The National Hospital Discharge Survey records for medical-legal, spontaneous, and "other" abortions (ICDA-8 640-641, 643, and 644 respectively) for 1970-1977 were analyzed to investigate the impact of liberalized access to abortion on abortion-related morbidity in the United States. The analysis suggests that in census regions where an increase in medical-legal abortions performed in hospitals occurred over the study period there was an associated decreased likelihood of a "spontaneous" or "other" abortion. The spontaneous and "other" abortion codes appear to have been used synonymously and a small number of each used to classify complications of both illegal and legal abortions performed outside hospitals. There was a significant reduction in length of stay for spontaneous and other abortions between 1970 and 1977. This is suggestive of decreasing severe presenting symptomatology for complicated abortion. The study further suggests that: during 1970-77 illegal abortions were largely replaced by legal procedures; increases in legal abortions beyond those replacing illegal have not resulted in increased rates of hospitalization for complicated abortion; and, the case morbidity rate for legal abortion appears to have declined.