•  
  •  
 

Document Type

Article

Abstract

Serial data recorded at eight ages from birth to two years were analyzed for 344 infants born between 1930 and 1970. The major purposes were to demonstrate the application of a particular mathematical model to these data and to analyze the associations among the parameters of the model and their associations with adult status. Other aims were to examine possible secular trends and intrafamilial associations. The infants studied included 87 parent-offspring pairs and 239 siblings. A three-parameter model fitted well to the data for each infant without significant correlations among the parameters. These parameters were estimated birth weight (0i), the intrinsic rate of growth (02) and the pattern of growth (03). In each sex, the average values for 03 suggested that growth in weight during infancy was a linear function of the square root of age. Only small proportions of the adult variance in weight, weight/stature2, or medial or lateral calf fat thicknesses were accounted for by the estimated parameters for growth in weight during infancy, but a greater proportion of the variance in stature during adulthood was explained. These parameter estimates were compared between four groups of infants differing in date of birth. For birth dates from 1930 to about 1952, there were significant but small decreases in the estimated birth weights (θ1) and the growth patterns (θ3) became more curvilinear. Later there were increases in the θ1 and θ3 values to groups with birth dates from 1952 to 1970. These changes probably reflect sampling bias. The parent-offspring correlations for the estimated parameters were not significant, but the sib-sib correlations were almost all significant and most approximated +0.5.The differences between these two sets of correlations were presumably due to greater environmental similarity for siblings than for parent-offspring pairs, when both the parents and their offspring were measured during infancy. The goodness of fit, the small number of parameters, their clear biological interpretation and their almost complete independence from each other, make the model appropriate for other analyses of serial growth data recorded during infancy

Download

Share

COinS