ANALYZING SIBSHIP CORRELATIONS IN BIRTH-WEIGHT USING LARGE SIBSHIPS FROM NORWAY

Data from the Medical Birth Registry of Norway were used to estimate s ibship correlations in large sibships (each with greater than or equal to 5 infants among singleton live births surviving the first year of life), while adjusting for covariates such as infant gender, gestation al age, maternal age, parity, and time since last pregnancy. This samp le of 12,356 full sibs in 2,462 sibships born in Norway between 1968 a nd 1989 was selected to maximize the information on parity, and a robu st approach to estimating both regression coefficients and the sibship correlation using generalized estimating equations (GEE) was employed . In concordance with previous studies, these data showed a high overa ll correlation in birth weight among full sibs (0.48 +/- 0.01), but th is sibship correlation was influenced by parity. In particular, the co rrelation between the firstborn infant and a subsequent infant was sli ghtly lower than between two subsequent sibs (0.44 +/- 0.01 vs. 0.50 /- 0.01, respectively). The effect of time between pregnancies was sta tistically significant, but its predicted impact was modest over the p eriod in which most of these large families were completed. While thes e data cannot discriminate whether factors influencing birth weight ar e maternal or fetal in nature, this analysis does illustrate how robus t statistical models can be used to estimate sibship correlations whil e adjusting for covariates in family studies. (C) 1997 Wiley-Liss, Inc .