Implementing a Permissive Policy: Hospital Abortion Services after Roe v. Wade

Abstract

3 hypotheses--community preferences incentives and disincentives and organizational norms--that past research identifies as possible explanations of variation in hospital abortion policies following the Supreme Court Roe v. Wade decision in 1973 are tested. Court with this decision established a national policy recognizing the right of a woman to terminate an unwanted pregnancy. A brief overview of judicial abortion policy and hospital responses to the abortion decision is included along with a review of the literature on policy implementation judicial impact and health policy. Responses from a national sample of hospital administrators revealed that almost half of the hospitals (47 percent) changed their abortion policies after 1973. A small number (5 percent) restricted their services but a much larger number (42 percent) expanded their services. Most increases in abortion services were by hospitals that changed their policy from providing only therapeutic abortions to providing elective abortions. Only a small number of hospitals changed from no abortions to therapeutic only. Data for this analysis are from a national random sample of hospitals. sample consists of all (N=325) general medical and obstetric hospitals listed in The Guide to the Health Care Field (1975) serving the largest city in 150 randomly selected counties in the U.S. dependent variable is hospital abortion policies in 1979; the variable indicates different levels of nonprovider policies as well as different levels of provider services. hypotheses suggest 3 types of independent variables--community preferences and demands incentives and disincentives and organizational norms--to explain variation in hospital abortion policy. Multiple indicators of each type of independent variable are used. Overall the results of the tests of the 3 hypotheses indicate that organizational norms have the strongest influence on abortion policies with community preferences having a moderate impact and incentive/disincentive considerations having little impact. Simple correlations fail to indicate the relative strength of each set of variables. Controlling for rival explanations it was found that only organizational variables related strongly to hospital abortion policies. With 2 exceptions--percent religious and the number of nonhospital providers in the county--all significant relationships between abortion policies and nonorganizational variables wash out.