Publication:

Interpreting Kaplan Meier’s survival curve in COVID-19 patients: a systematic review

Covid 19 is the causative agent of coronavirus disease in 2019-2020; was declared a pandemic by the World Health Organization on 11th March 2020. The article aims to interpret the functionality of Kaplan Meier’s survival curve used in various study designs. The search strategy and selection of published works in the MEDLINE database from the National Library of Medicine (NLM) was used to identify original journal articles published in English from March 1, 2020 to September 11, 2020. The search strings in title/abstract were ‘COVID-19’ or ‘coronavirus’ or and ‘Kaplan Meier curve’ that yielded 225 articles. Finally only 28 articles were reviewed. These were clinical trials (N=1), and retrospective (N=10) and prospective (N=1) observational studies (N=2), case control study (N=1) retrospective cohort study (N=6), medical record based study (N=3), observational studies (N=2), not specified (N=1), diagnostic test (N=1). 88% of articles were from China. Kaplan Meier’s curve was depicted in all the studies and 17 were found to be highly significant for KM curve, 2 were found significant as 6 were found to be non significant and in 3 no mention of significance of p value was given. It was concluded that it is the first systematic review to date related to interpreting Kaplan Meier’s survival curve in patients with COVID-19. The  graph has been used singularly, in multiples and cumulatively, thus giving visual assessment which enhances the interpretation of results for survival analysis, failure to time and time to event plots. It is not only used in RCT study designs but a range of observational and retrospective studies.