[Information on previous spontaneous abortions in the Medical Birth Registry]

Abstract

The reliability of information concerning previous spontaneous abortions in the Danish Medical Birth Register was evaluated by a historical follow-up study based on existing registers. The material consisted of data from the Medical Birth Register, the National Patient Register and the Central Person Register concerning 567 women who each had had at least two deliveries during the period 1978-1987. The number of previous abortions registered in the Birth Register at the first and last delivery during the period was compared with the number of hospital-treated abortions identified in the National Patient Register in the period between the two deliveries. Fifty-one women reported a greater number of spontaneous abortions to the Birth Register at the second delivery than at the first. In 57% this was in accordance with the diagnoses registered in the National Patient Register, incorrect reporting of a provoked abortion was a possibility in 16%. Thirty-five women were registered in the National Patient Register as having been treated in hospital for one or more spontaneous abortions. This agreed with the information in the Birth Register in 83% of the cases. Of 80 women who reported at least one spontaneous abortion at the first delivery 23% reported a lower number at the last delivery. It is concluded that the Danish Medical Birth Register's data on previous spontaneous abortions is only partly in agreement with the information contained in the National Patient Register. The difference can partly be explained by spontaneous abortions not treated in hospitals, but the agreement is lower than that of self-administered questionnaires. Misclassification between spontaneous and provoked abortion presumably occurs frequently.