Article Text
Abstract
Background Regarding controversial medical services, many have argued that if physicians cannot in good conscience provide a legal medical intervention for which a patient is a candidate, they should refer the requesting patient to an accommodating provider. This study examines what US physicians think a doctor is obligated to do when the doctor thinks it would be immoral to provide a referral.
Method The authors conducted a cross-sectional survey of a random sample of 2000 US physicians from all specialties. The primary criterion variable was agreement that physicians have a professional obligation to refer patients for all legal medical services for which the patients are candidates, even if the physician believes that such a referral is immoral.
Results Of 1895 eligible physicians, 1032 (55%) responded. 57% of physicians agreed that doctors must refer patients regardless of whether or not the doctor believes the referral itself is immoral. Holding this opinion was independently associated with being more theologically pluralistic, describing oneself as sociopolitically liberal, and indicating that respect for patient autonomy is the most important bioethical principle in one's practice (multivariable ORs, 1.6–2.4).
Conclusions Physicians are divided about a professional obligation to refer when the physician believes that referral itself is immoral. These data suggest there is no uncontroversial way to resolve conflicts posed when patients request interventions that their physicians cannot in good conscience provide.
- Principle-based Ethics
- conscience
- referral and Consultation
- treatment refusal
- philosophy of medicine
- legal aspects
- moral and religious aspects
- demographic surveys/attitudes
- cultural pluralism
Statistics from Altmetric.com
- Principle-based Ethics
- conscience
- referral and Consultation
- treatment refusal
- philosophy of medicine
- legal aspects
- moral and religious aspects
- demographic surveys/attitudes
- cultural pluralism
Introduction
Few issues in medicine pique professional and public interest more than debates over physician conscientious refusals.1–6 These debates take place within and are informed by broader disagreements over how to balance and prioritise different ethical principles and concerns in the practice of medicine. Physicians' freedom to refuse medical interventions for reasons of conscience has been defended on the grounds that medicine as a moral practice depends on physicians doing that which they in good faith believe is in the patient's interest, and also that physicians have a right to protect their integrity by acting according their values.7–10 Yet, critics argue that such refusals violate patient autonomy11–13 and unjustly make patients' access to healthcare services dependent on the personal values of individual physicians.6 14
A commonly proposed solution seeks to balance competing concerns by permitting refusals so long as the physician refers the patient to a provider who will accommodate the request.8 15–17 Dan Brock argues that this ‘conventional compromise’ respects individual physicians' integrity while fulfilling the medical profession's obligation to make the full range of legal medical interventions available to patients.15 Previous studies suggest that most physicians agree both that doctors are not obligated to do something they think is immoral and that they should provide a referral for services they are unwilling to provide themselves.18 19 But what about those situations in which a physician believes that making a referral is itself immoral? Brock and others have argued that physicians must refer in these cases or face professional sanction,15 20 but to date no empirical studies have examined the views of practicing physicians.
We examined data from a national survey to describe physicians' beliefs about whether or not they have a professional obligation to refer patients even when they believe the referral itself is immoral. In addition, we sought to clarify how theoretical ethics informs physicians' judgement in this area by asking physicians to indicate which bioethical principle—among beneficence, respect for autonomy, and justice21—is most important to their practice. Despite the prominence of these principles in medical ethics discourse, no empirical studies have assessed how physicians rank their priority with respect to clinical practice. Building on prior studies, we examined the relationships between believing that doctors are always obligated to refer, identifying autonomy as the most important principle in one's practice, and physicians' demographic, religious and sociopolitical characteristics.
Methods
The methods of this study have been described elsewhere.22 In 2009 we mailed a confidential, self-administered questionnaire up to three times to a random sample of 2000 practicing US physicians, aged 65 years or younger and from all specialties, selected from the American Medical Association Masterfile. The initial mailing included a gift, and an additional US$25 was promised to those who responded. The Mayo Clinic Institutional Review Board approved this study.
Questionnaire
Our primary criterion variable was agreement with the statement: ‘Physicians have a professional obligation to refer patients for all legal medical services for which the patients are candidates, even if the physician believes that such a referral is immoral’. We also asked: ‘Which of the following ethical principles is the most important in your practice as a physician? (1) Respect for autonomy—honouring the rights of patients to make decisions for themselves; (2) Justice—seeking fair treatment of patients based on medical need and fair distribution of healthcare resources; and (3) Beneficience/non-maleficence—promoting the wellbeing of patients and preventing illness, while minimising harm.’
Primary predictor variables were physicians' religious characteristics and sociopolitical views. Religious affiliation was categorised as: no religion, Jewish, Roman Catholic or Eastern Orthodox, non-evangelical Protestants (includes non-evangelical other Christians), evangelical Protestants (includes evangelical other Christians) and other religions. Religious salience23 24 was assessed with the question: ‘How important would you say your religion is in your life?’ Responses were: ‘the most important part of my life’, ‘very important’, ‘fairly important’, ‘not very important’ and ‘not applicable—I have no religion’; the last two categories were collapsed into one. Spirituality was measured by asking: ‘To what extent do you consider yourself a spiritual person?’ Responses were: ‘very spiritual’, ‘moderately spiritual’, ‘fairly spiritual’ and ‘not very spiritual’.
Additionally, we scored physicians on a scale of theological pluralism—the extent to which physicians believe that no religion is uniquely and comprehensively true. An earlier study found that physicians with high theological pluralism were more likely to endorse nondirective counsel in areas of moral controversy.25 We asked physicians to rate their level of agreement with three statements: (1) There is truth in one religion; (2) Different religions have different versions of the truth and each may be equally right in its own way; and (3) There is no one, true, right religion. Responses were scored on a four point scale from ‘agree strongly’ to ‘disagree strongly’. After reverse-scoring the first statement, responses were summed (Cronbach α=0.75) and scores trichotomised into low, moderate and high theological pluralism.
Sociopolitical views were measured by responses to the question, ‘How would you characterise yourself on social issues?’ Responses were: ‘conservative’, ‘moderate’, ‘liberal’ and ‘other’. Secondary predictors included age, sex, race, region of the country and medical specialty.
Statistical analyses
After generating population estimates from physicians' responses to each item, we used the χ2 test to examine associations between the two primary criterion variables, and between each criterion and each predictor. We then used multiple logistic regression to test whether bivariate associations remained after adjustment for relevant covariates. All analyses were conducted with Stata SE statistical software V.11.0. Respondents who left items blank were omitted from analysis of those items.
Results
Of the 2000 physicians surveyed, 5% (n=105) could not be contacted. Of 1895 eligible physicians, 1032 completed the survey, giving a cooperation rate of 55%.26 Table 1 displays the demographic, religious and sociopolitical characteristics of respondents.
As seen in table 2, the majority (57%) of respondents agreed that physicians have a professional duty to refer patients for all legal medical services for which the patients are candidates, even if the physician believes that such a referral is immoral. Almost two thirds (64%) indicated that beneficence was the most important ethical principle to their medical practice, one in four (26%) indicated respect for autonomy and one in 10 (10%) indicated justice.
Table 3 presents the incidence and odds of agreeing that physicians must refer even if they believe that referral is itself immoral, stratified by physicians' religious characteristics, sociopolitical views, and the ethical principle most important to their practice. After adjusting for potential covariates, physicians remained more likely to agree that they were obligated to refer if they had moderate or high theological pluralism (compared to low theological pluralism, OR 1.6, 95% CI 1.1 to 2.5 and OR 1.9, 95% CI 1.3 to 2.8, respectively), they self-identified as liberal (OR 2.4, 95% CI 1.5 to 3.8, compared to conservative) or they rated respect for autonomy as the most important ethical principle (OR 1.6, 95% CI 1.1 to 2.3, compared to beneficence/nonmaleficence).
After adjusting for relevant covariates, physicians' beliefs about referral were not associated with age, gender or region. Asian physicians were less likely than white physicians (OR 0.6, 95% CI 0.4 to 0.95), and obstetrician/gynecologists were more likely than general medicine physicians (OR 2.6, 95% CI 1.1 to 5.9), to agree that they are always obligated to refer (data not shown in tables).
In multivariate analyses, pediatricians were much less likely than general medicine physicians (OR 0.1, 95% CI 0.04 to 0.3) to indicate that autonomy is the most important ethical principle in their practice, but choosing autonomy was not associated with any religious, sociopolitical or demographic characteristics.
Discussion
In a large, contemporary survey of practicing US physicians from all specialties, we found that a small majority agrees that physicians have a professional obligation to refer patients for all legal medical services for which the patients are candidates, even if the physician believes that such a referral is immoral. This opinion is associated with being theologically pluralistic, sociopolitically liberal and/or believing that respect for patient autonomy is the most important bioethical principle in one's practice.
These data expand on previous findings about physicians' obligations when a patient requests a legal medical intervention to which their physician objects on moral grounds. Two prior studies found that most physicians (71%18 and 82%19) agree that when a patient requests a legal medical procedure to which the physician objects, the physician is obligated to provide a referral to a willing physician. This study asked explicitly about physicians' obligations when they object even to referral and finds that only slightly more than half of doctors believe that physicians are obligated to refer in those instances.
Previous research into conscience and medicine suggested that many physicians are ambivalent about their obligations in areas of moral controversy. In a prior study, 42% of physicians agreed that ‘a physician should never do what he or she believes is morally wrong, no matter what experts say’, 22% agreed that ‘sometimes physicians have a professional ethical obligation to provide medical services even if they personally believe it would be morally wrong to do so,’ and 36% agreed with both of these seemingly contradictory statements.19 The percentage of physicians in that study who believed that physicians are never obligated to violate their consciences corresponds very closely to the percentage of physicians in this present study (43%) who did not agree that physicians are obligated to make referrals that they believe are immoral.
Physician's conflicting opinions regarding referrals mirror disagreements among bioethicists, with leading figures both rejecting and defending physicians' right to refuse to refer if they believe a referral is immoral.15 27 Further complicating this issue is the reality that every clinical situation is unique; ethical rules do not always apply equally to different scenarios.28 Moreover, patients and physicians often come from different moral communities and disparate worldviews.29 As such, physicians and patients must at times negotiate complex clinical decisions without recourse to a shared ethical standard.
Our data highlight how this deliberative process depends to a real extent on the characteristics of the individual physician. Physicians who are more theologically pluralistic are more likely to believe they are always obligated to refer. Physicians who believe that neither their own nor any other religion is uniquely and comprehensively true, or that different religions or moral traditions may each be right in their own way, might sensibly accommodate requests that reflect the patient's moral valuations even if such valuations contradict those of the physician.
Likewise, physicians who describe their social views as liberal are also more likely to believe physicians should always refer. The term ‘liberal’ has many uses, so we are cautious to avoid overinterpreting this finding. However, this finding is consistent with what philosopher Charles Taylor calls ‘the liberalism of neutrality’, in which individuals make choices according to their own authentic convictions regarding what constitutes a good life.30 In such a framework, the state, and perhaps public professions like medicine, should remain neutral regarding patients' choices.
Nor is it surprising that physicians who prioritise respect for autonomy would be more accommodating of patient requests. The principle of patient autonomy receives great emphasis in the bioethics literature,31–33 and in our study one in four physicians rated autonomy as the most important bioethical principle in their clinical practice. However, we did not ask physicians to rank how they prioritise the ethical principles in morally complex scenarios and we cannot, therefore, infer which principle they believe is most important in such cases. Previous studies25 34 suggest that this proportion would probably have been higher if we had specified a morally complex scenario rather than physicians' general clinical practice. Further study is needed to draw these sorts of distinctions.
Together with earlier findings, these data make clear that consensus is narrow regarding how physicians should respond when patients request interventions to which their physicians have moral objections. Few would deny that physicians should be candid and forthcoming, taking care to not deceive or mislead the patient about the reason for the refusal or the options available. Likewise, it is widely recognised that patients have a legal right to seek all legal medical interventions, and that physician refusals for these services are made problematic and consequential for patients because professional licensing makes physicians the gatekeepers to most such interventions. Yet beyond this area of agreement, there are no uncontroversial solutions to the dilemmas posed by conscientious refusals to refer.
One proposed resolution would have physicians either leave the profession or choose specialties where they will not be asked to violate their consciences.3 14 20 Given the rapid evolution of medical practice, not to mention its segmentation and subspecialisation, those entering medical practice cannot fully anticipate whether a certain specialty will or will not coincide with their values in the future. Furthermore, this proposed resolution does not adequately address what is to be done with individuals who have a passionate interest in and aptitude for a particular clinical specialty, but who have misgivings about a small segment of that specialty's practice.
Another solution would have physicians inform patients, at the beginning of the physician-patient relationship or another reasonable time, what medical services they are and are not willing to provide.15 16 This would ostensibly enhance patient autonomy by allowing patients to seek out physicians who will at least accommodate their values. Many patients, however, have limited choices regarding their physicians, either because they live in rural or otherwise remote areas or because of their insurance status. In addition, it is unreasonable to expect patients to anticipate all circumstances that might transpire or the medical interventions they might one day request.27 35 Therefore, even if physicians make sincere efforts to proactively disclose their relevant objections to patients, conflicts will arise.
Future efforts to resolve problems posed by conscientious refusals should be informed by our findings. The conventional compromise, which permits conscientious refusals so long as physicians make timely referrals to accommodating providers, has been advanced as a way of protecting both physician integrity and patient autonomy. However, the compromise is unproblematic only when physicians can in good conscience make the referral. When they cannot, our data suggest that almost half (43%) of US physicians do not believe the conventional compromise applies. Policies that mandate referrals are therefore likely to be resisted by large portions of the profession. Less contentious, perhaps, would be policies that focus on meeting patients' interest in having increased access to controversial interventions without asking or requiring individual physicians to do what they believe is immoral.36
Our study suggests a possible role for healthcare institutions in mediating disputes over controversial medical services. Healthcare institutions have obligations not only to individual patients, but also to their broader communities.37 Moreover, healthcare institutions have the capacity to anticipate the sorts of conflicts that may emerge between physicians and patients, and to set up systems that minimise both the inconvenience to the patient and the complicity of the medical personnel.38 Some institutions are committed to providing all legal medical interventions. Others, such as Catholic hospitals, exclude those interventions that are inconsistent with their mission and identity. Either way, healthcare institutions can ask clinicians' to disclose clinically relevant objections, and should have policies and procedures to facilitate referrals, transfers of care, or other accommodations when patients' request interventions to which their physicians object.
There are additional limitations to this study. Although our response rate is consistent with other surveys of this type,39 there is a possibility that non-respondents differed in ways that biased our findings. Theological pluralism has internal consistency and has been found previously to account for difference in physicians' ethical judgements, but it remains a novel variable and should be considered provisional until further research affirms its validity. In addition, the structure of the questionnaire allowed respondents to imagine clinical scenarios specific to their practice. Future studies would benefit from vignettes that to some extent normalise how respondents think about conscientious refusals. Finally, the cross-sectional design of this study does not permit any causal inferences from statistical associations, nor can we say how physicians in fact behave in any specific instance.
Despite these limitations, this study indicates that physicians are divided about a professional obligation to refer if the physician believes that referral itself is immoral. Given the absence of consensus concerning a requirement to refer, at this time there remains no uncontroversial way to resolve conflicts posed when patients request interventions that their physicians cannot in good conscience provide.
Acknowledgments
Thanks to Michael Putman, Jiwon Helen Shin and Robert Stern for their insightful comments on earlier drafts of this manuscript, to Kenneth Rasinski for help with statistical analyses, and to the Mayo Clinic Survey Research Center and the University of Chicago Survey Lab for assistance in designing and implementing the study.
References
Footnotes
Funding This study is supported by grants from the National Center for Research Resources (NCRR), a component of the National Institutes of Health (NIH), the NIH Roadmap for Medical Research (1 KL2 RR024151, to JCT) and the National Center for Complementary and Alternative Medicine (1 K23 AT002749, to FAC). MPC was supported by funding from the Pritzker School of Medicine's Summer Research Program. RMA was funded by the Mayo Clinic Program in Professionalism and Ethics.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
Other content recommended for you
- Autonomy, religion and clinical decisions: findings from a national physician survey
- Obstetrician-gynaecologists' opinions about conscientious refusal of a request for abortion: results from a national vignette experiment
- Conflict and emotional exhaustion in obstetrician-gynaecologists: a national survey
- US primary care physicians’ opinions about conscientious refusal: a national vignette experiment
- Predictors of hospitalised patients' preferences for physician-directed medical decision-making
- The revised International Code of Medical Ethics: an exercise in international professional ethical self-regulation
- Discussing motherhood when the oncological prognosis is dire: ethical considerations for physicians
- Defending the four principles approach as a good basis for good medical practice and therefore for good medical ethics
- The impact of healthcare professionals’ personality and religious beliefs on the decisions to forego life sustaining treatments: an observational, multicentre, cross-sectional study in Greek intensive care units
- Ethics of sedation for existential suffering: palliative medicine physician perceptions - qualitative study