A 2020 article from JAMA2 reports the results from an observational study in Italy that investigates the association between mortality and potential risk factors for patients who were admitted to a hospital intensive care unit (ICU) for Covid-19 related morbidities. As per the researchers, "This observational study enrolled all consecutive patients with confirmed SARS-CoV-2 infection admitted to one of the Network ICUs from February 20 to April 22, 2020.The ICU and hospital outcomes of each patient were recorded on May 30, 2020. Cumulative follow-up time was 164 305 patientdays from ICU admission to end of follow-up for the 3988 patients (median follow-up time, 70 [range, 38-112] days; IQR, 61-70 days). " The outcome (event) of interest for this study is death. Patients admitted to the ICU were followed up until May 30, 2020. The follow-up time for each patient is the time from ICU admission to death for those who died, and the time from ICU admission to May 30, 2020 for those still alive as of May 30 2020. The following Kaplan-Meier curves show the estimated percentage surviving over time (days) separately by age at the time of ICU admission (< 64 years and >= 64 years), and for the entire sample (both age groups combined). The vertical dashes on the survival curves represent censoring times. (this is not always done on Kaplan-Meier curves, and does not change the interpretation) 2 Gresseli G, et al. Risk Factors Associated With Mortality Among Patients With COVID-19 in Intensive Care Units in Lombardy, Italy JAMA Intern Med. Published online July 15, 2020. doi:10.1001/jamainternmed.2020.3539 4 6. What is the approximate 20th percentile of the time- to-death distribution for patients >= 64 years old upon admission to the ICU? a. 25 days b. 6 days c. This cannot be estimated as the Kaplan-Meier curve does not reach 20% d. Percentiles cannot be estimated from a Kaplan-Meier curve