Abstract

Hospital admission rates before 2 years of age are reported for two complete birth cohorts born within a geographically defined area five years apart in 1980 and 1985. Admission rates for individual children in the cohort before 2 years of age showed an increase between 1980 and 1985 from 209 to 247 children/1000. The most significant increase took place in those children admitted three or more times (multiple admissions). Children admitted once only increased significantly but there was no significant change in the number of children admitted twice. A total of 2.3% of the total cohort in 1980 and 3.6% in 1985 were admitted three or more times and accounted for over 30% of all admissions. Age specific rates are reported for the 1985 cohort; 36% of those admitted had their first admission before 3 months of age and 76% before their first birthday. The disproportionate contribution of children with multiple admissions to the admissions total renders admission data from Hospital Activities Analysis invalid unless linked to individual children. Cohort data based on a geographically defined population provide the most reliable admission rates and permit an accurate analysis of trends. The significance of and reasons for the upward trend in early childhood admissions are discussed.