Sir.—Naeye and Peters1 have done an excellent job of reviewing the data from the National Collaborative Perinatal Study. However, before one concludes that intrapartum and postpartum events do not correlate with IQ values, one must be aware of the hazards of the assumptions made by the authors and/or the computer.
By choosing outcome at a patient age of 7 years, the authors have excluded all adverse outcomes that occurred before age 7 years in those children who were unavailable for follow-up examination, either through death or attrition. It is hard to predict if requiring a one-minute Apgar score of 6 or less with perinatal variables would alter correlations. Perhaps these same variables with higher scores are found in infants with low IQ values as well. Analysis without Apgar scores would be helpful.
The single regression analysis requires that each independent variable be given a priority. Even multiple stepwise linear