We read the article by Parihar and Passi(1) with interest. It was a very good attempt to study an important aspect of clinical practice. However, there are some problems with the article. First, the authors report a medical error rate of 35.5%, which is very high. The authors have this figure by dividing the number of medical errors by total number of patients under evaluation. However, it would have been better, if authors used total number of prescriptions as the denominator to report the rate of medical errors. Further, it is possible that in many cases multiple errors would have occurred in a single patient. Hence, the figure of 35.5% appears to be an erroneous conclusion. Second, on the same issue, if we take to understand that total numbers of prescriptions in the intensive care units are more compared to other service areas, the figure of number of errors would come down significantly. The authors also report the percentage of errors between 8 am to 8 pm and from 8 pm to 8 am. Again these figures can also be significantly be affected by total number of prescription under evaluation, rather than the time frame alone. We know that in our practice, in most cases the clinical rounds are mostly held during the day time and possibly more number of prescriptions are written during the day time. Third, the authors have not mentioned anything about the blinding of clinicians and other staff who were under scanner. Because it is quite possible that if the clinicians and staff were aware about the study, then the reported rates would be less than the actual rates in clinical practice, because of being conscious about avoiding mistakes. Fourth, the alarming aspect of the study was the fact that more number of errors were committed by Senior residents than Junior Residents. If this fact is true (after taking total number of prescriptions as denominator for the each group), than it is very important to look at the training aspect. Fifth, the authors have not mentioned about the time of assessment, because it would give us important lead as to whether the erroneous prescriptions were intercepted or not. Sixth, it has been frequently shown in Western studies(2) that the numbers of errors correlate with level of stress and psychiatric morbidity in the residents. The study could have been strengthened by including such measures.