To the Editor: We were pleased to read the interesting article by Anne et al. published in IJP [1]. We highlight a few points about the manuscript and seek clarification.

Sick newborn scores like SNAP II and SNAPPE II (Score for Neonatal Acute Physiology with Perinatal Extension‑II) are designed to be used within 24 h of admission. Yet here, authors have used this score after the babies have developed septicemia [2]. Thus, the scores obtained may not reflect the mortality risk as they do not consider the time elapsed since admission, treatment received, or pre-existing systemic pathologies.

Exclusion criteria don’t mention whether the babies who died within 72 h of enrolment are excluded. It is written in mixed model analysis that the outcome at day 14 is taken as an independent variable; it seems to be a dependent variable.

In Table 1: Gestational age and birth weight seem to be expressed as mean (standard deviation), but the header mentions only median and frequency, confusing readers. P values calculated by the Pearson chi-square test are not correlating with the values calculated by us using the same data. In Table 2, data are expressed in mean (SD) and frequency, but as per our calculations, the values seem to be of mean (confidence interval).

Paragraph two under discussion is the repetition of information from the introduction. The reasons for the failure of serial measurements in the SNAP II score to predict mortality better are not explained. As per the findings from a similar study by Meadow et al. [3], reason for this failure was explained as an increase in chances of survival with every passing day of NICU stay. It would have been more impactful if similar reasons were discussed.

We hope the discussion on the points raised will improve our understanding.