We read the article by Costantino et al. (1) that was conceived to assess the usefulness of hospitalization for older patients affected by COVID-19. The study examined 1,902 patients consecutively admitted to three large hospitals in Milan (Italy) at the time of the first wave of the pandemic. In the frame of an observational design, data were obtained from different sources (the emergency department [ED] database of the three participating hospitals and the population mortality data for Milan from January 1, 2015 to April 30, 2020 as released by the Italian National Institute of Statistics-ISTAT). The authors hypothesized that the mortality rates in older adults with COVID-19 were higher during the peak phases of the pandemic than during the off-peak phases when more access to hospitals was available. A peak phase was defined as a 7-day period with the highest mean number of patients with COVID-19 presenting at the three ED. In contrast, the off-peak phase was defined as the 14-day period starting after the 7-day peak, this length of the latter period being chosen to obtain a similar number of enrolled patients in both phases.
After analyzing the number of daily deaths across five years and according to age groups, a descriptive comparative analysis of the number of daily deaths in patients aged 80 years or older versus those younger than 80 was carried out. To compare mortality between age groups and peak phases, a standardized mortality ratio (SMR) was calculated by dividing the number of observed deaths by the mean number of deaths observed from 2015 to 2019. The SMRs for the peak and off-peak phases were also separately calculated for the two age groups. Finally, mortality during peak and off-peak phases was compared by calculating the ratio between the corresponding SMRs. The results of these analyses were that while the SMR for younger patients was lower during the off-peak phase than during the peak (1.40, 95% CI: 1.25–1.58 versus 1.98, 95% CI: 1.72–2.29, respectively), the SMR for older patients was similar in the two phases (2.48, 95% CI: 2.32–2.65 versus 2.28, 95% CI: 2.07–2.52, respectively).
With the main results stemming from this complex study design and analysis, we would like to comment on the interpretations provided by the authors in the Discussion. They state that mortality did not differ between the two periods in very old persons, suggesting that mortality of older persons with COVID-19 is related to the disease itself and not to lack of appropriate care. Further, the authors state that older patients are usually not offered intensive care nor non-invasive ventilation because these are not helpful. The authors also note that “…in a setting of limited resources and increased demand, hospitalization of elderly patients does not appear to affect their prognosis. Rather, hospitalization was found to worsen their quality of life and increase the overcrowding of health care facilities. Therefore, our data indicate that the focus for COVID-19-positive elderly patients should be providing healthcare, and possibly palliative care, outside of a hospital setting (i.e., at home)…”.
We believe that the reported results fail to support this kind of message and the authors’ conclusions. In particular, there is a lack of information on the disease severity, the patients’ clinical characteristics, and the provided treatments (before and during hospitalization) in both samples and the two periods of interest (i.e., peak and off-peak). Had older adults received the same opportunity to be hospitalized and/or be candidates for intensive care units as the younger ones during the first wave of the COVID-19 pandemic? Our view is that older patients were admitted to hospitals only when the disease was at a very advanced stage during both the peak and the off-peak phase. Moreover, when the first wave of COVID-19 pandemic hit the Lombardy region, the healthcare system was completely unprepared and disorganized, particularly in the context of general practitioners, who were not adequately supported in the management of the escalating emergency. Accordingly, in cases characterized by severe manifestations of COVID-19, the only opportunity of care was admission to ED and hospitals (2). At the peak of the pandemic, the hospitals were largely unable to accept all the severe patients, and older people were often those ending up not being hospitalized.
An Italian Registry of the Italian Society of Medicine (SIMI) that involved internal medicine wards in 41 large tertiary referral hospitals (3) and collected 3,044 hospitalized patients with COVID-19 (mean age 61+15 years) has recently shown that not only the patients’ age was associated to mortality, but also symptoms at admission and comorbidities (in particular, chronic heart failure and COPD). It is also noteworthy that when a cohort of older adults is evaluated, age is not the only risk determinant, because there is the need to consider physical and cognitive status, frailty, and exposure to polypharmacy. Indeed, age per se may be meaningless in front of the extreme heterogeneity of the older persons’ health status. Unfortunately, none of these factors seem to be considered nor reported by Costantino et al. (1).
All in all, these conclusions about the complex and new scenario of the COVID-19 pandemic are, in our opinion, not adequately supported by data nor the methodology used in this study. We strongly disagree with the statement made in the Discussion: “…While the data examined apply to the current COVID-19 epidemic, it is possible that the observation that hospitalization does not positively affect elderly patients’ prognosis may be true for other diseases, especially where effective therapy is lacking. The Choosing Wisely campaign has recently highlighted the need to reconsider the benefit(s) of hospital admission...”. The birth date cannot be considered an eligibility criterion for the allocation of care services. This kind of message may be misleadingly interpreted and potentially pave the way towards ageistic attitudes. It is surely necessary to have the current hospital-centered system evolve towards alternative care services for frail older persons, but such alternatives cannot be proposed without evidence stemming from randomized controlled studies (e.g., comparing a well-defined and standardized intervention and involving a clearly characterized population).
In our opinion, this article is at risk of reporting asymmetric results. It does not seem to adequately consider the complexity of addressing the clinical needs of frail older persons outside of the hospital setting. These difficulties translate into limited access to care, inadequate capacity to address basic necessities, and increased risk of health-related adverse events. For this reason, we recommend a cautious interpretation of the findings. We also hope that the article will not be used to promote ageistic attitudes in the clinical setting and the society. The modernization of the healthcare system as a whole will only happen through the avoidance of obsolete paradigms (such as the use of chronological age for allocation of services) and the promotion of personalization of care (based on the individual’s biology and functions).


Alessandro Nobili, Alessandra Marengoni, Matteo Cesari, Giuseppe Bellelli, Fabiano di Marco, Andrea Gori and Pier Mannuccio Mannucci

References
1. Costantino G, Solbiati M, Elli S, Paganuzzi M, Massabò D, Montano N, Mancarella M, Cortellaro F, Cataudella E, Bellone A, Capsoni N, Bertolini G, Nattino G, Casazza G. Utility of hospitalization for elderly individuals affected by COVID-19. PLoS One. 2021 Apr 26;16(4):e0250730. doi: 10.1371/journal.pone.0250730. PMID: 33901228; PMCID: PMC8075227.
2. Garattini L, Zanetti M, Freemantle N. The Italian NHS: What Lessons to Draw from COVID-19? Appl Health Econ Health Policy. 2020 Aug;18(4):463-466. doi: 10.1007/s40258-020-00594-5. PMID: 32451979; PMCID: PMC7247917.
3. Corradini E, Ventura P, Ageno W, Cogliati CB, Muiesan ML, Girelli D, Pirisi M, Gasbarrini A, Angeli P, Querini PR, Bosi E, Tresoldi M, Vettor R, Cattaneo M, Piscaglia F, Brucato AL, Perlini S, Martelletti P, Pontremoli R, Porta M, Minuz P, Olivieri O, Sesti G, Biolo G, Rizzoni D, Serviddio G, Cipollone F, Grassi D, Manfredini R, Moreo GL, Pietrangelo A; SIMI-COVID-19 Collaborators. Clinical factors associated with death in 3044 COVID-19 patients managed in internal medicine wards in Italy: results from the SIMI-COVID-19 study of the Italian Society of Internal Medicine (SIMI). Intern Emerg Med. 2021 Apr 24:1–11. doi: 10.1007/s11739-021-02742-8. Epub ahead of print. PMID: 33893976; PMCID: PMC8065333.