1. We have utilized the latest scoring system that Glascoe(2) delineates in her manual in our study. The reason why we identify only 4 and not 5 groups of parents is because the PEDS was administered to the parents by research worker and was not self-administered as reported by Glascoe in several of her studies. Therefore, communication barriers due to not understanding of the questions were not found in our sample. Moreover, only those children who were accompanied by their parent/s were administered the PEDS and included in the study.

2. Unlike USA where a wide variety of screening and diagnostic tests are available for use with children of all ages, there are only limited number of tests covering a limited age range available in India(4) and one has to very often use tests developed in the West. We have used the DP II extensively(5) after some modi-fications to make it relevant for use with Indian children and find it more useful than the Indian adaptation of the Vineland Social Maturity Scale (VSMS)(6) which is one of the most commonly used screening test for young children in India.

3. In the present study, we have used the DP II(3) as the gold standard as it is considered the best-standardized develop-mental screening test for use in the pediatric clinic(7). The DP II has five sub scales, namely Physical, Self-Help, Social Academic and Communication. The test can be administered in several different ways and also interpreted in at least two different ways(3). The use of different methods of administration and inter-pretation may influence the test results. The test can be administered to parents in an interview format or interview format may be combined with direct adminis-tration to the child of some of the test items, when the validity of the parental report is of concern. It is important to recognize that although the correlations between the two methods are high, literature indicates that parents may report higher functioning for their child leading to under detection of developmental problems. Glascoe in her studies has used the parental interview format for administration, whereas we have used the second method of administration.

Secondly, there are two methods of interpreting the DP II and this may also vastly influence which child is labelled normal or delayed. In the DP II, each scale produces a developmental age that is subtracted from the child’s chronological age. The resulting "months differential" is compared with a cut-off that indicates whether the child is advanced, normal, borderline or delayed in their develop-ment. Glascoe(8,9) in her studies uses this method, wherein she compares the developmental age of the child on the Academic sub scale to designate the child as delayed. Using this method she reports only 4.3% of the children tested as delayed(8). In the second method, the academic age of the child is converted to an IQ equivalent score (IQE) i.e., the ratio of the academic age to chronological age and the product multiplied by 100. Scores less than 70 are interpreted as having failed screening. The academic sub scale of DP II assesses a range of skills necessary for success in school including language, cognition and school achieve-ment. The authors of DP II have reported moderate to high correlations between IQE and measures of intelligence(3). Using IQE scores less than 70 we have found 16.5% of our sample of children aged 0-5 years attending well child clinic in the department of Pediatrics to be delayed(1). Our results are in line with previous research, which has identified about 20% of the children to be delayed using other diagnostic tests including Bayley Developmental Scales, Kaufman Assessment Battery for Children, Stanford-Binet intelligence scale(10,11). It is obvious that the method of interpretation of scores of DP II used by Glascoe(10) detects delayed children at unacceptably low rates and perhaps misses some cases. Therefore rather than dismiss the DP II as being inaccurate, it would be more meaningful to use the second method of interpreting scores on the DP II as we have done in our study.

4. Finally, we agree with Dr. Glascoe that perhaps PEDS would benefit from an alternative scoring system for use in India. We are in the process of publishing a study using PEDS on a larger sample and would look at our data via logistic regression analyses as suggested by her.