Congenital anomalies are the leading cause of infant mortality among US whites and the second leading cause among US blacks. Only a few population-based studies had previously examined mortality from congenital anomalies. Researchers from the California Birth Defects Monitoring Program utilized population-based registry data to examine the effects of major anomalies on neonatal mortality. In addition, since low birth weight (LBW) is a risk factor for neonatal mortality and highly associated with congenital anomalies, they were particularly interested in the interrelations between LBW, anomalies, and neonatal mortality. The table below shows the distribution of the 174,533 live births included in the study by birth weight category (LBW vs. normal birth weight (NBW)), anomaly status (any vs. none), and neonatal mortality (deceased vs. lived). [Adapted from an assignment developed by Prof. Linda Cowan, Univ. of Oklahoma] Neonatal mortality Any anomaly, LBW (< 2500 gm) Any anomaly, NBW ( 2500 gm) No anomaly, LBW No anomaly, NBW Deceased 471 188 381 151 Lived 688 8448 9834 154372 b. We want to determine whether birth weight is a confounder of the relationship between anomaly status and neonatal mortality or it is an effect modifier – modifying the effect of anomaly status on the odds of neonatal mortality – - in this study population. Construct two appropriate 2x2 tables to investigate if birth weight is an effect modifier or confounder, or neither in this study (use rule of thumb: use 2 10% difference for confounding and 2 33% difference for effect modification). [ Note: M-H Adjusted OR = ] Neonatal mortality Any anomaly, LBW (< 2500 gm) Any anomaly, NBW ( 2500 gm) No anomaly, LBW No anomaly, NBW Deceased 471 188 381 151 Lived 688 8448 9834 154372 (Please show your work)
