Cognitive and behavioural outcomes of non-organic failure to thrive

Abstract

In a study of failure to thrive in 1987-8, 54 children (21%) from an annual cohort attending two clinics in a deprived area of Newcastle Upon Tyne were identified as having fallen across two or more major weight centiles for a month or more during the first 18 months of life. They were studied with 52 normally growing controls selected from the same clinics. Eighty nine per cent of these children were traced for a follow up study at age 6-7 years old. This follow up study is reported in this thesis. IQ was assessed using the Weschler Preschool and Primary Scale of Intelligence. The Teacher's Report Form (Achenbach, 1991) was used to assess behaviour problems in the groups. Testers were unaware of the clinical status of the children. Height was routinely measured at school entry and the original data were analysed to determine age at the lowest centile point and severity of fall in weight gain. In an independent samples analysis, a small but statistically significant difference in height at school entry age was found, but there was no statistically significant difference between the cases and controls in IQ (mean IQ 83.6 and 87 respectively, P=0.16), or ratings of behaviour problems (TRF median problems reported 23 and 14, Mann U=1.05, P=0.297). Teacher ratings did not reach conventional levels of statistical significance in any subsequent analysis. A within case group analysis of growth data was carried out to determine if there was a larger effect on a subset of cases sharing characteristics of growth failure. The effects of chronicity, age at the lowest centile point and severity of failure to thrive were analysed. A significant association was found between IQ and severity of failure to thrive (P=.03).Analysis of weight gains showed that while the screening criterion used was sensitive, identifying a group of children with a median rate of weight gain below the 10th centile for expected weight gain, 6 had fallen no lower than their expected weight gain and 17 were only mildly growth retarded. However, the measure for severity of fall used in this study is not only a sensitive criterion, but can also distinguish between a normal fall in rate of weight gain towards the population mean and an abnormal fall away from the mean and it was this measure that was significantly associated with IQ.