a picket line to obtain an abortion is undoubtedly stressful; crossing the same picket line for a contraceptive visit—or even worrying that obtaining contraception could require confronting protesters—might be enough to avoid making the trip altogether and cause a couple to rely instead on less effective, nonprescription methods of birth control.
With regard to the second issue—that the ratio between unintended pregnancies resolved by abortion rather than childbearing may be affected by opposition to abortion—there are some data to suggest a possible connection. Recently, the number of abortions performed in the United States has begun to decline; in 1992, 1.5 million abortions occurred versus 1.6 million in 1990 (Henshaw and Van Vort, 1994), and a decreasing proportion of all pregnancies, including unintended pregnancies, are now being resolved by abortion (Henshaw and Van Vort, 1994; Henshaw et al., 1991) (see also Chapter 2). It is possible that at least one reason for such trends is the discomfort and fear among both patients and providers caused by the harassment described above. Other factors that may help to explain the decreasing number of abortions include a changing age structure in the population, with more women in the older age groups among whom abortion is less common, and less punitive attitudes towards nonmarital childbearing (Henshaw and Van Vort, 1994).
An additional factor that is probably contributing to the decrease is the decline in the number of facilities that perform abortions. Between 1978 and 1992, for example, the number of counties that reported the presence of at least one abortion provider declined by more than 30 percent, such that by 1992 more than 80 percent of all counties in the United States and more than half of all metropolitan counties had no abortion provider at all. Similarly, it is estimated that over the last decade and more, the number of clinics, hospitals, and physician offices that perform abortions has declined by approximately 65 a year (Henshaw and Van Vort, 1994). Again, this may reflect fear among providers, some of whom are understandably reluctant to work at an clinic that provides abortion services. The decline in the number of abortion providers may also reflect the limited training in the procedure that many physicians now receive, particularly as compared to several years ago. Goldstein (1995:A11) recently reported that "in 1975, two years after Roe, all but 7 percent of U.S. medical schools offered training in abortion to obstetrics residents, and 26 percent required it. By 1992, one third were not given any training, even when residents requested it, and 12 percent included it as a requirement." Similarly, Westoff and colleagues (1993) report that most programs training family practice physicians as well as obstetrician-gynecologists do not require competency in this procedure. Although the Accreditation Council for Graduate Medical Education recently revised its requirements for obstetrics-gynecology residency programs to insist that training in abortion be provided (with a few narrowly defined exceptions for religious or moral objections), these new requirements will not