Hospital Religious Affiliation and Emergency Contraceptive Prescribing Practices

Abstract

With access to reproductive health care eroding, examination of prescribing of contraception, including emergency contraception (EC), is important. We examined whether working in a family practice affiliated with a religious institution changes the likelihood of a provider prescribing EC. Our survey asked about EC prescribing practices in a range of situations. As predicted, practitioners in non–religiously affiliated practices reported higher rates of prescribing EC than those in religiously affiliated practices. In both cases, however, the practitioners’ prescribing patterns were inadequate.

At a time when federal conscience clauses,1 state legislative initiatives,2 and fundamentalist religious opposition are challenging women’s access to abortion, it is important to examine physician prescribing practices with regard to contraception. Physicians continue to be the gatekeepers for all non–over-the-counter contraceptive methods, and access to contraception has a significant impact on women’s need for abortion services. In recent years, physician and public health organizations have advocated for greater availability and use of levonorgestrel-based emergency contraception (EC) to prevent unintended pregnancy after a woman has had unprotected intercourse. This US Food and Drug Administration (FDA)–approved treatment is safe, and 75% to 95% effective (efficacy decreases over time), but must be initiated within 120 hours, or 5 days, after unprotected intercourse.3^–5 Because of this short time frame, during which women must obtain a prescription, have it filled, and take the medication, many medical organizations, including the American Academy of Family Physicians,6 the Society of Adolescent Medicine,7 and the American Public Health Association,8 have evaluated the relevant data and taken a public position that EC should be available over the counter, without a prescription.

According to the World Health Organization, the only contraindication to levonorgestrel-based EC is that it not be given to women with an already established pregnancy.9 This is not because it works as an abortifacient, but rather because it is completely ineffective in this setting, and so its use is inappropriate. It is not harmful to a fetus if taken mistakenly. Unfortunately, the FDA chose not to make it available over the counter, and Plan B (two 0.75-mg tablets of levonorgestrel) is currently available in the United States primarily by prescription, and directly from pharmacists, in only 7 states. Because of this restriction, many organizations, including the American Public Health Association, have recommended that it be prescribed in advance, wherever possible, to women of reproductive age.8

We sought to examine how family practitioners in residency training programs are prescribing EC. Family physicians provide the only available health care in large areas of the United States, including many medically underserved rural and inner-city communities.10 We compared practitioners from family medicine residencies associated with religiously affiliated institutions with those from non–religiously affiliated institutions in 6 demographically similar settings. Four practices are in New York City proper (2 religiously affiliated and 2 non–religiously affiliated), and the other 2, 1 religiously affiliated and 1 non–religiously affiliated, are located in New Jersey, near New York City. Like most family medicine training clinics, all serve a predominantly medically underserved urban patient population.

Because of the consolidation of many medical institutions, a growing number of hospitals and health care facilities are now affiliated with religious institutions.11^,12 In fact, Catholic-affiliated institutions are now among the largest nongovernmental owners of hospitals.13 These religious institutions place restrictions on the practice of medicine and limit women’s choices for reproductive health care.13^–16 Many of these hospitals have family medicine and gynecology residency programs where new generations of practitioners are being trained. We report on a survey of clinical scenarios comparing the prescribing of EC by family practitioners from religiously affiliated and non–religiously affiliated institutions.

METHODS

We performed a cross-sectional study comparing the self-reported provider prescribing practices of clinicians from 6 residency programs in the New York City metropolitan area. Three residency programs had Catholic religious affiliations and 3 did not. A resident in each program distributed the surveys to all practitioners, asking each to fill it out anonymously and return it. Faculty physicians, faculty nurse practitioners, and residents were included as participants. The surveys from the non–religiously affiliated institutions were handed out and collected in July and August of 2003, and the religiously affiliated group surveys were handed out and collected in February and March of 2004. The survey had 9 questions regarding prescribing practices of EC. We used a χ² test to compare answers to the 9 questions. Statistical significance was defined at P < .05.

RESULTS

Surveys from 81% of 93 eligible participants in the non–religiously affiliated group and 95% of 80 eligible participants in the religiously affiliated group were obtained. Faculty physicians and family medicine residents comprised the majority of respondents; nurse practitioners represented less than 10% of respondents in any individual institution, and less than 4% of the total respondents to the survey.

In 7 out of the 9 clinical scenarios, clinicians from non–religiously affiliated institutions would prescribe EC more readily than those in religiously affiliated institutions (P < .05). The 2 scenarios in which no differences were detected were: teaching women already using contraceptive pills how to use them as EC and providing EC prescriptions to men.

The results of the statistically significant questions are shown in Table 1 ▶. Non–religiously affiliated clinical settings where prescriptions were more readily given included advanced prescribing during routine checkups, visits by women of reproductive age for other primary care medical issues, and pregnancy testing visits. Non–religiously affiliated providers were more likely to give refills on prescriptions for EC. Phone-in prescribing, when women needed the medication urgently, was also more readily available from non–religiously affiliated institutions than from religiously affiliated practices. This was the case despite the fact that, around the time that the religiously affiliated group was being surveyed, there was a moderate amount of coverage in New York–area papers of the FDA consideration of over-the-counter status for EC (non–religiously affiliated institutions were surveyed 6 months earlier than the religiously affiliated institutions).17^–23

TABLE 1—

Survey Responses (%) From Religiously Affiliated (RA) and Non–Religiously Affiliated (NA) Institutions

	All/Some of the Time		Conditionallya		Rarely or Neverb
	RA	NA	RA	NA	RA	NA
Routine check-ups
Q1: I prescribe EC during check-ups or routine exams to a woman not using a continuous method.	10.4	41.7	35.1	21.4	54.5	36.9
Q2: I prescribe EC to a woman who comes in for a urine pregnancy test, who is not pregnant, and who does not desire a continuous method.	15.6	48.2	23.4	18.1	61.0	33.7
New patients
Q4: When an EXISTING patient calls to request a prescription for EC, the nurses or I call it in without requiring an office visit.	32.9	78.5	36.8	13.9b	30.3	7.6
Q5: When a NEW patient calls to request a prescription for EC, the nurses or I call it in without requiring an office visit.	5.5	26.0	56.2	62.3	38.4	11.7
Ordering for any reason, timely filling, and refills
Q6: I ask female patients of reproductive age in my office for any reason if they need a prescription for EC.	12.3	39.8	32.9	16.9	54.8	43.4
Q7: When I order EC, I give refills.	13.7	55.4	15.1	8.4	71.2	36.1
Q8: When I prescribe EC in advance, I encourage women to fill the prescription so they have the medication on hand should they need it.	23.3	46.3	16.4	14.6	60.3	39.0

Note. EC = emergency contraception.

^a For routine check-ups and ordering for any reason, timely filling, and refills, the conditional response was “only if patient asks.” For new patients, the conditional response was “patient must come in to be seen first.”

^b For new patients, response was “would not accept an appointment.”

We also asked the residents administering the survey to report on the amount of education in teaching afternoons or grand rounds on the topic of EC. None of the religiously affiliated institutions had devoted any formal education time to the topic. Two of the 3 non–religiously affiliated residencies had covered the topic, each for part of a teaching hour on contraception.

Finally, we assessed knowledge of official hospital or residency policies on prescribing EC. Results indicate that policies on EC were not clearly communicated in either the non–religiously affiliated nor the religiously affiliated institutions, and that different practitioners had different impressions of what the rules might be.

DISCUSSION

The results of our survey show that clinicians in non–religiously affiliated institutions are more likely to be both responsive to women’s requests for the use of EC and proactive about prescribing EC to women during routine visits. This was a strong association, despite a lack of awareness of the institution’s stance on prescribing of EC.

There is a wide range of restrictions, loose to rigid, applied to reproductive health care practices in religiously affiliated teaching programs.11^,13^,16^,24 Although the official policy for all Catholic-affiliated medical institutions is available online,25 in many cases the attending physicians in religiously affiliated institutions have never received written policies on what they may or may not prescribe. Thus, they practice according to either what they think they may be able to do or what they think may be approved of by the administration of their particular institution.

In the case of EC, we see that, even in non–religiously affiliated practices, access to EC is not 100%. Practitioners in the non–religiously affiliated practices would give a prescription without an office visit only 79% of the time when an established patient calls requesting EC because she has had unprotected sexual intercourse and wants to prevent a pregnancy. Of course, this 79% reflects only the willingness of the practitioner to give a prescription if the patient’s phone message were to actually reach the provider; there are further difficulties for patients in actually getting the prescription filled.26^,27

Because many young women do not have an established physician, if they were to call non–religiously affiliated practices as a new patient, and if they were to actually reach a practitioner, the practitioners report that they would phone in an EC prescription only 25% of the time. Sixty-two percent of the time, the non–religiously affiliated practitioners reported that a new patient must come in to be seen first. Because appointment availability in most New York City residency practices is not within EC’s 5-day window of effectiveness, this restriction effectively limits access or allows access only as the medication efficacy declines. In the religiously affiliated programs, providers report that existing patients would receive EC only 31% of the time, and new patients only 5% of the time, without an office visit. For women who are already patients in these practices, it would be difficult to prevent an unintended pregnancy after a condom breaks or after any episode of unprotected sexual intercourse. For women who are not enrolled patients, getting help would be highly unlikely.

There were limitations to this study. The provider practice patterns were self-reported, which allows for inherent bias. The survey questions about prescribing were phrased as “I prescribe,” which yielded answers that reflect personal preferences of providers and not the actual policies of their health care institutions. Barriers in the health care delivery system may also prevent patients from receiving EC, even if their providers want to prescribe it to them. On the other hand, there is the possibility that providers may have reported what they believed to be their clinic’s protocol for prescribing EC. Because several factors could be at work here, it is difficult to determine whether the next step is to address provider attitude, actual institutional restrictions, or perceived restrictions. Further study is warranted in this area.

Another shortcoming of this study was that the surveys were not completed simultaneously; however, because of the media coverage noted previously, that should only have made the differences less striking. In addition, because of insufficient power, 2 response categories, “all of the time” and “some of the time,” were combined into 1, as there was not a strong trend toward either category, and selection of these two responses were generally evenly divided.

These results have implications for the ability of women to easily obtain a medication that has been determined to be safe under all medical conditions.9 Levonorgestrel EC is 75% to 95% effective at preventing an unintended pregnancy when taken within 120 hours after unprotected intercourse,5 and the range of effectiveness reflects the decline of efficacy over time.3^,4 Thus, practitioners need to prescribe it readily, and women need easy, timely access to practitioners who will do so. This survey demonstrates that religious affiliation clearly creates a deterrent to prescribing EC in a wide range of clinical scenarios.

Family medicine educators have a responsibility to see that their residents are educated in the full spectrum of reproductive health practices, regardless of institutional affiliation of the residency program. The faculty should teach and model for the residents the prescribing of a full range of contraceptives in a patient-centered manner. If they cannot educate residents and then readily prescribe the medications that women need, perhaps residencies should be housed only in hospitals that are not working under religious restrictions regarding the delivery of medical care. Perhaps accreditation for religiously affiliated residencies should be questioned. Religiously based policy restrictions leave residents vulnerable to learning less, and patients vulnerable to receiving less, than the full range of reproductive health care that would be available in a non–religiously affiliated setting. Finally, in order for women to receive appropriate reproductive health services, all faculty members should question the efficacy of their delivery system if in-office visits are required for a medication whose effectiveness is time dependent.