Abstract
Context
While religion often informs ethical judgments, little is known about the views of American clergy regarding controversial end-of-life ethical issues including allowing to die and physician-aid in dying or physician-assisted suicide (PAD/PAS).
Objective
To describe the views of U.S. clergy concerning allowing to die and PAD/PAS.
Methods
A survey was mailed to 1665 nationally representative clergy between 8/2014 to 3/2015 (60% response rate). Outcome variables included beliefs about whether the terminally ill should ever be “allowed to die” and moral/legal opinions concerning PAD/PAS.
Results
Most U.S. clergy are Christian (98%). Clergy agreed that there are circumstances in which the terminally ill should be “allowed to die” (80%). A minority agreed that PAD/PAS was morally (28%) or legally (22%) acceptable. Mainline/Liberal Christian clergy were more likely to approve of the morality (56%) and legality (47%) of PAD/PAS, in contrast to all other clergy groups (6%–17%). Greater end-of-life medical knowledge was associated with moral disapproval of PAD/PAS (adjusted odds ratio [AOR], 1.51; 95% CI, 1.04 to 2.19), p=0.03). Those reporting distrust in healthcare were less likely to oppose legalization of PAD/PAS (AOR 0.93; 95% CI, 0.87 to 0.99, p<0.02). Religious beliefs associated with disapproval of PAD/PAS, included “life’s value is not tied to the patient’s quality of life” (AOR 2.12; 95% CI, 0.1.49 to 3.03, p<0.001) and “only God numbers our days” (AOR 2.60; 95% CI, 1.77 to 3.82, p<0.001).
Conclusion
Most U.S. clergy approve of “allowing to die” but reject the morality or legalization of PAD/PAS. Respectful discussion in public discourse should consider rather than ignore underlying religious reasons informing end-of-life controversies.
Keywords: Bioethics, physician aid in dying, physician assisted suicide, religion, spirituality
INTRODUCTION
Respectful public discourse about the ethics of care at the end of life (EOL) requires attention to the reasons that underlie the opinions offered on both sides of moral controversies, reasons that often are religious, or have a religion-like character.1,2 In discourse about PAD/PAS, this includes fundamental views regarding the meaning of a good life and a good death, the nature of suffering and freedom, and the role of medicine in human life. Since many US citizens are religious,3–6 it is important to understand how religion shapes opinions on EOL care if they are to be included in public discourse about such controversial issues. Political philosophers such as Habermas7 and Sandel8 have argued that mutual respect among citizens of a good society that includes both religious and non-religious persons requires an attempt to understand the positions of others on their own terms.
The majority of Americans rely upon their religious beliefs to cope with life and its challenges, especially with increasing age9 and within serious illness.10,11 Likewise, multiple studies have reported associations between religion and attitudes about legalization of PAD/PAS within the American population,12,13 among patients,14,15 U.S. physicians,16 and concerning suicide generally in international comparisons.17 Additionally, half of terminally-ill patients are visited by their community religious leaders,2,18 who have formed opinions about a good death19 and of how to provide spiritual care at the EOL,20,21 and whose religious care is associated with end-of-life medical discussions2 and outcomes.22 The salient role of community clergy has also been recognized by the medical community in EOL care, including within national palliative care guidelines.23,24 However, while recognized to have a clear role at life’s end, what religious leaders believe about controversial EOL ethical issues such as PAD/PAS is not well-defined. For example, prominent religious leaders, from Desmond Tutu to Pope Francis, have both supported and opposed the legalization of PAD/PAS.25,26 Yet little is known about what proportion of religious leaders favor or oppose PAD/PAS or why they hold these opinions.
The National Clergy Project on End-of-Life Care is an NCI-funded cross-sectional study of a nationally random sample of religious leaders in the U.S. designed to measure attitudes related to EOL ethics, including opinions concerning ‘allowing to die’ and PAD/PAS. Community clergy were chosen because of their direct influence over congregants’ ethical perspectives, and because they have a symbolic moral voice that many Americans look to for guidance in the formation of their own beliefs. This study aims to describe religious leaders’ attitudes and opinions of PAD/PAS, and identify predictors of clergy viewpoints.
METHODS
Sample
Methods for the study are previously reported.2 From August 2014 to February 2015, a confidential, self-administered, eight-page questionnaire in English and Spanish (see Supplementary Appendix) was developed by an interdisciplinary, expert panel, and mailed to a random sample of 2000 practicing U.S. clergy. Clergy were randomly selected from a third-party business file (InfoGroup, Inc. Papillion, NE) intended to include all houses of worship in the United States (N=368,407). Of the 2000 potential respondents, an estimated 16.8% could not be contacted due to incorrect addresses and telephone numbers or because the institution no longer existed leaving an actual potential sample of 1665. The study oversampled minorities to compare clergy views based on race. Clergy received up to four mailings, a telephone call, and email, and were offered a $10 gift card in the initial mailing. The study was approved by the Dana-Farber/Harvard Cancer Care institutional review board.
Dependent Variables
Sometimes Allow to Die
Clergy responded to the previously used Pew question13 assessing views regarding treatment at the EOL: “Which comes closer to your view: In all circumstances, doctors and nurses should do everything possible to extend the life of a patient, or, sometimes there are circumstances where a patient should be allowed to die?”
Physician Aid in Dying or Physician-Assisted Suicide
All participants rated on a five-point scale their degree of agreement with five statements that assessed ethical opinions related to PAD/PAS. The question frame (see Table 2) was developed by a diverse expert panel, which included different opinions on PAD/PAS in order to neutrally frame the questionnaire.27 Clergy responded to statements that assessed their opinions concerning the morality and legality of PAD/PAS. Participants also responded to statements that assessed PAD/PAS views including whether choosing the time of death gives dignity, when pain is unrelenting and uncontrollable, and the applicability of the term “suicide.”
Table 2.
U.S. Religious Leaders’ Attitudes on Controversial End-of-Life Ethical Decisions (N=1005).
| Question and Response | No./Total No. | (%) |
|---|---|---|
| Which comes closer to your view? In all circumstances, doctors and nurses should do everything possible to extend the life of a patient. Or, sometimes there are circumstances where a patient should be allowed to die. | ||
|
| ||
| Always extend life | 154/972 | (15.8) |
|
| ||
| Sometimes let a patient die | 776/972 | (79.8) |
|
| ||
| Not sure | 33/972 | (3.4) |
|
| ||
| We would like to ask you about what some call “physician aid in dying” and others call “physician-assisted suicide.” This refers to patients who doctors say cannot be cured by medicine and will likely die in less than six months. Some states allow patients to request from their doctor a dose of drugs intended to cause death. Some argue that this gives patients a level of choice within dying and avoids unnecessary suffering. Others see this as an act of killing because the drugs, not the disease, cause death. To what extent do you agree with the following statements? | ||
|
| ||
| No./Total No. | (%) | |
| It is immoral no matter the circumstances | ||
|
| ||
| Disagree Strongly | 135/969 | (14.0) |
|
| ||
| Disagree Somewhat | 138/969 | (14.2) |
|
| ||
| Not Sure | 60/969 | (6.2) |
|
| ||
| Agree Somewhat | 163/969 | (16.8) |
|
| ||
| Agree Strongly | 473/969 | (48.8) |
|
| ||
| It should be a legal right to end life this way | ||
|
| ||
| Disagree Strongly | 519/963 | (53.9) |
|
| ||
| Disagree Somewhat | 142/963 | (14.8) |
|
| ||
| Not Sure | 89/963 | (9.2) |
|
| ||
| Agree Somewhat | 145/963 | (15.0) |
|
| ||
| Agree Strongly | 68/963 | (7.1) |
|
| ||
| Choosing the time of death gives back dignity | ||
|
| ||
| Disagree Strongly | 491/966 | (50.9) |
|
| ||
| Disagree Somewhat | 168/966 | (17.4) |
|
| ||
| Not Sure | 82/966 | (8.5) |
|
| ||
| Agree Somewhat | 161/966 | (16.7) |
|
| ||
| Agree Strongly | 64/966 | (6.6) |
|
| ||
| It is morally OK if pain is unrelenting and uncontrollable | ||
|
| ||
| Disagree Strongly | 404/962 | (42.0) |
|
| ||
| Disagree Somewhat | 166/962 | (17.2) |
|
| ||
| Not Sure | 85/962 | (8.9) |
|
| ||
| Agree Somewhat | 201/962 | (21.0) |
|
| ||
| Agree Strongly | 106/962 | (11.0) |
|
| ||
| It is suicide even when the patient is actively dying | ||
|
| ||
| Disagree Strongly | 173/962 | (18.0) |
|
| ||
| Disagree Somewhat | 153/962 | (15.9) |
|
| ||
| Not Sure | 88/962 | (9.1) |
|
| ||
| Agree Somewhat | 160/962 | (16.6) |
|
| ||
| Agree Strongly | 388/962 | (40.4) |
Independent Variables
Demographics
Clergy age, race, gender, geographic location, educational level, congregational position, and religious/denominational affiliations were collected by database or self-report. Clergy estimated average congregational annual household income.
End-of-Life Medical Knowledge
Clergy completed a 9-item questionnaire (see on-line supplement) on knowledge of hospice, palliative care, pain treatment, and ICU-care, generating a composite score on EOL knowledge (possible scores 0 to 9).
Distrust in the Healthcare System
Clergy completed a modified 4-item validated questionnaire assessing level of distrust in the healthcare system,28 generating a composite score on distrust (possible scores 4 to 20), with higher scores meaning increasing distrust.
Life’s Value Not Tied to Quality of Life
On a five-point scale participants rated their level of agreement with the statement: “The value of a patient’s life is not tied to the patient’s quality of life.”
Pain and Suffering have Purpose
On a five-point scale participants rated their level of agreement with a congregant stating: “I endure painful medical procedures because suffering is part of God’s way of testing me.”
Only God Numbers our Days
On a five-point scale clergy indicated the level of importance to express to dying congregants that “Only God numbers our days.”
Analysis
Weighted analysis accounts for sampling strategy and differences in response rates according to respondents’ race including Black clergy (11.2/22.4=0.5), Hispanic clergy (4.4/8.4=0.52) and White/Other (84.4/69.2=1.22). Multivariate (MVA) logistic regression analyses were used to identify predictors of clergy viewpoints on sometimes “allowing to die” and PAD/PAS. Multivariable models adjusted for clergy gender, age, years in ministry, race, educational level, geographical region, congregational median income, higher EOL medical knowledge, distrust in the healthcare system, and agreement with religious beliefs including the value of life apart from quality of life, spiritual purpose in suffering, and that only God numbers our days.
All reported p-values are two-sided and considered significant when less than 0.05. Statistical analyses were performed with STATA (Stata/IC 14.1, College Station, TX).
RESULTS
Among eligible clergy, the response rate was 60% (1005/1665) based on the American Association for Public Opinion Research definition IV.29 Case weights accounted for different response rates among White (69%), Black (43%), and Hispanic (43%) clergy.
Sample Characteristics
Clergy and congregational characteristics are listed in Table 1. Most religious leaders surveyed identified with Christianity (98%).
Table 1.
National Characteristics of United States Religious Leaders and Associated Congregations (N=1005).
| No./Total No. | (%) | |
|---|---|---|
| Clergy Demographic Information | ||
| Male gender | 816/982 | (83.1) |
| Age, Mean (SD) | 54.3 | (13.2) |
| Self-Reported Race/Ethnicity | ||
| Asian | 12/952 | (1.3) |
| Black or African-American | 104/952 | (10.9) |
| American Indian or Alaskan Native | 5/952 | (0.5) |
| White or Caucasian | 809/952 | (85.0) |
| Other | 34/952 | (3.6) |
| Do you consider yourself Hispanic or Latino? | 37/952 | (3.9) |
| Region | ||
| South | 385/983 | (39.2) |
| Mid-West | 292/983 | (29.7) |
| Northeast | 146/983 | (14.8) |
| West | 160/983 | (16.3) |
| Current Position | ||
| Senior, Solo, Interim Minister | 919/974 | (94.4) |
| Associate or Assistant Minister | 28/974 | (2.9) |
| Lay (non-ordained) Minister | 16/974 | (1.6) |
| Highest Level of Education | ||
| Non-College Graduate | 51/952 | (5.3) |
| 4-year Bachelor’s Degree | 109/952 | (11.5) |
| Non-Masters Certificate from Seminary | 118/952 | (12.4) |
| Master’s Degree (e.g., Master of Divinity) | 517/952 | (54.3) |
| Doctor of Ministry | 112/952 | (11.8) |
| Ph.D. | 45/952 | (4.7) |
| Clergy Religious Information | ||
| Religious Identity | ||
| Buddhist | 2/959 | (0.2) |
| Orthodox | 15/959 | (1.6) |
| Jewish | 5/959 | (0.6) |
| Jehovah’s Witness | 1/959 | (0.1) |
| Latter-Day Saints | 19/959 | (2.0) |
| Muslim | 2/959 | (0.3) |
| Roman Catholic | 85/959 | (8.9) |
| Protestant | 781/959 | (81.5) |
| Christian Other | 39/959 | (4.0) |
| Other | 8/959 | (0.9) |
| Christian Tradition | ||
| Fundamentalist | 41/896 | (4.6) |
| Evangelical | 344/896 | (38.4) |
| Pentecostal | 97/986 | (10.8) |
| Mainline | 174/986 | (19.4) |
| Liberal or Progressive | 109/896 | (12.1) |
| Catholic | 56/896 | (6.3) |
| Orthodox | 13/896 | (1.5) |
| None Apply | 62/896 | (7.0) |
| Congregational Information | ||
| Average Annual Household Income in Congregation | ||
| < $40,001 | 261/932 | (28.1) |
| $40,001 – $60,000 | 349/932 | (37.5) |
| $60,001 – $75,000 | 205/932 | (22.0) |
| > $75,001 | 116/932 | (12.5) |
| Average Congregational Weekly Attendance | ||
| < 51 | 153/953 | (16.0) |
| 51–100 | 289/953 | (30.3) |
| 101–250 | 297/953 | (31.1) |
| 251–500 | 106/953 | (11.1) |
| >501 | 61/953 | (6.4) |
| Racial Composition of Congregation | ||
| 100% of Congregation of one race | 193/952 | (20.2) |
| 75% to 99% of congregation of one race | 650/952 | (68.2) |
| 50% – 74% of congregation of one race | 94/952 | (9.9) |
| < 50% of congregation of one race | 15/952 | (1.6) |
Views on End-of-Life Ethical Decisions
A large majority of religious leaders agreed that “sometimes there are circumstances where a patient should be allowed to die” (80%) versus doing “everything possible to extend the life of a patient” (16%).
A majority of religious leaders indicated (Table 2) that PAD/PAS is immoral no matter the circumstances (66%) and should not be legal (69%) in the United States; 6% were unsure on the moral question and 9% were unsure concerning legality; 28% disagreed that PAD/PAS is immoral no matter the circumstances and 22% affirmed that PAD/PAS should be a legal right. A majority disagreed that choosing the time of death gives back dignity (68%), while 23% agreed. A smaller majority (59%) disagreed, while 32% agreed that PAD/PAS is moral if pain is unrelenting or uncontrollable. Regarding whether PAD/PAS should be classified as “suicide” if the patient is actively dying, 57% agreed while 34% disagreed.
Predictors of Allowing to Die
Table 3 shows multivariate predictors of religious leaders’ likelihood of agreeing that there are circumstances where a patient should be allowed to die. Mainline and Liberal clergy (97%) were more likely to endorse ‘allowing to die’ compared to other denominational affiliations such as Pentecostals (63%), Fundamentalist (77%), Evangelicals (81%) or Catholics (84%). In addition, Black (60%) and Hispanic (58%) religious leaders were less likely than Whites (88%) to agree in ‘allowing to die.’ Religious leaders serving in congregations of higher income and with higher measures of distrust in the healthcare system were more likely to agree with ‘allowing to die’; similarly, clergy who believed that suffering holds spiritual purpose were less likely (AOR 0.59; 95% CI, 0.38 to 0.96, p=03).
Table 3.
Demographic Predictors of U.S. Religious Leaders’ Agreement that “Sometimes there are circumstances where a patient should be allowed to die.”
| “Sometimes Allow to
Die”1 N = 793 |
||
|---|---|---|
| AOR (95% CI)2 | P* | |
| Male Gender | 0.81 (0.60, 1.08) | 0.16 |
| Age | 1.01 (0.99, 1.03) | 0.36 |
| Years in Ministry | 1.00 (0.98, 1.02) | 0.83 |
| Race | ||
| White | REFERENCE | |
| Black/African American | 0.24 (0.13, 0.45) | <0.001 |
| Hispanic | 0.27 (0.12, 0.59) | 0.001 |
| Other Race | 0.29 (0.08, 1.11) | 0.07 |
| Higher Congregational Income | 1.47 (1.16, 1.86) | 0.001 |
| U.S. Region | ||
| Northeast | REFERENCE | |
| Midwest | 2.25 (1.04, 4.89) | 0.04 |
| South | 1.06 (0.54, 2.10) | 0.87 |
| West | 1.08 (0.48, 2.40) | 0.87 |
| Educational Level | ||
| < Master of Divinity Degree | REFERENCE | |
| ≥ Master of Divinity Degree | 1.02 (0.62, 1.68) | 0.07 |
| Denominational Identity | ||
| Mainline/Liberal | REFERENCE | |
| Fundamentalist | 0.16 (0.04, 0.57) | <0.01 |
| Evangelical | 0.24 (0.11, 0.56) | 0.001 |
| Pentecostal | 0.15 (0.06, 0.37) | <0.001 |
| Roman Catholic | 0.28 (0.09, 0.90) | 0.03 |
| Eastern Orthodox | 0.20 (0.03, 1.44) | 0.11 |
| Latter-Day Saints (Mormon) | 0.16 (0.02, 1.72) | 0.13 |
| Other Christian | 0.39 (0.14, 1.09) | 0.07 |
| Other World Religions | 0.09 (0.02, 0.43) | <0.01 |
| Medical Knowledge and Trust | ||
| Higher EOL Medical Knowledge3 | 1.30 (0.83, 2.05) | 0.25 |
| Distrust in the Healthcare System4 | 1.12 (1.03, 1.20) | <0.01 |
| Religiously-Informed Beliefs | ||
| Life’s Value Not Tied to Quality of Life5 | 0.73 (0.45, 1.17) | 0.19 |
| Pain & Suffering have Spiritual Purpose6 | 0.59 (0.38, 0.96) | 0.03 |
| Only God Numbers Our Days7 | 0.68 (0.42, 1.08) | 0.10 |
Bold denotes statistical significance.
“Sometimes Allow to Die” is defined as endorsement of “Sometimes there are circumstances where a patient should be allowed to die” in response to the question: “Which comes closer to your view? In all circumstances, doctors and nurses should do everything possible to extend the life of a patient. Or, sometimes there are circumstances where a patient should be allowed to die.”
Multivariate regression analysis adjusted for gender, age, years in ministry, position, race, congregational median income, geographical region, congregational median income. (Position defined as 1=Senior/Solo Minister, 0=all else, geographical region defined as by U.S. census 1=Northeast, 2=Midwest, 3=South and 4=West. Race defined as 1=White, 2=Black/African American, 3=Hispanic, 4=Other).
Higher End-of-Life Medical Knowledge was based on a median split of a 9-item summed score where “Lower EOL Knowledge” was defined as a score of 5 or less and “Higher EOL Knowledge” was defined as a score of 6 or higher. Total scores ranged from 0 to 9.
Distrust in Healthcare is a continuous variable based on a 4-item summed score with possible scores ranging from 4 to 20, with increasing scores meaning greater distrust.
QOL Does Not Measure Life’s Value is measured by the question: “The value of a patient’s life is not tied to the patient’s quality of life.” Response options dichotomized between “Agree somewhat” and “Agree strongly” versus “Disagree somewhat,” “Disagree strongly,” and “Not Sure.”
Pain and Suffering have Purpose is measured based on clergy agreement with a congregant stating: “I endure painful medical procedures because suffering is part of God’s way of testing me.” Agreement measured on a 5-point scale from “not at all” to “completely.” Agreement was defined as endorsement from “a little” to “Completely” whereas disagreement was “not at all.”
Only God Numbers Our Days” is measured based on clergy indicating that it was a pastoral priority to “Express that only God numbers our days” when visiting a congregant with less than six months to live. Participants were dichotomized if they answered “not at all,” “a little,” or “somewhat”; and considered to endorse the statement if they answered “Quite a bit” or “Completely.”
Figure 1 portrays clergy responses to “allowing to die” and legalization of PAD/PAS.
Figure 1. Religious Leaders’ Perspectives on ‘Sometimes Allow to Die’ and Views of Legalization of Physician-Assisted Suicide/Physician Aid in Dying.
The largest clergy group (52%) said “Sometimes allow to die” but opposed PAD/PAS. A second group (20%) affirmed “Sometimes allow to die” and favored PAD/PAS. A third group (14%) said “Always extend life” and opposed PAD/PAS.
Multivariate Predictors of Religious Leaders’ Moral and Legal Opinions on PAD/PAS
Table 4 provides associations of religious leaders’ opinions that PAD/PAS is immoral and should be illegal.
Table 4.
Demographic Predictors of U.S. Religious Leaders’ Perspectives on Physician Aid in Dying/Physician-Assisted Suicide.
| Immoral No Matter
Circumstances N = 826 |
Should Not Be Legal N = 825 |
|||
|---|---|---|---|---|
| AOR1 (95% CI) | P* | AOR1 (95% CI) | P* | |
| Male Gender | 0.75 (0.58, 0.97) | 0.03 | 1.00 (0.74, 1.36) | 0.98 |
| Age | 0.99 (0.97, 1.00) | 0.13 | 0.99 (0.97, 1.00) | 0.12 |
| Years in Ministry | 1.00 (0.98, 1.01) | 0.59 | 1.00 (0.98, 1.01) | 0.88 |
| Race | ||||
| White | REFERENCE | REFERENCE | ||
| Black/African American | 1.09 (0.63, 1.92) | 0.74 | 0.95 (0.54, 1.68) | 0.86 |
| Hispanic | 0.66 (0.28, 1.54) | 0.34 | 0.53 (0.22, 1.22) | 0.14 |
| Other Race | 2.77 (0.64, 12.1) | 0.17 | 2.83 (0.64, 12.6) | 0.17 |
| Higher Congregational Income | 1.03 (0.87, 1.22) | 0.68 | 0.95 (0.81, 1.12) | 0.60 |
| U.S. Region | ||||
| Northeast | REFERENCE | REFERENCE | ||
| Midwest | 1.10 (0.62, 1.96) | 0.74 | 1.16 (0.65, 2.09) | 0.60 |
| South | 0.64 (0.37, 1.12) | 0.12 | 0.79 (0.45, 1.40) | 0.43 |
| West | 0.64 (0.34, 1.21) | 0.17 | 0.77 (0.40, 1.46) | 0.42 |
| Educational Level | ||||
| < Master of Divinity Degree | REFERENCE | |||
| ≥ Master of Divinity Degree | 0.73 (0.47, 1.14) | 0.17 | 0.82 (0.52, 1.28) | 0.38 |
| Denominational Identity | ||||
| Mainline/Liberal | REFERENCE | REFERENCE | ||
| Fundamentalist | 5.53 (1.71, 17.9) | <0.01 | 5.38 (1.67, 17.3) | <0.01 |
| Evangelical | 3.45 (2.19, 5.44) | <0.001 | 4.44 (2.77, 7.09) | <0.001 |
| Pentecostal | 5.24 (2.61, 10.6) | <0.001 | 5.96 (2.91, 12.2) | <0.001 |
| Roman Catholic | 6.41 (2.73, 15.1) | <0.001 | 8.92 (3.48, 22.9) | <0.001 |
| Eastern Orthodox | 2.97 (0.30, 29.0) | 0.35 | 2.96 (0.30, 29.1) | 0.35 |
| Latter-Day Saints (Mormon) | 2.55 (0.41, 15.6) | 0.31 | 2.63 (0.44, 15.9) | 0.29 |
| Other Christian | 3.07 (1.65, 5.71) | <0.001 | 2.47 (1.34, 4.55) | <0.01 |
| Other World Religions | 0.62 (0.17, 2.33) | 0.48 | 1.15 (0.33, 3.98) | 0.82 |
| Medical Knowledge and Trust | ||||
| Higher EOL Medical Knowledge2 | 1.51 (1.04, 2.19) | 0.03 | 1.04 (0.71, 1.51) | 0.86 |
| Distrust in the Healthcare System3 | 0.96 (0.90, 1.03) | 0.26 | 0.93 (0.87, 0.99) | 0.02 |
| Religiously-Informed Beliefs | ||||
| Life’s Value Not Tied to Quality of Life4 | 1.91 (1.34, 2.71) | <0.001 | 2.12 (1.49, 3.03) | <0.001 |
| Pain & Suffering have Spiritual Purpose5 | 1.44 (0.93, 2.22) | 0.11 | 1.34 (0.85, 2.12) | 0.21 |
| Only God Numbers Our Days6 | 2.93 (2.02, 4.26) | <0.001 | 2.60 (1.77, 3.82) | <0.001 |
Bold denotes statistical significance.
Multivariate regression analysis adjusted for gender, age, years in ministry, race, congregational median income, geographical region, congregational median income. (Geographical region defined as by U.S. census 1=Northeast, 2=Midwest, 3=South and 4=West. Race defined as 1=White, 2=Black/African American, 3=Hispanic, 4=Other).
Higher End-of-Life Medical Knowledge was based on a median split of a 9-item summed score where “Lower EOL Knowledge” was defined as a score of 5 or less and “Higher EOL Knowledge” was defined as a score of 6 or higher. Total scores ranged from 0 to 9.
Distrust in Healthcare is a continuous variable based on a 4-item summed score with possible scores ranging from 4 to 20, with increasing scores meaning greater distrust.
QOL Does Not Measure Life’s Value was measured by the question: “The value of a patient’s life is not tied to the patient’s quality of life.” Response options dichotomized between “Agree somewhat” and “Agree strongly” versus “Disagree somewhat,” “Disagree strongly,” and “Not Sure.”
Pain and Suffering have Purpose was measured based on clergy agreement with a congregant stating: “I endure painful medical procedures because suffering is part of God’s way of testing me.” Agreement measured on a 5-point scale from “not at all” to “completely.” Agreement was defined as endorsement from “a little” to “Completely” whereas disagreement was “not at all.”
“Only God Numbers Our Days” was measured based on clergy indicating that it was a pastoral priority to “Express that only God numbers our days” when visiting a congregant with less than six months to live. Participants were dichotomized if they answered “not at all,” “a little,” or “somewhat”; and considered to endorse the statement if they answered “Quite a bit” or “Completely.”
Demographics
Other than gender, no other demographic factors including age, race, congregational income, U.S. region or educational level were associated with opinions related to the morality or legality of PAD/PAS (Table 4). Analysis also examined clergy views comparing those serving in locations that have legalized PAD/PAS as of the time of the survey (OR, WA, MT, and VT) versus the 46 U.S. states where laws prohibit it, but results were comparable.
Denominational identity
Affiliation was the strongest predictor of both moral and legal views among religious leaders (Table 4). Clergy in Mainline and Liberal denominations were most likely to believe PAD/PAS to be moral and should be legal (56%/47%), in contrast to Fundamentalists (14%/6%), Evangelicals (17%/11%), Pentecostals (11%/11%), and Catholics (14%/11%), who were considerably less likely to agree.
Medical Knowledge and Distrust
Clergy with higher scores in EOL medical knowledge were more likely than those who scored lower to conclude that PAD/PAS is immoral (AOR 1.51; 95% CI, 0.1.04 to 2.19, p=0.03). However, those with greater distrust in the healthcare system were less likely to oppose legalization (AOR 0.93; 95% CI, 0.87 to 0.99, p<0.02) compared to clergy who had more trust.
Religiously-Informed Beliefs
Certain beliefs were associated with negative opinions toward PAD/PAS (Table 4). Religious leaders who agree that the value of a patient’s life is not tied to the patient’s quality of life” (59%) were more likely to view PAD/PAS as immoral (AOR 1.91; 95% CI, 1.34 to 2.71, p<0.001), and should be illegal (AOR 2.12; 95% CI, 0.1.49 to 3.03, p<0.001). Likewise, those who believe that suffering has spiritual purpose (27%) were more likely to hold that PAD/PAS does not give back dignity. Clergy who uphold the belief that “only God numbers our days” (82% “a little” to “completely”; 50% “quite a bit/completely”) opposed the legality of PAD/PAS (AOR 2.60; 95% CI, 1.77 to 3.82, p<0.001).
Multivariate Predictors Concerning Dignity, Unrelenting Pain, and Terminology
As shown in Table 5, denominational identity and religiously informed beliefs consistently predicted religious leaders’ responses to statements related to dignity, unrelenting and uncontrollable pain, and adoption of suicide terminology for PAD/PAS. Mainline/Liberals were most likely to agree that PAD/PAS gives back dignity (47%), was moral if pain is unrelenting/uncontrollable (58%), and should not be termed “suicide” (66%). Contrastingly, fewer Christian Fundamentalists, Pentecostals, Evangelicals, or Catholic leaders agreed that PAD/PAS gives dignity (6%–11%), is morally warranted even with unrelenting/uncontrollable pain (14%–19%), and most believe “suicide” is a warranted term (68–74%).
Table 5.
Predictors of Religious Leaders’ Views Concerning PAD/PAS and Dignity, Pain, and Terminology.
| Choosing time of death gives
back dignity N = 812 |
Morally OK if pain is
unrelenting and uncontrollable N = 824 |
It is suicide even when
actively dying N = 818 |
||||
|---|---|---|---|---|---|---|
| AOR1 (95% CI) | P* | AOR1 (95% CI) | P* | AOR1 (95% CI) | P* | |
| Denominational Identity | ||||||
| Mainline/Liberal | REFERENCE | REFERENCE | REFERENCE | |||
| Fundamentalist | 0.10 (0.01, 0.82) | 0.03 | 0.30 (0.11, 0.84) | 0.02 | 5.14 (1.83, 14.4) | <0.01 |
| Evangelical | 0.30 (0.18, 0.51) | <0.001 | 0.29 (0.19, 0.46) | <0.001 | 2.43 (1.58, 3.73) | <0.001 |
| Pentecostal | 0.18 (0.08, 0.43) | <0.001 | 0.17 (0.08, 0.35) | <0.001 | 2.89 (1.56, 5.38) | 0.001 |
| Roman Catholic | 0.25 (0.09, 0.64) | <0.01 | 0.17 (0.07, 0.41) | <0.001 | 3.28 (1.63, 6.62) | 0.001 |
| Eastern Orthodox | -- | 0.40 (0.04, 3.74) | 0.42 | -- | ||
| Latter-Day Saints (Mormon) | -- | 1.82 (0.35, 9.42) | 0.47 | 1.17 (0.22, 6.13) | 0.85 | |
| Other Christian | 0.40 (0.20, 0.81) | 0.01 | 0.43 (0.24, 0.78) | <0.01 | 2.24 (1.27, 3.98) | <0.01 |
| Other World Religions | 1.27 (0.36, 4.47) | 0.71 | 1.44 (0.43, 4.79 | 0.55 | 1.27 (0.37, 4.27) | 0.70 |
| Medical Knowledge and Trust | ||||||
| Higher EOL Medical Knowledge2 | 1.44 (0.93, 2.23) | 0.10 | 0.79 (0.55, 1.14) | 0.21 | 1.27 (0.91, 1.78) | 0.16 |
| Distrust in the Healthcare System3 | 1.07 (0.99, 1.15) | 0.07 | 1.05 (0.99, 1.12) | 0.10 | 0.98 (0.92, 1.04) | 0.43 |
| Religiously-Informed Beliefs | ||||||
| Life’s Value Not Tied to Quality of Life4 | 0.49 (0.33, 0.73) | <0.001 | 0.54 (0.38, 0.75) | <0.001 | 2.33 (1.69, 3.21) | <0.001 |
| Pain & Suffering have Spiritual Purpose5 | 0.49 (0.28, 0.87) | 0.02 | 0.66 (0.43, 1.02) | 0.07 | 0.91 (0.62, 1.32) | 0.61 |
| Only God Numbers Our Days6 | 0.44 (0.28, 0.68) | <0.001 | 0.53 (0.37, 0.77) | 0.001 | 1.80 (1.29, 2.52) | <0.01 |
Bold denotes statistical significance.
Multivariate regression analysis adjusted for gender, age, years in ministry, race, higher congregational income, educational level, geographical region, and all covariates listed in model.
Higher End-of-Life Medical Knowledge was based on a median split of a 9-item summed score where “Lower EOL Knowledge” was defined as a score of 5 or less and “Higher EOL Knowledge” was defined as a score of 6 or higher. Total scores ranged from 0 to 9.
Distrust in Healthcare is a continuous variable based on a 4-item summed score with possible scores ranging from 4 to 20, with increasing scores meaning greater distrust.
QOL Does Not Measure Life’s Value was measured by the question: “The value of a patient’s life is not tied to the patient’s quality of life.” Response options dichotomized between “Agree somewhat” and “Agree strongly” versus “Disagree somewhat,” “Disagree strongly,” and “Not Sure.”
Pain and Suffering have Purpose was measured based on clergy agreement with a congregant stating: “I endure painful medical procedures because suffering is part of God’s way of testing me.” Agreement measured on a 5-point scale from “not at all” to “completely.” Agreement was defined as endorsement from “a little” to “Completely” whereas disagreement was “not at all.”
“Only God Numbers Our Days” was measured based on clergy indicating that it was a pastoral priority to “Express that only God numbers our days” when visiting a congregant with less than six months to live. Participants were dichotomized if they answered “not at all,” “a little,” or “somewhat”; and considered to endorse the statement if they answered “Quite a bit” or “Completely.”
DISCUSSION
This is the first study among a representative sample of U.S. clergy showing that a large majority believe that there are circumstances where the terminally-ill should be allowed to die (80%). Furthermore, a minority of U.S. religious leaders endorse moral (28%) and legal (22%) rationales in favor of PAD/PAS. While most reject that PAD/PAS offers patients dignity, smaller majorities consider PAD/PAS immoral with unrelenting and uncontrollable pain (59%) or agree that PAD/PAS should be termed “suicide” when a patient is actively dying (57%). These findings suggest a level of diversity in how religion and religious leaders view PAD/PAS, including a notable minority who are proponents in certain circumstances. The strongest predictors related to opinions concerning PAD/PAS were not demographic factors (e.g. race or socioeconomic status), but denominational identity (Mainline/Liberals who were most favorable of PAD/PAS versus most other major religious identities), and specific religious beliefs, followed by clergy understanding of medical knowledge, and their level of distrust in healthcare.
In comparison to the U.S. population, clergy are more accepting of ‘allowing to die’ but also more circumspect pertaining to PAD/PAS. Notably, while 31% of the general population say “do everything possible to save the patient’s life,”13 only 16% of clergy agree. Likewise, whereas 47% of the U.S. population approve of the legality of PAD/PAS,13 only 22% of religious leaders approve. This suggests that many religious leaders perceive a middle ground between accepting cirucmstances of ‘allowing to die’ while largely disfavoring PAD/PAS. Figure 1 combines response options between ‘allowing to die’ and clergy opinions of legalization. The largest group comprises of religious leaders (52%) who agree that while there are circumstances when patients should be allowed to die, PAD/PAS should not be legal. Fewer either combine perceptions such as ‘allowing to die’ with PAD/PAS legalization (20%), or its opposite, ‘always extend life’ and PAD/PAS illegalization (14%). Hence, a majority of clergy appear to maintain an ethical distinction that accepts circumstances for physicians to allow the terminally-ill to die but do not include circumstances where a physician intends death.
In addition to religious leaders’ viewpoints, this study aims to identify the underlying rationales driving persepctives on PAD/PAS. We hypothsized both non-religious variables and three particular religious beliefs that would be relevant.
Regarding non-religious influences, both medical knowledge and distrust in healthcare were associated with PAD/PAS in multivariate analysis. Religious leaders who have higher scores of EOL medical knowledge, a composite measure that inlcuded understanding about pain treatment and palliative care, were more likely than clergy with lower medical knowledge to believe that PAD/PAS was immoral no matter the circumstances (Table 4). One possible explanation is that as clergy understand palliative care’s present-day abilities to mitigate pain within terminal-illness, they may be more likely to conclude that PAD/PAS is morally inconsistent with their faith tradition. Additionally, those reporting greater distrust in healthcare were more likely to accept “allowing to die” and legalization of PAD/PAS. This association is surprising and reasons for this association are less clear. One hypothesis is that increasing trust in the healthcare system is connected to a confidence in the medical profession’s aim to both strive for the patient’s health against all odds and allieviate suffering. Contrastingly, distrust in healthcare may result in a legal preference for all decision-making powers to reside exclusively with a terminally-ill patient rather than on external authorities such as physicians, who are perceived to not always have the patient’s sole interests in mind. Higher trust may correlate with a view that decesion-making need not rest upon patients alone, whereas distrust may indicate that termination of ones life is a decsions that should legally rest on the patient alone. Future study should include evaluation of EOL medical knowledge and healthcare distrust in other population samples.
Three particular religious beliefs were hypothesized to influence clergy opinions of PAD/PAS. First, a majority of clergy agreed (59%) that the value of a patient’s life is not tied to the patient’s quality of life. This was the only religious belief associated with all five PAD/PAS questions (Table 4–5). This belief suggests that because human life has intrinsic value, even burdensome circumstances at the EOL do not detract from life’s sanctity. Nonetheless, this belief was not associated with doing everything possible to extend life (Table 3), corroborating that most clergy do not equate ‘allowing to die’ with medical actions intending death. Second, a majority of religious leaders recognized a belief that “only God numbers our days,” which was associated with dispositions opposing PAD/PAS (Table 4–5). This belief highlights that the locus of authority in the timing of death rests solely in the prerogative of God, who superintends life and death. Within this rationale, PAD/PAS is interpreted as a human action that rejects divine authority by conferring the locus of authority on the individual patient to choose the timing of death. Clergy who did not endorse this statement may believe that God grants freedom of will including timing over one’s own death. Third, a minority of clergy (27%) affirmed that physical pain and suffering has spiritual purpose, and this was associated with views of ‘always extend life’ (Table 3) and opposition to choosing the time of death gives back dignity (Tables 5). This view presupposes that though pain is an inherent evil and not to be sought,19 within the burdensome experience of terminal-illness, the virtue of patience within suffering holds promise of transcendent meaning and exemplifies human dignity despite illness’ oppression. Though PAD/PAS discontinues a temporary trial of physical pain, it simultaneously severs the patient from the possibility of spiritual growth yielded within suffering at the EOL. Clergy who reject this view may perceive no redeemable or necessary meaning within pain and suffering. Hence, all three religious issues seem to hold a role in forming religious leaders’ viewpoints of PAD/PAS.
Lastly, larger religious and secular trends in the U.S. may also partially suggest, (barring unforseen changes in immigration, fertility patterns, etc.) how religion might influence future attitudes of PAD/PAS. Within the general population there has been a continued decline among those who identify with Christianity, from 78% in 2007 down to 71% in 2014.3 Correspondingly, those with no religious affiliation have increased, from 16% to 23% as of 2014.3 An increasing number of Americans may thus form opinions concerning PAD/PAS without traditional religious influence. On the other hand, the Mainline and Liberal Protestant traditions have in the last few decades dramatically decreased in proportional size to the population (e.g. from 18%–2007 to 15%–2014) as well as in absolute numbers (from 41 million to 36 million).3 In contrast, conservative Protestant groups such as White and Black Evangelicals continue to grow in absolute numbers, e.g, Evangelicals expanded from 59.8 million in 2007 to 62.2 million in 2014.3 If these trends continue, it suggests that an increasing proportion of religious leaders will likely oppose its legalization or practice. However, it is unclear how those without religious affiliation will perceive the moral authority of clergy or if a greater proportion of clergy opposing PAD/PAS would affect future public viewpoints.
This study has important limitations to note. Although it seems likely that more basic religious identies and beliefs precede opinions regarding PAD/PAS, the study design demonstrates association only not causation. Additionally, although religious leaders are likely adequate representatives for their respective traditions, it is not clear if religious lay people would answer similarily regarding particular religious beliefs or if they hold similar levels of medical knowledge or distrust in the healthcare system. The interplay between religious beliefs, medical knowledge, and trust needs to be evaluated in future studies within the general population and among patients facing life-threatening illness. Finally, because our sample is derived from U.S. congregations, data on views of religious leaders of other world religions is limited, though other studies suggest similar attitudes toward suicide more generally.17
CONCLUSION
This study demonstrates that U.S. clergy agree that sometimes the terminally-ill should be allowed to die, and while a notable minority of religious leaders are proponents of PAD/PAS, most do not agree in the morality or legalization of PAD/PAS. The depth and respectfulness of public discourse and deliberation concerning controversies in care at the EOL may be improved through the evaluation and discussion of the underlying reasons, including religious reasons, that inform opinions on these topics7,8.
Supplementary Material
Acknowledgments
Funding support was provided by the National Cancer Institute (NCI) #CA156732 and the John Templeton Foundation, and The Issachar Fund, and supporters of the Initiative on Health, Religion, and Spirituality within Harvard University.
Research staff provided critical support including Christine Mitchell, MDiv, PhD candidate, Rebecca Quinones, MTS, Audra Hite, BS, and Sarah Novereske, BS.
Footnotes
Disclosures: No authors have any conflicts of interest or disclosures to report.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Contributor Information
Michael J. Balboni, Department of Psychosocial Oncology and Palliative Care, Dana-Farber Cancer Institute, Boston, MA. Initiative on Health, Religion, and Spirituality within Harvard, Boston, MA
Adam Sullivan, Department of Biostatistics, Brown University, Providence, RI
Patrick T. Smith, Harvard Medical School Center for Bioethics, Boston, MA
Danish Zaidi, Harvard Medical School Center for Bioethics, Boston, MA
Christine Mitchell, Harvard Medical School Center for Bioethics, Boston, MA
James A. Tulsky, Department of Psychosocial Oncology and Palliative Care, Dana-Farber Cancer Institute, Boston, MA. Division of Palliative Medicine. Department of Medicine, Brigham and Women’s Hospital, Boston, MA
Daniel P. Sulmasy, Georgetown University, Washington DC
Tyler J. VanderWeele, Departments of Epidemiology and Biostatistics, Harvard School of Public Health, Boston, MA. Initiative on Health, Religion, and Spirituality within Harvard, Boston, MA
Tracy A. Balboni, Departments of Radiation Oncology and Psychosocial Oncology and Palliative Care, Dana-Farber Cancer Institute, Boston, MA. Brigham and Women’s Hospital, Boston, MA. Initiative on Health, Religion, and Spirituality within Harvard, Boston, MA
WORKS CITED
- 1.Sulmasy DP. Ethos, mythos, and thanatos: spirituality and ethics at the end of life. J Pain Symptom Manage. 2013;46:447–51. doi: 10.1016/j.jpainsymman.2013.03.016. [DOI] [PubMed] [Google Scholar]
- 2.Balboni MJ, Sullivan A, Enzinger AC, et al. United States Clergy Religious Values and Relationships to End-of-Life Discussions and Care. J Pain Symptom Manage. 2017 doi: 10.1016/j.jpainsymman.2016.12.346. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.America’s Changing Religious Landscape. Pew Research Center; 2014. http://www.pewforum.org/2015/05/12/americas-changing-religious-landscape/ [Google Scholar]
- 4.Gallup. Religion; 2016. http://www.gallup.com/poll/1690/religion.aspx
- 5.Crabtree SPB. Gallup Poll News Service. 2009. What Alabamians and Iranians have in common; A global perspective on American’s religiosity offers a few surprises. [Google Scholar]
- 6.Franck R. Religious Decline in the 20th Century West: Testing Alternative Explanations. Public Choice. 2014;159:385–414. [Google Scholar]
- 7.Habermas J. Religion in the Public Square. European Journal of Philosophy. 2006;14:1–25. [Google Scholar]
- 8.Sandel MJ. Justice : what’s the right thing to do? New York: Farrar, Straus and Giroux; 2009. [Google Scholar]
- 9.Atchley RC. Spirituality and aging. Baltimore: Johns Hopkins University Press; 2009. [Google Scholar]
- 10.MacLean CD, Susi B, Phifer N, et al. Patient preference for physician discussion and practice of spirituality. J Gen Intern Med. 2003;18:38–43. doi: 10.1046/j.1525-1497.2003.20403.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Balboni TA, Vanderwerker LC, Block SD, et al. Religiousness and spiritual support among advanced cancer patients and associations with end-of-life treatment preferences and quality of life. J Clin Oncol. 2007;25:555–60. doi: 10.1200/JCO.2006.07.9046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Periyakoil VS, Kraemer H, Neri E. Multi-Ethnic Attitudes Toward Physician-Assisted Death in California and Hawaii. J Palliat Med. 2016 doi: 10.1089/jpm.2016.0160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Pew Research Center. Views on End-of-Life Medical Treatment. Nov 21, 2013. [Google Scholar]
- 14.Emanuel EJ, Fairclough DL, Emanuel LL. Attitudes and desires related to euthanasia and physician-assisted suicide among terminally ill patients and their caregivers. JAMA : the journal of the American Medical Association. 2000;284:2460–8. doi: 10.1001/jama.284.19.2460. [DOI] [PubMed] [Google Scholar]
- 15.Emanuel EJ, Onwuteaka-Philipsen BD, Urwin JW, Cohen J. Attitudes and Practices of Euthanasia and Physician-Assisted Suicide in the United States, Canada, and Europe. Jama. 2016;316:79–90. doi: 10.1001/jama.2016.8499. [DOI] [PubMed] [Google Scholar]
- 16.Curlin FA, Nwodim C, Vance JL, Chin MH, Lantos JD. To die, to sleep: US physicians’ religious and other objections to physician-assisted suicide, terminal sedation, and withdrawal of life support. Am J Hosp Palliat Care. 2008;25:112–20. doi: 10.1177/1049909107310141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Stack S, Kposowa AJ. Religion and Suicide Acceptability: A Cross-National Analysis. Journal for the Scientific Study of Religion. 2011;50:289–306. doi: 10.1111/j.1468-5906.2011.01568.x. [DOI] [PubMed] [Google Scholar]
- 18.Mamiya L. Pulpit&Pew: Research on Pastoral Leadership. 2006. River of Struggle, River of Freedom: Trends Among Black Churches and Black Pastoral Leadership. [Google Scholar]
- 19.LeBaron VT, Cooke A, Resmini J, et al. Clergy Views on a Good Versus a Poor Death: Ministry to the Terminally Ill. J Palliat Med. 2015 doi: 10.1089/jpm.2015.0176. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.LeBaron VT, Smith PT, Quinones R, et al. How Community Clergy Provide Spiritual Care: Toward a Conceptual Framework for Clergy End-of-Life Education. J Pain Symptom Manage. 2015 doi: 10.1016/j.jpainsymman.2015.11.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sanders JJ, Chow V, Enzinger AC, et al. Seeking and Accepting: U.S. Clergy Theological and Moral Perspectives Informing Decision Making at the End of Life. J Palliat Med. 2017 doi: 10.1089/jpm.2016.0545. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Balboni TA, Phelps AC, Gallivan K, Paulk ME, Wright AA, Steinhauser K, VanderWeele TJ, Prigerson HG. Provision of Spiritual Support to Advanced Cancer Patients by Religious Communities and Associations with Medical Care at the End of Life. JAMA Internal Medicine. 2013;173:1109–17. doi: 10.1001/jamainternmed.2013.903. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Puchalski C, Ferrell B, Virani R, et al. Improving the quality of spiritual care as a dimension of palliative care: the report of the Consensus Conference. J Palliat Med. 2009;12:885–904. doi: 10.1089/jpm.2009.0142. [DOI] [PubMed] [Google Scholar]
- 24.NCP Clinical Practice Guidelines for Quality Palliative Care. National Consensus Project; 2009. ( http://www.nationalconsensusproject.org/guideline.pdf) [Google Scholar]
- 25.Cagney H. Anglican bishops declare support for assisted dying. Lancet Oncology. 2014;15:e368. doi: 10.1016/s1470-2045(14)70340-7. [DOI] [PubMed] [Google Scholar]
- 26.National Catholic Reporter Newsletter. (December 30, 2014)
- 27.Magelssen M, Supphellen M, Nortvedt P, Materstvedt LJ. Attitudes towards assisted dying are influenced by question wording and order: a survey experiment. BMC Med Ethics. 2016;17:24. doi: 10.1186/s12910-016-0107-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Shea JA, Micco E, Dean LT, McMurphy S, Schwartz JS, Armstrong K. Development of a revised Health Care System Distrust scale. J Gen Intern Med. 2008;23:727–32. doi: 10.1007/s11606-008-0575-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.http://www.aapor.org/AAPORKentico/Communications/AAPOR-Journals/Standard-Definitions.aspx.