Alberta's infant mortality rate: the effect of the registration of live newborns weighing less than 500 grams.

VOLUME 89, NO. 3 Infant mortality is a frequently used measure of population health status. Because of concern that Canada’s infant mortality rate (IMR) was rising, Joseph and Kramer examined trends in Canada’s IMR and the effect of registering extremely low birthweight (under 500 grams) infants as live births. A live birth is defined as any infant who, after delivery, “...breathes or shows any other evidence of life, as beating of the heart, pulsation of the umbilical cord or definite movement of voluntary muscles...”. This definition does not, however, refer to the viability of the newborn and as a result there may be differences in how live births are recorded in different jurisdictions. The World Health Organization has recommended the removal of newborns weighing less than 500 grams from the calculation of IMR when they are to be compared across regions. Joseph and Kramer, noting an increase in the number of under 500 gram newborns registered as live births and assuming a 90% mortality rate for these newborns, recalculated the Canadian IMR. For 1993, this decreased the IMR for Canada from 6.3 to 5.5 per 1,000 live births. In a second report, they excluded all live births weighing less than 500 grams and again demonstrated an impact on the IMR. Joseph and Kramer concluded that infant mortality rates should be modified for the proportion of registered live births which are under 500 grams. One of the limitations acknowledged by Joseph and Kramer was that they did not directly link births with deaths to determine the actual mortality rate. We undertook the present analysis to determine the effects of the registration of live births weighing less than 500 grams on Alberta’s IMR by linking data on birth and death registrations and comparing the results with a chart review strategy described elsewhere.