Preliminary analysis of developmental observations in a survey of morbidity in infants.

Assessment of developmental status, particularly the identification of severe developmental disability, is an important outcome measure of the effect of many types of perinatal and early infant health programs. This report examines the feasibility of the use of a brief battery of developmental observations by a trained lay interviewer in the home as part of a large-scale survey of morbidity in one-year-old infants. A total of 4989 home visits was completed, and of these 3179 were to infants of low birthweight. The interviewers obtained a high degree of cooperation from the infants in this setting. Internal consistency, as measured by coefficient alpha and item-total score correlations, and inter-rater agreement for each observation on a small random sample (137) of revisits indicated reliability sufficient for research purposes for all except receptive language items. Good agreement was found between diagnostic and health are variables obtained through the interview with the mother and a developmental status classification based on the observations. While preliminary, these results are of importance in the design of infant follow-up programs.