Illegal abortion: an attempt to assess its cost to the health services and its incidence in the community.

This article describes a study designed to test a method for assessing the cost to the health services of illegally induced abortion and the feasibility of estimating the incidence of induced abortion by a field interviewing approach. The participating centers included three hospitals in Ankara, Turkey; three hospitals in Ibadan, Nigeria; one hospital in Caracas and one in Valencia, Venezuela; and two hospitals in Kuala Lumpur, Malaysia. Hospitalized abortion cases were classified as induced or spontaneous or as "probably induced," "possibly induced," or "unknown" according to a classification scheme comprising certain medical criteria. The sociodemographic characteristics of induced and spontaneous abortion cases were subjected to discriminant function analysis and the discriminating variables best characterizing the induced versus the spontaneous abortion groups were identified for each center. On the basis of this analysis, the "probably" and "possibly" induced and "unknown" categories were further classified as induced or spontaneous abortion, with stated probabilities. Thus an overall estimate is made of the proportion of all hospitalized abortions that can be considered illegally induced outside the hospital. Selected results on costs of induced and spontaneous abortion are shown. The method further tested the feasibility of obtaining valid survey data on abortion from the communities studied by re-interviewing the women hospitalized for induced and spontaneous abortion six months later in their homes. This exercise showed a degree of under-reporting of abortion that varied widely among centers, even among women who had admitted illegal induction at the time of hospitalization. The feasibility of estimating the incidence of illegal abortion by field studies is discussed in the light of these findings.