• Authors used semi-Delphi technique to arrive at pre-defined objectives. What were the intended objectives/end points of Delphi method?

• What was the method of raters’ orientation?

• Authors assessed appropriateness of hospitalization of 274 patients during a period from July-September 2015 from five medical colleges in Delhi and surrounding states. This comes out as 0.6 case/day/medical college. Considering that all these are public hospitals, number of hospitalizations would have been much higher. What was the sampling method to select only 274 cases?

• Sampling should also have taken account of season when minimum/least number of admission season for better validity and more accurate representative of actual population. There are more chances of appropriateness of admission criteria in peak season than in trough season of admissions. Bianco, et al. [2] reported that inappropriate admission was significantly higher if it occurred during the daytime. Considering the circumstances in such hospitals (less chances of follow-up, biased health seeking behavior for male children, distance from home to hospital, availability of night conveyance, ignorance, loss of wages for attending health facility), admission at night time tend to be more inappropriate as compared to day time. Were these factors taken into consideration?

• Sample consisted of 274 patients from Northern India (Delhi-NCR) and 622 from Southern India (Coimbatore and Kollam). The population from Sothern India consisted of mostly infants which were part of another study, and taken from 146 hospitals. So overall, almost two-thirds of sample population were infants making the tool less valid for older age groups. Besides that, not all states of India have same Doctor and bed availability [3]. Thus, samples should have been more representative of the country to have more valid applicability of results.