A validation study of the Physician Notices of Birth database in the regional municipality of Ottawa-Carleton.

Abstract

The Regional Municipality of Ottawa-Carleton maintains a database of information obtained from routinely assembled Physician Notices of Birth Forms. The validity of this database was called into question when it was discovered that substantial over-reporting of stillbirth existed when compared to the hospital caseroom logbooks. The literature shows that key variables in perinatal research, such as birthweight, gestational age, mother's age and sex of baby are generally reasonably accurate. To date, however, there is no piece of research investigating chance-corrected agreement of whether a baby was born alive or still. The study problem examined in this paper is the quality of the Ottawa-Carleton Health Department's Physician Notices of Birth (PNOB) database. All stillbirths and babies with unknown outcomes born in an Ottawa hospital to women residing in Ottawa-Carleton were included in the study in addition to a random sample of livebirths. Study variables were gestational age, mother's age, birthweight, number of previous livebirths, sex, postal code and whether the baby was born alive. (Abstract shortened by UMI.)