Obstetrician–gynecologists, religious institutions, and conflicts regarding patient care policies

Abstract

Objective

To assess how common it is for obstetrician-gynecologists working in religiously-affiliated hospitals or practices to experience conflict with those institutions over religiously-based policies for patient care, and to identify the proportion of obstetrician-gynecologists who report that their hospitals restrict their options for treating ectopic pregnancy.

Study Design

Mailed survey, nationally representative sample of 1,800 practicing obstetrician–gynecologists.

Results

The response rate was 66%. Among obstetrician–gynecologists who practice in religiously affiliated institutions, 37% have had a conflict with their institution over religiously-based policies. These conflicts are most common in Catholic institutions (52%, adjusted Odds Ratio 8.7, 95% Confidence Interval 1.7–46.2). Few report that their options for treating ectopic pregnancy are limited by their hospitals (2.5% at non-Catholic institutions vs. 5.5% at Catholic, p=0.07).

Conclusion

Many obstetrician-gynecologists who practice in religiously-affiliated institutions have had conflicts over religiously-based policies. The effects of these conflicts on patient care and outcomes are an important area for future research.

Introduction

Religious denominations sponsor a significant share of health care institutions in the United States.¹ Catholic hospitals account for 16% of admissions to community hospitals,² and four of the ten largest health systems are Catholic.³ Such institutions often have policies regarding patient care that are derived from religious teachings, and at times those policies lead to conflicts with physicians regarding how best to care for patients. Popular media have reported recently on cases in which Catholic moral teaching has conflicted with physicians’ judgments about patient care,⁴ and a national survey of internists and family physicians found that one in five of those who had worked in religiously affiliated institutions had experienced conflict with the institution over religiously-based policies for patient care.⁵ Obstetrician-gynecologists’ experiences of conflict over religious hospital policies have not been formally examined in the literature.

Obstetrician-gynecologists are the physicians perhaps most likely to be impacted by religiously-based policies for patient care. Hospitals sponsored by a range of religious denominations restrict abortion,⁶ and Catholic institutions in particular prohibit many common and professionally accepted practices related to sexuality and reproduction. For example, the Ethical and Religious Directives for Catholic Health Care Services (hereafter ‘the Directives’), which are authoritative for all Catholic health care institutions in the United States, prohibit abortion, sterilization, contraception, and most uses of assisted reproductive technologies.⁷

One area of ambiguity has been how Catholic teaching applies to the treatment of ectopic pregnancy. The Directives state, “In case of extrauterine pregnancy, no intervention is morally licit which constitutes a direct abortion.”⁷ In the past, many had interpreted Catholic teaching as banning any direct treatment of ectopic pregnancy unless the fallopian tube had already ruptured.⁸ Today Catholic ethicists generally agree that salpingectomy may be used to treat an ectopic pregnancy (without the need to wait for tubal rupture), because in removing the diseased fallopian tube, the fetus is destroyed indirectly as a secondary effect.^8–10 However, Catholic ethicists still disagree about the moral permissibility of salpingostomy and methotrexate, two safe and effective methods that are supported by the American College of Obstetrics and Gynecology.¹¹ There are Catholic ethicists who endorse their use,¹⁰ but others argue that when the fetus has heart tones (and therefore under Catholic teaching is treated as a living person), performing a salpingostomy (to remove the embryo while leaving the fallopian tube in place) or giving methotrexate constitutes a direct abortion.⁹ In interviews some physicians working at Catholic hospitals report that their hospitals prohibit them from offering methotrexate for women with ectopic pregnancies.¹² To our knowledge, no previous research has quantitatively assessed obstetrician-gynecologists’ experiences with hospital policies that would restrict options for treating ectopic pregnancy.

This study surveyed a nationally–representative sample of practicing obstetrician–gynecologists to characterize those who practice in religiously-affiliated institutions and to determine the prevalence and correlates of physician–institution conflicts over religiously-based policies for patient care. The study also measured the proportion of obstetrician-gynecologists who say that the policies of their institution limit their options for treating ectopic pregnancy, and how that proportion varied by the religious affiliation of the institution.

Materials and Methods

Data

The methods of this study have been reported elsewhere.¹³ From October 2008 to January 2009, we mailed a self–administered confidential survey to a stratified random sample of 1,800 practicing obstetrician–gynecologists aged 65 years or younger. We obtained our sample from the American Medical Association Physician Masterfile, a database intended to include all practicing physicians in the U.S. To increase minority representation (especially minority religious perspectives), we used validated surname lists to create four strata.^14–16 We sampled 180 physicians with typical South Asian surnames, 225 physicians with typical Arabic surnames, 180 physicians with typical Jewish surnames, and 1,215 other physicians (from all those whose surnames were not on one of these ethnic lists). Physicians received up to three separate mailings of the questionnaire; the first included $20, and the third offered an additional $30 for participating. Physicians also received an advance letter and a postcard reminder after the first questionnaire mailing. The University of Chicago Institutional Review Board approved this survey. The requirement for written consent was waived, as is typical for confidential, self-administered surveys.

Variables

For the present study, we asked respondents, “Is your primary place of practice religiously affiliated?” (yes/no). Those who indicated “yes” were asked, “What is the religious affiliation of that hospital/practice?” (Jewish, Roman Catholic, Christian non–Catholic, other), and “Have you ever had a conflict with that hospital/practice over religiously-based policies for patient care?” (yes/no).

We also presented the following clinical vignette: “A 24-year-old patient presents with left lower quadrant pain. Vaginal ultrasound reveals a 7-week ectopic pregnancy implanted in the fallopian tube, with fetal heart tones present.” We then asked respondents, “Assuming it was technically feasible and you had the appropriate surgical skills, would you be willing to: a) “perform a salpingostomy in this case?” (yes/no), and b) “perform a salpingectomy in this case?” (yes/no) In addition, we asked, “Do the policies of your hospital or employer limit the options you have for treating ectopic pregnancy in cases like this one?” (yes/no).

Predictors were physician age, gender, race/ethnicity, region, immigration status (born in the US or immigrated), religious affiliation, and importance of religion. Participants indicated their religious affiliation as Hindu, Muslim, Catholic (Roman Catholic or Eastern Orthodox], Jewish, Evangelical Protestant, non – Evangelical Protestant, other, or none. They were also asked, “How important would you say your religion is in your own life?” Response options were: ‘not very important in my life’, ‘fairly important in my life’, ‘very important in my life’, and ‘the most important thing in my life’.

Statistical analysis

We utilized chi square tests for bivariate analyses and logistic regression for multivariate analyses. We carried out all analyses using the survey design adjusted commands in STATA release 11.0 (StataSoft Corp, College Station, TX). All analyses were adjusted using probability weights to account for oversampling of physicians by ethnic surname and to account for differential response rates among physicians from each of the four different strata. In this way we were able to generate estimates for the population of obstetrician-gynecologists currently practicing in the U.S. Missing data were excluded from analyses, and we considered findings significant at the p < 0.05 level.

Results

Of 1,800 physicians sampled, 40 were ineligible for this study because they either had retired or had an invalid address. The overall response rate of the survey was 66% (1154/1760). Among respondents, 19 were missing data on whether they worked in a religiously affiliated institution, and an additional 7 were missing data on whether they had experienced conflict with their institution, leaving an analytical sample of 1128.

Approximately 22% (n=241) of U.S. obstetrician–gynecologists primarily practice in religiously-affiliated institutions. The majority of these (59%, N=143) practice in Catholic institutions; 23% (n=56) in Christian non-Catholic institutions, 8% (n=19) in Jewish institutions, 9% (n=21) in institutions with other religious affiliations, and 1% (n=2) not reported. Those who work in the Northeast are less likely to work in religiously-affiliated institutions than those in the South, Midwest or West (Table 1). Those for whom religion is not personally important are also less likely to work in religiously-affiliated institutions than are their colleagues who rate religion as fairly, very, or most important. However, obstetrician-gynecologists working in religious hospitals are themselves religiously diverse and do not differ from other obstetrician-gynecologists with respect to religious affiliations. Physicians who identify as Roman Catholic are no more likely (when controlling for other characteristics) to work in a Catholic hospital (OR = 1.7 compared to those reporting no religious affiliation, 95% CI 0.7 – 4.1, data not reported in a table).

Table 1.

Characteristics of obstetrician-gynecologists, by whether they practice in a religiously-affiliated institution (n=1128)

	Practice in Religiously Affiliated Institution?
Characteristic	Yes, N=241	No, N=887	p(Χ2)
Age in years, mean (SD)	47.3 (9.0)	47.7 (9.2)	0.58
	Na (%)b	Na (%)b
Sex			0.58
Male	120 (21.0)	485 (79.0)
Female	121 (22.5)	402 (77.5)
Race/ethnicity			0.43
White, non-Hispanic	177 (23.4)	583 (76.6)
Black, non-Hispanic	11 (17.3)	54 (82.7)
Hispanic or Latino	12 (18.2)	51 (81.8)
Asian	33 (17.9)	163 (82.1)
Other	3 (14.4)	19 (85.6)
Geographic Region			0.002
Northeast	37 (12.5)	246 (87.5)
South	85 (24.1)	278 (75.9)
Midwest	67 (27.1)	179 (72.9)
West	52 (22.0)	182 (78.0)
Immigration history			0.25
Born in the U.S.	179 (22.5)	622 (77.5)
Immigrated to U.S. at any age	60 (18.8)	255 (81.2)
Religious affiliation			0.32
No religion	21 (17.3)	96 (82.7)
Hindu	15 (15.4)	73 (84.6)
Jewish	38 (25.5)	118 (74.5)
Muslim	9 (14.5)	44 (85.5)
Catholic	58 (22.5)	200 (77.5)
Evangelical Protestant	22 (24.5)	68 (75.5)
Non-evangelical Protestant	61 (21.0)	233 (79.0)
Other religion	15 (34.0)	31 (66.0)
Importance of religion			0.02
The most important	35 (23.5)	120 (76.5)
Very important	86 (23.4)	287 (76.6)
Fairly important	77 (25.9)	240 (74.1)
Not very important	40 (14.4)	227 (85.6)

^aCounts do not sum to 241 or 887 for all variables due to partial nonresponse.

^bPercentages are adjusted for survey design to estimate the portion of all practicing U.S. obstetrician-gynecologists with a given characteristic who practice in a religiously- or non-religiously affiliated institution. For example, 21.0% of all male ob-gyns are estimated to practice in religiously-affiliated institutions.

Among physicians who work in religiously affiliated institutions, 37% (N=90) have had a conflict with their institution regarding religiously-based policies for patient care. Those who work in Catholic institutions were most likely to report such conflicts (52%). Although age, immigration history, religious affiliation, and religious motivation were all associated in bivariate analyses with having had a conflict (Table 2), only working in a Catholic institution remained significant after adjusting for other variables (OR = 8.7, 95% CI = 1.7 – 46.2).

Table 2.

Likelihood of having experienced conflicts with religiously-affiliated institutions over religiously-based policies for patient care, by physician characteristics

	Have had conflict over religiously-based policies (n=90)
Physician Characteristics	na (%)b	p(Χ2)	Multivariable OR (95% CI)
Gender
Male	40 (31)	0.07	1.0 (ref)
Female	50 (43)		1.4 (0.7–2.9)
Geographic Region
Northeast	11 (30)	0.53	1.0 (ref)
South	31 (38)		1.6 (0.5–5.3)
Midwest	30 (55)		1.1 (0.3–3.9)
West	18 (31)		0.4 (0.1–1.8)
Immigration history
Born in the U.S.	75 (41)	0.003	1.0 (ref)
Immigrated to U.S. at any age	15 (18)		0.4 (0.1–1.5)
Religious affiliation (physician)
No religion	8 (44)	0.002	1.0 (ref)
Hindu	7 (35)		1.4 (0.2–12.9)
Jewish	16 (41)		1.6 (0.3–8.1)
Muslim	2 (22)		0.6 (0.1–3.7)
Catholic	21 (35)		0.7 (0.2–2.9)
Evangelical Protestant	1 (5)		0.1 (0.0–1.3)
Non-evangelical Protestant	25 (41)		0.9 (0.2–3.6)
Other religion	10 (76)		4.4 (0.2–22.9)
Importance of religion
Most important	8 (20)	0.010	1.0 (ref)
Very important	30 (30)		1.0 (0.3–3.4)
Fairly important	34 (49)		1.8 (0.5–6.1)
Not very important	18 (48)		1.9 (0.4–8.9)
Hospital religious affiliation
Other religious facility	3 (16)	<0.001	1.0 (ref)
Jewish facility	1(9)		0.6 (0.0–8.4)
Christian, non-Catholic facility	9 (17)		1.9 (0.3–11.7)
Catholic facility	77 (52)		8.7 (1.7–46.2)c

^aCounts do not sum to 90 for all variables due to partial nonresponse

^bPercentages are adjusted for survey design to estimate the portion of all practicing U.S. obstetrician-gynecologists working in religious institutions with a given characteristic who have had conflict. For example, 31% of all male ob-gyns working in religious institutions are estimated to have had a conflict.

^cp<0.05

With respect to treating an ectopic pregnancy with fetal heart tones present, the great majority of obstetrician-gynecologists would be willing to perform a salpingectomy and/or a salpingostomy (Table 3). Furthermore, few (n =31, 2.9%) reported that policies of their institution limit the options they have for treating ectopic pregnancy in similar cases: 2.5% of those who work in non-Catholic institutions versus 5.5% in Catholic institutions (p=0.07).

Table 3.

Willingness to perform salpingectomy and salpingostomy, and reports of hospital/employer policies that limit options for treating ectopic pregnancy, among U.S. obstetrician-gynecologists.

Physician wiliness to perform selected procedures to treat ectopic pregnancy, n	n (%)a
Salpingectomy, 1111	1006 (91.6)
Salpingostomy, 1114	1057 (95.1)
Hospital/employer limits options for treating ectopic pregnancy, n
All obstetrician-gynecologists, 1111	31 (2.9)
By hospital/practice affiliation
Non-religious, 871	21 (2.4)	P=0.14b
Religious, 240d	10 (4.4)
Roman Catholic, 143	7 (5.5)	P=0.07c
Christian, non-Catholic, 56	3 (4.6)
Jewish, 18	0 (0)
Other, 21	0 (0)

^aSurvey design-adjusted percentages of ob-gyns who responded “yes” to each question

^bComparison of respondents who work at religiously-affiliated vs. non-religiously-affiliated institutions

^cComparison of respondents who work at Catholic vs. all other (non-Catholic) institutions

^dResponses do not sum to 240 because 2 respondents did not report the religious affiliation of their hospital/practice

Comment

Among obstetrician–gynecologists who practice in religiously affiliated institutions, more than one in three has had a conflict with their institution over religiously–based patient care policies. This is true for more than half of those who work in Catholic facilities. As expected, these conflicts appear to be more common among obstetrician-gynecologists than was reported among general internists and family physicians in a prior study.⁵

These conflicts may have implications for both physicians and patients. Yoon and colleagues found that obstetrician–gynecologists who have religious–based ethical conflicts with patients and colleagues exhibit higher rates of emotional exhaustion and lower levels of empathy.17 Physicians may wish to ask detailed questions about hospital policies before signing a contract for employment, medical privileges, or office space, in order to minimize these conflicts. Similarly, patients seeking care may wish to ask about hospital policies that affect the treatments their physicians will be allowed to offer. However, particularly in rural areas and certain regions of the country, there is not always a wide variety of institutions for practitioners and patients alike to choose from.¹⁸ Furthermore, new conflicts can arise when previously non–religious facilities merge with religious ones and longstanding physicians and patients find themselves working under new policies.¹⁹

Based on obstetrician-gynecologists’ experiences, hospital policies do not frequently restrict options for treating ectopic pregnancy. While physicians at Catholic hospitals were slightly more likely (p=0.07) to report institutional restrictions than those at non-Catholic hospitals, restrictions were uncommon in all institutions. These findings suggest that although Catholic ethicists debate whether using salpingostomy and methotrexate constitute direct abortion, few institutions prohibit these practices. Confusion on this issue can lead to unnecessary delays (e.g., if physicians transfer patients to other institutions) and potentially to patient harms (e.g. from ruptured pregnancy).¹² Therefore, leaders of religiously-affiliated institutions should work to clarify and educate physicians about their policies regarding which (if any) treatments of ectopic pregnancy are prohibited. Further research is warranted to understand those less common cases in which physicians’ choices in treating ectopic pregnancy are restricted by their hospitals.

This study has several limitations. First, we only surveyed obstetrician–gynecologists, not other physicians who may provide care to patients with ectopic pregnancies, including emergency and family physicians. In addition, survey non-respondents might differ from respondents in terms of religiosity, potential for conflict, or other characteristics in ways that would bias the study’s findings. Information on religious affiliation, religiosity, and conflict was self–reported, and thus is subject to measurement error. We did not ask if respondents were aware of specific religiously-based policies in their hospitals, so it is possible that physicians disagree with policies they are unaware of and thus under-report conflict. We also did not ask whether obstetrician-gynecologists working in secular hospitals had ethical or other patient-care conflicts with their hospitals. In addition, limited survey space kept us from asking about the qualitative aspects of physicians’ conflicts with religious hospitals, if and how religious restrictions affected patient care, or the strategies they use to resolve them. In ongoing research, we are inviting survey respondents to participate in qualitative interviews to elicit more detail about the nature of their conflicts and relationships with their hospitals. Finally, our study cannot directly assess how institutional policies constrain physicians’ decisions or otherwise affect patients.

Notwithstanding these limitations, this study suggests that conflict over religiously-based patient care policies is common among obstetrician-gynecologists who work in religiously affiliated institutions, particularly Catholic institutions. Further research should explore the actual effects on patients of the Catholic Directives and other religiously-based patient care policies.