Association between birth weight and birth arm circumference of neonates in rural Egypt

In many developing countries including Egypt, the birth weights (BWs) of most babies go unrecorded because they are born at home. Since it is difficult to record BW in the community setting, birth arm circumference (BAC) has been used as a valid proxy for BW to identify at risk babies with low birth weight (LBW). However, the validity of BAC as an indicator of the actual BW has not been assessed fully. We did this study among neonates in rural Bilbeis, Egypt, to examine the association between BW and BAC, to assess whether BAC can serve as an indicator of the actual BW, and to determine the validity indices of 3 different cut-off levels of BAC as indicators of LBW. During 1987-88, the weights and arm circumference of 148 neonates were recorded within 2 weeks of birth. We observed a strong and highly significant positive linear correlation between BW and BAC (r = 0.6188, p = 0.0001). The BAC cut-off value of < 9.5 cm was associated with the optimal combination of sensitivity (50%) and specificity (88.4%) as an indicator of LBW. In linear regression analysis BAC was found to be a significant (p = 0.00001) predictor of BW. The means of the model predicted weights for males, females, and both genders together were identical to the corresponding means of the observed values. In contingency table analysis, for 78 (52.7%) of the 148 neonates studied the observed and predicted BW categories were identical. For another 59 (39.9%) neonates, agreement with the next lower or higher BW category was observed. We feel that BAC can be used as an indicator of the actual BW in settings where routine recording of BW is currently not practicable.