Feeding and failure to thrive in early infancy

• Main Author: Colquhoun, Fiona
• Published: Durham University 2003
• Subjects: 618.9239
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.397622

The aims of this thesis were fourfold. The first aim was the early identification of cases of non organic failure to thrive in a community based study and of randomly selected controls. Cases and controls were identified at six weeks of age using Thrive Index (Wright et al, 1994). The Thrive Index was used to measure growth velocity from birth to six weeks using two weights (birth weight and six weeks weight).The second aim was to analyse the familial characteristics of case and control families to investigate whether the frequency of failure to thrive over this period was higher in more deprived families, or families with other social characteristics. The characteristics recorded were maternal education, wage earning status, home and car ownership number of previous children and religious affiliation. The results showed that there were no statistically significant differences between the two groups or any of these variables. The third aim was to investigate the early feeding behaviour of the case and control infants. This was done using two separate approaches. The first was a six week feeding questionnaire given to all mothers asking them to provide information about their infant's feeding behaviour. Case infants were more likely to be fed on demand than set times (Chi-square =5.035, df=l, p=0.025). Also, mothers of cases reported their infants’ appetite to be poorer than that of controls (Mann-Whitney U = 1494, z=-2.179, p=0.02). The second approach was to directly observe and measure the infants feeding behaviour when the infants were aged between eight and twelve weeks. This was carried out blind to eliminate experimental bias. The sucking behaviour was analysed using a method described in Woolridge and Drewett (1986). Cases and controls did not differ on any of the recorded sucking behaviour characteristics. The fourth aim was to monitor the growth of cases and controls over one year. Using regression analyses it was found that only sex predicted weight gain to the end of the first year.