The risk of redefining rationing

Abstract

<h3>In Reply.</h3> —The purpose of our article was to examine inconsistencies in the coding of race and Hispanic ethnicity in infant birth and death certificates and to show the implications of inconsistent coding for the calculation of infant mortality rates.¹We found that although 1.2% of infants coded as "white" at birth were coded with a different race at death and 4.3% of "black" infants at birth had a different race coded at death, 43.2% of infants coded with other races at birth were coded with different races at death. Japanese and Filipino infants were more likely to be coded with a different race than with the same race at death. Most infants whose race was inconsistently coded were coded as white at death. Racial classification at birth is most likely to be accurate; thus, we compared classifications at death with this standard. We believe that if an individual