths at, respectively, 2, 4 and 6 months since the onset of the epidemic, by age group. The large differences in case fatality rates by age (ranging from less than 3% in people younger than 60 years, up to more 30% in people 80 years or older) are well known and associated, among other factors, with higher rates of chronic comorbidities in elder populations. It also explains—as we have previously demonstrated—the high number of COVID-19 deaths in Italy, the country with one of the highest average age in the world (Signorelli et al. 2020).

Table 1 COVID-19 cumulative deaths and case-fatality rates (CFRs) per age group at three different time points (Italy 2020)

More interestingly, as it emerges from Table 1, age-specific COVID-19 case-fatality rates seem not to vary over time, supporting the hypothesis that the less severe COVID-19 clinical outcomes (i.e. decreasing overall death rates) reported in recent times in Italy and Europe might be due to infection community transmission in younger populations.

We acknowledge the limitations linked to using surveillance data to derive insight on the COVID-19 epidemiology, linked to under-notification, heterogeneous testing strategies and time lag between notification and death (average time 12 days; Italian National Institute of Health 2020). Sill, we believe this data provide solid evidence to support the argument that should infection start to circulate again among older adults and elder populations (including those living in nursing homes), this might have detrimental clinical impacts in the weeks to come. If we add that no conclusive evidence has emerged so far on other potential factors influencing COVID-19 epidemiology, including the clinical effect of viral mutations or the impact of environmental temperature and weather conditions, it is of crucial importance to keep effective containment and mitigation measures in place and protect vulnerable subgroups of the population.