Much attention has been focused on the potential of medical abortion to change the character of pregnancy termination services in the United States. Because medical abortion regimes do not require surgical training and, hypothetically, can be administered by any physician with prescription-writing privileges, the availability of medical abortion could theoretically expand women's access to induced abortion services, especially in areas that now lack surgical providers.¹ In addition, medical abortion may make the delivery of abortions in private physicians' offices more feasible, and thus allow women to bypass freestanding clinics, which have been the target of harassment by antiabortion protesters.

The best known of the medical abortion regimes involves the use of the antiprogestin mifepristone (also known as RU-486 or the "French abortion pill") and the prostaglandin misoprostol. This combination is currently unavailable for general use in the United States, even though it has received provisional approval from the Food and Drug Administration (FDA). However, an abortion rights advocacy group (Abortion Rights Mobilization) made up to 10,000 doses of mifepristone available for research purposes; thus, there presumably will be a small but steady number of mifepristone abortions in the United States into the foreseeable future.²

Methotrexate, another drug used in combination with misoprostol for medical abortion, is already approved in the United States for other purposes, and thus its "off-label" use for abortion provision is legally permissible. Starting in the early 1990s, accounts of successful pregnancy termination using methotrexate started to appear in both the medical and popular literature.³ Interest in this method grew rapidly within the community of abortion providers. In spring 1996, the National Abortion Federation (NAF) devoted its annual postgraduate symposium to medical abortion, including both mifepristone and methotrexate, and the Planned Parenthood Federation of America announced its intention to conduct trials of methotrexate abortions on some 2,500 patients. Moreover, NAF now lists some 97 individuals and clinics as providers of medical abortion.

Against this backdrop of growing interest in medical abortion, I conducted exploratory interviews in the fall of 1996 with long-term surgical providers of abortion about their responses to medical abortion. A small number of surveys have examined the prospective involvement of physicians in providing abortions if mifepristone became available,⁴ and a handful of analyses are available on patients' responses to medical abortion.⁵ However, only one study has undertaken in-depth interviews with surgical providers to learn about their reactions to medical abortion.⁶

The rationale for such an exploratory study is a presumption that medical abortion will not spread to a wider range of providers (i.e., those not now offering surgical abortions) until these regimes have been tested by veteran abortion providers. Thus, only by examining reactions to medical abortion in this highly committed group can we gain clues as to the likelihood of widespread diffusion of these techniques among the larger population of United States physicians.

Drawing on the NAF membership base, I interviewed 25 experienced providers of surgical abortion: physicians, mid-level practitioners and clinic counselors (two Canadian providers were among those interviewed). Some interviewees had extensive experience with medical abortion, having been involved in various clinical trials of mifepristone or methotrexate; some had only been involved with medical abortion for several months; some were still undecided about whether to adopt these new methods, and a handful—at least at the time of the interview—were emphatically against them.

I questioned providers about their initial decision to offer medical abortion; what changes in office routines this new practice entailed; and any emergencies and complications they encountered. Respondents were also asked to speculate on the prospect that medical abortion might attract new providers. What follows, therefore, may be seen as a "snapshot" of an early moment in the adoption of medical abortion techniques by U.S. abortion providers.

Doctors' Attitudes

Why Add Medical Abortion?

In all, 20 of those interviewed had already had some experience with medical abortion; all of these had used methotrexate, and six had used mifepristone. Their motivations included excitement at an innovation in a field that has seen little technical change for the last 20 years; perception of competitive pressures from other providers in their regions who were offering women these options; and "principled" commitment to make all possible options available to their patients. Individuals often reported a combination of the above, as the following case of a female obstetrician-gynecologist in private practice on the West Coast suggests: "When I first heard about it [methotrexate], I decided not to do it, because it wasn't FDA-approved. But...friends of mine [other obstetrician-gynecologists] were calling, saying they were doing it, patients were starting to inquire...and then after the last NAF meeting, I saw a lot of other people were doing this...so I went back and immediately read the articles that had been published about it, and started doing it...."

Although this physician was enthusiastic about the innovation, she acknowledged feeling apprehensive when she started offering methotrexate abortions. Although protocols allowed methotrexate use up to seven weeks after the last menstrual period, she decided to impose a cutoff date of six weeks. "I intuitively decided that the earlier you take it, the better it would work, and also the less bleeding a person would have when they abort."

The physician also speculated that the cutoff might also help avoid middle-of-the-night dilation and curettage to treat heavy bleeding, "I wanted to be able to offer the patient an option which is satisfactory for everyone but I didn't want it to cause me more trouble than it was worth...."

Changes in Office Routine

Even the most committed practitioners of medical abortion acknowledge that adding this option has—in the short run at least—been cumbersome and challenging. First, the quality and quantity of counseling for medical abortion is different from that involved in surgical abortion. Several doctors reported that patients are misinformed about the procedure and that the counseling can take far longer than for surgical abortions. It is also essential to counsel sufficiently to screen out inappropriate candidates.

The doctors who were interviewed handle this issue in different ways. Some doctors prepare a telephone tape for patients interested in medical abortion or have a staff person record such a tape. All have prepared extensive literature for patients, comparing medical and surgical abortion and identifying the primary disadvantages of medical abortion: the long period of bleeding that may extend for up to six weeks, and not knowing when the abortion will actually take place.

Some doctors in private practice make a point of personally counseling all prospective medical abortion candidates. In clinics that have incorporated medical abortion provision, the counseling or nursing staff typically take on such duties. An abortion provider in the South, who personally counsels all medical abortion recipients himself, acknowledges that this counseling "takes twice as much of my time as talking to a surgical patient."

The additional time taken for counseling for medical abortions is not just for the patient's benefit, but for the counselor's benefit as well. The counselor—whether the physician or a staff person—must decide whether the prospective candidate is suitable for the method, whether she is responsible enough to return for the necessary visits, whether she is likely to report complications promptly and whether she has an adequate support system (e.g., that can get her to a hospital emergency room if necessary). Doctors, clinic administrators and counselors who do such screening sometimes encounter ambiguous cases where the patient meets all formal criteria but "seems" unreliable. In such instances, said one administrator, you find a way to "talk her out of it."

Some providers have discovered, in the course of counseling medical abortion patients, that some women do not consider the procedure a "real" abortion. At a national meeting on medical abortion, counselors from clinics already providing such procedures told of women who call clinics saying: "I...don't believe in abortion. But I can't be pregnant. Can you give me that pill that will make me stop being pregnant?"

This situation raises professional and philosophical dilemmas—and conceivably legal ones—for abortion providers. The cardinal rule of abortion counseling is that a woman should not get an abortion unless it is her freely made and informed choice.⁷ This principle was designed to remove the possibility of coercion, but it was thought to be self-evident what an "abortion" is. Abortion providers now have a methodology about which there is much patient misinformation, and which—given that it seems to induce a miscarriage—can allow some women the illusion that an abortion has not taken place at all. Given the willingness of certain antiabortion organizations to recruit plaintiffs who will sue providers for allegedly misleading patients,⁸ providers of medical abortion must take special pains to ensure that patients fully understand what is involved.

The question of how medical abortion is performed has significant implications for the feasibility of integrating medical abortions with surgical abortions (and with other office routines). Issues such as whether the misoprostol is taken orally or vaginally and whether it is administered in the office or at home are still being clarified. Providers who participated in the initial United States clinical trials of mifepristone had been required to have their patients take the misoprostol orally in the office, and then remain there for four hours. This protocol caused considerable disruption in their offices and clinics, largely because of the demand for bathrooms to accommodate the nausea and diarrhea that resulted. Some providers concluded that medical abortions could not be smoothly integrated with other office or clinic activities or would have to be done on separate days.

A family practitioner in the Northwest who had participated in both mifepristone and methotrexate trials was deeply concerned about what the ultimate protocols governing mifepristone would be: "If...you have to insert the misoprostol in the office, that is going to discourage providers—appropriately—from using it." However, a subsequent trial that permitted women to insert the misoprostol vaginally at home has proved successful.⁹

Medical Abortion and Provider Skills

Given that a number of longtime abortion providers already report finding the surgical procedure tedious and routine,¹⁰ an important question concerning medical abortion is the effect of taking away the surgery, arguably the greatest "skill" factor. An obstetrician-gynecologist in the Northeast who recently started performing methotrexate abortions in a freestanding clinic candidly acknowledges that she is doing so out of "political commitment" rather than personal preference: "I think most physicians like myself...will find them harder to do, in the sense that I'm trained as a surgeon. I'm not trained to watch people miscarry. When people come in miscarrying to the hospital, I'm trained to evacuate the uterus...I think there's an underlying anxiety to [medical abortion] that is just not there when you do a [surgical] procedure and it's done in a few minutes..."

The West Coast obstetrician-gynecologist cited earlier concedes a similar threat to her identity: "[Medical abortion] makes you feel like an internist, not a gynecologist...I guess a surgical abortion gives you a sense of satisfaction in that you did something that someone else couldn't do...with medical abortion, you feel a little bit less needed, because 'everybody' may end up doing them..."

On the other hand, a Canadian family practitioner and longtime provider of surgical abortion who has performed some 1,300 methotrexate abortions observes that medical abortion "is actually more family doctor stuff. And it feels more natural for a family doctor, giving medication and helping somebody through the side effects...I started doing these in 1993 and now it's more than half of my abortion practice."

Many providers also feel an ambiguity about what actually constitutes "performing the abortion"—in sharp contrast to their experience of surgical abortions. The steps typically involved in a mifepristone abortion are counseling the woman before the abortion, taking an ultrasound, administering the mifepristone tablets, inserting the misoprostol vaginally or having the patient ingest it orally, providing as much phone consultation as needed and then verifying, usually by an additional ultrasound, the termination of the pregnancy.

The methotrexate regime is similar, except that instead of having the patient herself take mifepristone in tablet form, the provider—or a staff member—gives the patient an injection of mextrotrexate. Among the small pool of medical abortion providers interviewed, most physicians give the methotrexate shot themselves, although one acknowledged that these were the first shots she had given in years: "The nurses had to teach me to give shots! But in our state we felt we were on safer ground legally if I was giving the shot."

In contrast, a physician who provides methotrexate abortions in his private family medicine practice in the Southwest has the nurse inject the methotrexate: "I don't give the injections for anything else, immunizations on kids, anything like that. So why should I here? I don't feel I need to do that." Besides the nurse giving the shot, the patient is counseled by either a nurse or counselor and has her ultrasound taken by a technician; the physician's role in the procedure is to insert the misoprostol. This, too, could be done by others, and the physician readily acknowledges that he does it because it is important for legal reasons that he has a part in the medical abortion sequence.

Status of Ultrasound

One of the most interesting, and potentially most consequential, of all issues within medical abortion services is whether use of ultrasound is essential. Virtually all of the doctors interviewed routinely made ultrasound part of this service. Nearly all providers of surgical abortions now use ultrasound machines. However, whether use of ultrasound should be part of the required protocol of medical abortion is another matter.

Most respondents emphatically felt it unthinkable to do medical abortions without ultrasound, both to adequately size the very early pregnancies involved and to ascertain that the abortion has been completed. On the other hand, one respondent, a family practice physician, did not think it absolutely essential, pointing out that in France (where medical abortion was pioneered) ultrasound is not always used.

The use of ultrasound will ultimately affect who can be reasonably expected to provide medical abortions. While medical abortion may be more amenable to the family practice model ("managing a case") than to the surgical model ("completing a procedure"), obstetrician-gynecologists are far more likely than family practice physicians to own an ultrasound machine.

The reluctance of the family practice physician mentioned above to require use of ultrasound for medical abortion was directly tied to her concerns about increasing the pool of abortion providers: "[Requiring ultrasound] is going to...decrease access. You have to get an ultrasound in your office? That's $20,000... so you're not going to do ultrasounds yourself, but require that everyone get one? Well, now it's going to cost your patients an extra $150 [and] the radiologists are going to say, 'What's with all these early ultrasounds?'"

A related issue pertains to the cost of medical abortions: Virtually all of those who were interviewed and were offering medical abortions charge the same as they do for a first-trimester surgical abortion. Partly this is because doing so was a requirement for those participating in clinical trials; others did not want their patients choosing an unproven method of abortion on the basis of cost. However, several speculated that if the process is protracted, requiring several ultrasounds and additional office visits, a medical abortion will inevitably be more costly than a surgical abortion. Moreover, in the few instances when either a surgical procedure is needed to complete a medical abortion or when a woman must be hospitalized because of excessive bleeding, the costs will be considerably higher.

Finally, it is still too early to know how insurance plans will respond to medical abortion, especially whether they will reimburse for medical abortions at a rate comparable to surgical abortion. This issue is particularly worrisome for providers working in a managed care environment, which might be especially inhospitable to the extensive counseling and several ultrasounds associated with medical abortion.

Satisfaction with Medical Abortion

Virtually everyone who reported performing medical abortions found the experience satisfactory. The worst fear that had been voiced by many before doing medical abortions (being called to the emergency room in the early morning because a patient was hemorrhaging) had not come to pass. Indeed, most respondents were pleasantly surprised by how few emergency calls of any nature had occurred with medical abortion.

Another major fear—that many patients would not return for the required additional visits, either to have the misoprostol inserted or to ascertain the termination of the pregnancy—also proved unfounded. No one reported a case of the "nightmare" scenario—a patient who received a methotrexate injection or who took mifepristone but then did not receive the misoprostol. Providers uniformly attributed this excellent record to the extensive counseling and screening of potential patients by themselves and their staffs.

Potential for Growth in Providers

All of the 25 people interviewed were asked if they knew of anyone in their medical networks who did not provide surgical abortions, but had taken on medical abortion provision; only one could point to such an instance. A professor of obstetrics and gynecology at an East Coast university medical center that has recently started to provide methotrexate abortions recounted a telling incident: "I got a phone call from an internist who runs a program with lots of internists and family practice doctors, and they wanted to start doing it. But when he heard what was involved, he said, 'Well, we'll wait a while.' He had believed it's a pill that you give and the patient goes home and has the abortion and that's it [but] when we started talking about ultrasounds and dating the pregnancies...and [what to do] if the patient doesn't come back [for misoprostol], he said 'Well maybe that's not what I thought.'"

Although she still sees an important future for medical abortion, an East Coast-based professor of obstetrics and gynecology concedes now that her "fantasy of two years ago that everyone was going to do it" was unrealistic. For her, the issue of accurately sizing pregnancies—which to her implies the use of ultrasound—is a major obstacle in expanding the pool of providers. She sees expansion in the near future coming from within the ranks of obstetrician-gynecologists who already own ultrasound machines.

Acknowledging the possibility of a regional bias, she stated that on the East Coast, relatively few family practitioners are involved in obstetrics, and hence the pool of non-obstetrician-gynecologists with experience in sizing pregnancies is relatively small. In contrast, physicians located on the West Coast, where it is more common for family practitioners to be involved in obstetrics, were more persuaded that such doctors might enter into abortion provision.

A major issue when contemplating expanded access among medical providers who do not perform surgical abortions but will provide medical abortion is surgical backup services for the minority of medical abortion patients who will need them (perhaps 4-5%). Theoretically, the backup physician in such cases does not have to be an identified "abortion provider"; it could be whomever the family doctor has previously sent his or her patients to for an incomplete miscarriage.

But in the eyes of most respondents, the larger politics of abortion would inevitably intrude into these relationships. Whether a provider of medical abortion will feel comfortable in getting backup surgical services from a colleague, or in asking that colleague to do his or her ultrasounds for medical abortion patients, or in sending a patient to the local hospital in an emergency situation—all of these will depend heavily on the individual's sense of the larger climate surrounding abortion in that particular community.

Conclusions

Medical abortion appears to be an acceptable innovation to those who are already engaged in providing surgical abortion. While somewhat more cumbersome and time-consuming in terms of office routines, medical abortion is not so disruptive as to be unacceptable. Their worst fears concerning complications, especially those concerning the possibility of emergency room hospital visits, have not materialized. Moreover, if mifepristone becomes generally available in the United States, there quite likely will be even higher satisfaction among abortion providers and even more use among those now performing only surgical abortion, because this drug is more predictable and takes less time to induce an abortion.

However, these seasoned providers seem to expect that medical abortion will appeal to a relatively small portion of their patients. Again, more women might be likely to choose this option if mifepristone becomes widely available. The most optimistic scenario was offered by a family practice physician from the Northwest, who predicted that if mifepristone were available, maybe as many as 50% of United States women would choose medical abortion—"if the education was done right...I'm looking at 25 years from now, realistically. At another generation."

However, as long as the only available option is methotrexate, even this physician concedes a very low degree of acceptance: "It's very effective, but for an American woman who has got kids, jobs, teaching, travel, everything else—to tell her that her miscarriage is going to happen sometime in the next 7-45 days just doesn't work very well."

There was also a general consensus among those interviewed that, if legally, permitted, medical abortion is a highly logical vehicle to draw mid-level practitioners into direct abortion provision.¹¹ But, in the immediate future at least, medical abortion's potential for expanding access has perhaps been overstated.

Realistically, perhaps the first instances of truly "new" abortion providers will be obstetrician-gynecologists who quietly offer such services to selected private patients in an office setting—and who do not advertise such services.

But whether medical abortion will eventually attract many new providers no doubt depends on several factors: the ultimate status of mifepristone, with respect both to final FDA approval and to appropriate manufacturing and distributing arrangements; the actual protocols governing misoprostol insertion (in particular, whether such insertion needs to be done in the office); the continued appearance of scientific reports in medical journals reporting on the safety and efficacy of these methods; and a wide variety of issues that always surround abortion, ranging from how various methods of abortion fare in managed care contracts to whether abortion providers are adequately shielded from antiabortion harassment in particular communities.

In sum, there is every reason to expect that in a field as uniquely contested as abortion services, innovation will take place cautiously, in spite of the high degrees of efficacy and safety reported thus far for medical abortion. Perhaps the most important lesson to be drawn from these interviews, therefore, is that there is no simple "technological fix" to the problems of abortion delivery in the United States—problems that are inseparable from the larger political climate surrounding abortion.