Primary care physicians’ and psychiatrists’ willingness to refer to religious mental health providers

Abstract

Background:

Recent decades have witnessed some integration of mental health care and religious resources.

Aim:

We measured primary care physicians’ (PCPs) and psychiatrists’ knowledge of religious mental health-care providers, and their willingness to refer there.

Methods:

A national survey of PCPs and psychiatrists was conducted, using vignettes of depressed and anxious patients. Vignettes included Christian or Jewish patients, who regularly or rarely attended services. We asked whether physicians knew of local religious mental health providers, and whether they would refer patients there.

Results:

In all, 896/1427 PCPs and 312/487 psychiatrists responded. Half of PCPs (34.1%–44.1%) and psychiatrists (51.4%–56.3%) knew Christian providers; fewer PCPs (8.5%–9.9%) and psychiatrists (15.8%–19.6%) knew Jewish providers. Predictors included the following: patients were Christian (odds ratio (OR) = 2.2–2.9 for PCPs, 2.3–2.4 for psychiatrists), respondents were Christian (OR = 2.1–9.3 for PCPs) and respondents frequently attend services (OR = 3.5–7.0 for PCPs). Two-thirds of PCPs (63.3%–64%) and psychiatrists (48.8%–52.6%) would refer to religious providers. Predictors included the following: patients regularly attend services OR = 1.2 for PCPs, 1.6 for Psychiatrists, depression vignette only), respondents were Christian (OR = 2.8–18.1 for PCPs, 2.3–9.2 for psychiatrists) and respondents frequently attend services (OR = 5.1–6.3 for PCPs).

Conclusion:

Many physicians would refer patients to religious mental health providers. However, less religious PCPs are less knowledgeable about local religious providers.

Keywords

Religion spirituality therapy referrals primary care anxiety depression

Introduction

Religion and psychiatry were historically perceived as incompatible. Freud (1964) criticized religion as neurosis, and generations of psychiatrists and other psychotherapists were significantly less religious than most of American society (Bergin & Jensen, 1990; Franzblau, D’Agostino et al., 1975; Marx & Spray, 1969). However, recent evidence suggests that psychiatrists today are more religious than their predecessors (Curlin, Odell, et al., 2007), most psychiatrists value religion as important to patients (Curlin, Lawrence, et al., 2007) and empirical data support that many patients suffering from mental health concerns benefit from religion (Koenig, 2009; Koenig, Larson, & Weaver, 1998). And the success of Alcoholics Anonymous (drawing on Christianity) and Dialectical Behavioral Therapy (drawing on Buddhism) attest to the potential instrumental benefits of integrating religious elements into mental health care.

Yet, in a previous study, we observed that physicians who are themselves less religious appear to have a bias against referring patients to religious resources (Curlin, Odell, et al., 2007). We gave nonpsychiatrist physicians a vignette of a patient experiencing 2 months of ‘deep grieving’ after his wife’s death and found that nonreligious physicians were less likely to refer the patient to a clergy member or a religious counselor. This observation raised a variety of questions that the original data could not answer. Do psychiatrists share this (apparent) bias against referring patients to religious resources? Is there really an ideological bias against religious mental health counseling, or are physicians instead motivated by concerns about availability? Is this pattern limited to ‘deep grieving’, or is it generalizable to other psychiatric conditions? What role does the patient’s religion play in shaping physicians’ recommendations?

The current survey was designed as a follow-up study to address these questions. We conducted a national survey of primary care physicians (PCPs) and psychiatrists – two specialties especially likely to see patients with mental health concerns. The full survey included vignettes about depression (Lawrence, Rasinski, Yoon, Meador, et al., 2012) anxiety (Lawrence, Rasinski et al., 2013a), medically unexplained symptoms (Lawrence, Rasinski et al., 2013b) and substance abuse (Lawrence, Rasinski, Yoon, Koenig, et al., 2012). Elsewhere, we report physicians’ tendencies to encourage patients to utilize their religious communities in general (Lawrence, Rasinski, Yoon, Meador, et al., 2012; Lawrence, Rasinski et al., 2013a). Here, we report physicians’ beliefs about referring anxious or depressed patients specifically to religious mental health professionals. We measured whether physicians knew of religious mental health professionals nearby, and also how likely physicians were to refer patients to such professionals. We hypothesized that physicians’ knowledge of religious mental health providers, and their willingness to refer, would vary with how frequently the physician attended religious services, and with the physician’s religious affiliation.

The questions addressed in this study are important for several reasons. Many patients prefer to be treated by someone within their religious tradition (Belaire, Young, & Elder, 2005; Kelly & Aridi, 1996; Mitchell & Baker, 2000; Walker, Worthington, Gartner, Gorsuch, & Hanshew, 2011). Many clergy are not knowledgeable about mental health problems (Domino, 1990; Leavey, Loewenthal, & King, 2007). There are concerns that secular mental health professionals are not well equipped to treat patients with religious or spiritual concerns (although empirical support for this sentiment is lacking) (Genia, 1994; Young, Cashwell, Wiggins-Frame, & Belaire, 2002). The number of religious mental health counselors – especially Christian counselors – has grown substantially, increasing the likelihood that physicians will interact with them and their patients. For instance, the American Association of Christian Counselors (founded in 1986) now has over 50,000 members (American Association of Christian Counselors, n.d.); 45 theological seminaries in the United States and Canada offer Master’s degrees in counseling or therapy, with many programs leading to state licensure (Association of Theological Schools, n.d.); and the American Psychological Association (American Psychological Association, n.d.) currently accredits clinical psychology doctoral programs at five schools who are members of the Council for Christian Colleges and Universities (n.d.) and have designed their programs to integrate historical Christian doctrines with current psychological treatments (Azusa Pacific University, n.d.; Biola University Rosemead School of Psychology, n.d.; George Fox University, n.d.; Regent University, n.d.; Wheaton College, n.d.). Moreover, the medical literature is largely devoid of information on religious counseling.

Method

Between September 2009 and June 2010, we mailed a confidential self-administered questionnaire to a stratified random sample consisting of 1504 US PCPs and 512 US psychiatrists 65 years old or younger. The sample was generated from the American Medical Association Physician Masterfile, a database intended to include all practicing US physicians. To increase minority religious group representation in the primary care sample, we used validated surname lists (Lauderdale, 2006; Lauderdale & Kestenbaum, 2000; Sheskin, 1998) to create four strata and oversampled from these strata. We sampled (a) 121 PCPs with typical South Asian surnames, (b) 171 PCPs with typical Arabic surnames, (c) 86 PCPs with typical Jewish surnames and (d) 1126 additional PCPs (from all those whose surnames were not on one of these ethnic lists). The psychiatrist sample was not sufficiently large to warrant oversampling by ethnic surname. Physicians received up to three separate mailings of the questionnaire. The first mailing included a US$20 bill and the third offered an additional US$30 for participation. Data were double-keyed, cross-compared and corrected against the original questionnaires. The study was approved by the University of Chicago institutional review board. The survey is available online at (https://pmr.uchicago.edu/page/physician-responses-common-mental-and-behavioral-health-concerns).

The present analysis focuses on two vignette experiments that were embedded in the survey. The first vignette focused on a patient with depression, and the second focused on a patient with anxiety. Each vignette included three patient characteristics that were experimentally varied in a factorial between-subjects design. In each vignette, the patient either reported problems in his or her personal life or reported no problems. The patient was either Christian or Jewish, and the patient either rarely attended church/synagogue or regularly attended. The patients in the two vignettes were made opposite with respect to these three characteristics (e.g. a Christian patient in the first vignette was followed by a Jewish patient in the second). Vignettes are summarized in Figure 1.

Figure 1.

Texts for the vignette experiments.

The first primary criterion measure asked, ‘To the best of your knowledge, are there any mental health care professionals in your area who base their practice on (Christian or Jewish) teachings?’ The mental health-care professional’s religion (Christian or Jewish) was matched to the vignette patient’s religion (Christian or Jewish). Response options were yes/no. The question was asked after each vignette.

The second primary criterion measure asked, ‘How likely would you be to refer this patient to one of those professionals?’ (If physicians did not know of anyone who based his or her practice on Christian or Jewish teachings, the question item asked them to ‘assume there were’ providers nearby). Four response options ranged from not at all likely to very likely. For multivariable analyses, we dichotomized the variable to very likely/somewhat likely and not very likely/not at all likely. The question was asked after each vignette.

To test our hypotheses, we used respondents’ self-reports of their religious characteristics. Attendance at religious services was classified as never, once a month or less and twice a month or more. Religious affiliation was classified as follows: no affiliation, Hindu, Jewish, Muslim, Catholic (includes 27 Eastern Orthodox Christians), Evangelical Protestant, Non-Evangelical Protestant and other religion.

Statistical analysis

Stratum weights are described in more detail elsewhere (Lawrence, Rasinski, Yoon, Meador, et al., 2012; Lawrence, Rasinski et al., 2012; Lawrence, Rasinski, Yoon, Koenig, et al., 2013a). Briefly, the design weight accounted for the oversampling by ethnic surname; the post-stratification adjustment weight accounted for varying response rates between US and international medical school graduates and between physicians with different clinical roles; and the final weight was the product of the design weight and the post-stratification adjustment rate. Applying stratum weights enabled us to generate estimates for all US PCPs and psychiatrists.

We tested the effect of the experiment by constructing a saturated analysis, looking for main effects and interactions between the three experiment variables. Any variables or interactions contributing significant effects were retained in multivariable analyses examining our hypotheses about religious physicians differing from nonreligious physicians.

Results

Response

The response rate was 63% (896/1427) for PCPs and 64% (312/487) for psychiatrists, after excluding 77 PCPs and 25 psychiatrists who had invalid addresses or were no longer practicing. Details of the response rate and respondent demographics are described elsewhere (Lawrence, Rasinski, Yoon, Meador, et al., 2012; Lawrence, Rasinski, Yoon, Koenig, et al., 2012; Lawrence, Rasinski et al., 2013a). Overall responses (without regard to the experimental manipulations) are summarized in Table 1.

Table 1.

Overall results, not accounting for the vignette experiment design.

Variable	Primary care physicians		Psychiatrists
	Depression vignette	Anxiety vignette	Depression vignette	Anxiety vignette
	N (%)^a	N (%)^a	N (%)^a	N (%)^a
To the best of your knowledge, are there any mental health professionals in your area who base their practice on Christian (or Jewish)^b teaching?
Yes	195 (22.1)	227 (26.1)	100 (32.9)	119 (38.6)
No	684 (77.9)	648 (73.9)	204 (67.1)	189 (61.4)
(Assuming there were) How likely would you be to refer this patient to one of those professionals?
Very likely	206 (26.1)	191 (24.3)	45 (15.1)	35 (11.8)
Somewhat likely	297 (37.2)	311 (39.7)	112 (37.5)	110 (37.0)
Not very likely	219 (28.1)	225 (28.1)	103 (34.5)	109 (36.7)
Not at all likely	76 (8.6)	72 (7.9)	39 (13.0)	43 (14.5)

Percentages are adjusted for survey design and estimate the percentage of US primary care physicians and psychiatrists.

The religion of the mental health professional was matched to the patient’s religion.

Effect of experiment among PCPs

We constructed a saturated analysis, including whether the patient had problems, whether the patient and the hypothetical mental health provider were Christian or Jewish and whether the patient regularly or rarely attended services. PCPs were more likely to know of a Christian mental health provider than a Jewish provider (main effect). Specifically, following the depression vignette, 34.1% knew of a local mental health provider who practiced according to Christian teachings, but only 9.9% knew someone who practiced according to Jewish teachings (odds ratio (OR) = 2.2, 95% confidence interval (CI) = [1.8, 2.7]). Likewise, following the anxiety vignette, 44.1% knew of a local Christian provider, while 8.5% knew of a local Jewish provider (OR = 2.9, 95% CI = [2.4, 3.6]) (Table 2).

Table 2.

Results from the saturated analysis of the experimental variables, for each vignette. It shows main effects and interactions affecting physicians’ knowledge of religious providers, and whether they would refer patients there.

	Physician knows of a local mental provider who bases his or her practice on the same religion as the patient				Physician is (or would be) somewhat/very likely to refer this patient to a local mental health provider who bases his or her practice on the same religion as the patient
	Depression vignette		Anxiety vignette		Depression vignette		Anxiety vignette
	N (%)^a	OR, 95% CI^b	N (%)^a	OR, 95% CI^b	N (%)^a	OR, 95% CI^b	N (%)^a	OR, 95% CI^b
Primary care physicians^c
Patient reports no problems		Referent		Referent		Referent		Referent
Patient reports problems		NS^d		NS		NS		NS
Patient is Jewish	41 (9.9)	1.0 referent	39 (8.5)	1.0 referent	262 (68.5)	1.0 referent	229 (59.0)	1.0 referent
Patient is Christian	154 (34.1)	2.2, [1.8, 2.7]	188 (44.1)	2.9, [2.4, 3.6]	241 (58.5)	0.8, [0.7, 0.95]	273 (68.9)	1.2, [1.1, 1.5]
Patient rarely attends religious services		Referent		Referent	231 (58.5)	Referent		Referent
Patient regularly attends religious services		NS		NS	272 (68.0)	1.2, [1.04, 1.4]		NS
Problems × religion		NS		NS		NS		NS
Problems × attendance		NS		NS		NS		NS
Religion × attendance		NS		NS		NS		NS
Problems × religion × attendance		NS		NS		NS		NS
Psychiatrists^c
Patient reports no problems		Referent		Referent		Referent		Referent
Patient reports problems		NS		NS		NS		NS
Patient is Jewish	25 (15.8)	1.0 referent	29 (19.6)	1.0 referent		Referent		Referent
Patient is Christian	75 (51.4)	2.4, [1.8, 3.1]	90 (56.3)	2.3, [1.8, 3.0]		NS		NS
Patient rarely attends religious services		Referent		Referent	57 (40.4)	1.0 referent		Referent
Patient regularly attends religious services		NS		NS	100 (63.3)	1.6, [1.3, 2.0]		NS
Problems × religion		NS		NS		NS		NS
Problems × attendance		NS		NS		NS		NS
Religion × attendance		NS		NS		NS		NS
Problems × religion × attendance		NS		NS		NS		NS

OR: odds ratio; CI: confidence interval.

Percentages are adjusted for survey design and estimate the percent of US primary care physicians and psychiatrists.

Multivariable ORs were adjusted for all variables in the saturated analysis.

Primary care physicians and psychiatrists were analyzed in separate models.

NS means variable is not significantly different than the referent category.

The patient’s religious affiliation (which was the same as the hypothetical mental health provider’s affiliation) also had a main effect on whether PCPs would refer the patient. The direction of the effect varied between the depression vignette and the anxiety vignette. For the depression vignette, 68.5% of primary care providers would refer a Jewish patient to a Jewish provider, while 58.5% would refer a Christian patient to a Christian provider (OR = 0.8, 95% CI = [0.7, 0.95]). However, following the anxiety vignette, more primary care providers would refer a Christian patient to a Christian provider (68.9%, OR = 1.2, 95% CI = [1.1, 1.5]) than would refer a Jewish patient to a Jewish provider (59.0%) (Table 2).

There was also a main effect from how frequently the patient attends religious services, but only for the depression vignette. For a patient who rarely attends services, 58.5% of PCPs would refer the patient to a religious mental health provider. But for a patient who frequently attends services, 68.0% would refer the patient to a religious provider (OR = 1.2, 95% CI = [1.04, 1.4]) (Table 2).

Effect of experiment among psychiatrists

Psychiatrists also were more likely to know of Christian than Jewish mental health providers. Following the depression vignette, 51.4% of psychiatrists knew of a Christian mental health provider, while 15.8% knew of a Jewish provider (OR = 2.4, 95% CI = [1.8, 3.1]). Following the anxiety vignette, 56.3% knew of a Christian provider, and 19.6% knew of a Jewish provider (OR = 2.3, 95% CI = [1.8, 3.0]) (Table 2).

For psychiatrists, the only variable that affected referrals was how frequently the patient attends religious services, and this only had a main effect for the depression vignette. When a depressed patient rarely attends services, 40.4% of psychiatrists would refer to a religious mental health provider. But if a depressed patient regularly attends services, 63.3% of psychiatrists would refer to a religious provider (OR = 1.6, 95% CI = [1.3, 2.0]) (Table 2).

Effect of physician attendance at religious services

PCPs who frequently attend religious services were more likely to know of a religious mental health provider in the area and were more likely to refer the patient there. This pattern held after adjusting for the patient’s religious affiliation and how frequently the patient attends services. Compared with PCPs who never attend religious services, PCPs who attend twice a month or more were more likely to know of a religious mental health provider for a depressed patient (30.5% versus 12.9%, OR = 3.5, 95% CI = [1.9, 6.6]) and for an anxious patient (33.3% versus 8.0%, OR = 7.0, 95% CI = [3.3, 14.7]). In addition, relative to nonattenders, PCPs who frequently attend services were more likely to refer the depressed patient to a religious mental health provider (74.2% versus 36.7%, OR = 5.1, 95% CI = [3.1, 8.5]) and more likely to refer the anxious patient to a religious mental health provider (79.6% versus 38.2%, OR = 6.3, 95% CI = [3.7, 10.6]) (Table 3).

Table 3.

Variance of knowledge of local religious mental health providers, and willingness to refer, with how frequently the physician attends religious services.

	Physician knows of a local mental provider who bases his or her practice on the same religion as the patient				Physician is (or would be) somewhat/very likely to refer this patient to a local mental health provider who bases his or her practice on the same religion as the patient
	Depression vignette		Anxiety vignette		Depression vignette		Anxiety vignette
	N (%)^a	OR, 95% CI^b	N (%)^a	OR, 95% CI^b	N (%)^a	OR, 95% CI^b	N (%)^a	OR, 95% CI^b
Primary care physicians^c
Never attends services	15 (12.9)	1.0 referent	10 (8.0)	1.0 referent	37 (36.7)	1.0 referent	38 (38.2)	1.0 referent
Attends 1/month or less	70 (17.4)	1.5, [0.8, 2.9]	94 (24.5)	3.9, [1.8, 8.3]	221 (61.0)	2.6, [1.6, 4.1]	204 (57.2)	2.2, [1.3, 3.5]
Attends 2/month or more	105 (30.5)	3.5, [1.9-6.6]	115 (33.3)	7.0, [3.3, 14.7]	233 (74.2)	5.1, [3.1, 8.5]	247 (79.6)	6.3, [3.7, 10.6]
Psychiatrists^c
Never attends services		Referent		Referent		Referent	19 (38.0)	1.0 referent
Attends 1/month or less		NS^d		NS		NS	66 (42.6)	1.2, [0.6, 2.3]
Attends 2/month or more		NS		NS		NS	58 (65.2)	3.0, [1.4, 6.2]

OR: odds ratio; CI: confidence interval.

Percentages are adjusted for survey design and estimate the percentage of US primary care physicians and psychiatrists.

Multivariable ORs are adjusted for whether the patient is Christian or Jewish, and whether the patient regularly or rarely attends religious services.

Primary care physicians and psychiatrists were analyzed in separate models.

NS means variable is not significantly different than the referent category.

Psychiatrists’ knowledge of religious mental health providers in the local area was not significantly different between psychiatrists who frequently attend religious services, and those who never attend (after adjusting for the patient’s religious affiliation and frequency of attending services). However, compared with psychiatrists who never attend services, those who attend twice a month or more were more likely than nonattenders to refer an anxious patient to a religious mental health provider (65.2% versus 38.0%, OR = 3.0, 95% CI = [1.4, 6.2]) (Table 3).

Effect of physician religious affiliation

Christian PCPs were more likely than unaffiliated physicians to know of a religious mental health provider (after adjusting for the patient’s religious affiliation and the patient’s attendance at services). Following the depression vignette, 13.0% (referent) of unaffiliated physicians knew of a religious mental health provider, while 22.9% of Catholics (OR = 2.1, 95% CI = [1.01, 4.4]), 41.6% of Evangelical Protestants (OR = 5.8, 95% CI = [2.7, 12.5]) and 24.1% of Non-Evangelical Protestants (OR = 2.2, 95% CI = [1.1, 4.6]) knew of a religious provider. Likewise, following the anxiety vignette, 14.6% (referent) of unaffiliated physicians knew of a religious mental health provider, while 24.5% of Catholics (OR = 2.1, 95% CI = [1.04, 4.4]), 50.3% of Evangelical Protestants (OR = 9.3, 95% CI = [4.3, 20.2]) and 32.3% of Non-Evangelical Protestants (OR = 3.5, 95% CI = [1.8, 7.0]) knew of a religious provider (Table 4).

Table 4.

Variance of knowledge of local religious mental health providers, and willingness to refer, with the physicians’ religious affiliation.

	Physician knows of a local mental provider who bases his or her practice on the same religion as the patient				Physician is (or would be) somewhat/very likely to refer this patient to a local mental health provider who bases his or her practice on the same religion as the patient
	Depression vignette		Anxiety vignette		Depression vignette		Anxiety vignette
	N (%)^a	OR, 95% CI^b	N (%)^a	OR, 95% CI^b	N (%)^a	OR, 95% CI^b	N (%)^a	OR, 95% CI^b
Primary care physicians^c
No affiliation	13 (13.0)	1.0 referent	15 (14.6)	1.0 referent	33 (39.0)	1.0 referent	32 (38.0)	1.0 referent
Hindu	2 (3.9)	0.2, [0.1, 1.2]	7 (18.2)	1.5, [0.5, 4.5]	15 (40.1)	1.1, [0.5, 2.4]	12 (36.9)	1.0, [0.4, 2.4]
Jewish	20 (21.2)	1.7, [0.7, 4.0]	12 (14.3)	1.1, [0.4, 2.7]	48 (53.4)	1.8, [0.95, 3.3]	43 (50.6)	1.8, [0.9, 3.4]
Muslim	8 (16.4)	1.6, [0.5, 5.3]	6 (11.2)	0.6, [0.2, 2.0]	27 (55.3)	1.8, [0.8, 4.1]	24 (50.1)	1.5, [0.7, 3.5]
Catholic	47 (22.9)	2.1, [1.01, 4.4]	50 (24.5)	2.1, [1.04, 4.4]	116 (62.9)	2.8, [1.6, 4.9]	126 (68.5)	3.6, [2.0, 6.4]
Evangelical Protestant	38 (41.6)	5.8, [2.7, 12.5]	48 (50.3)	9.3, [4.3, 20.2]	82 (90.3)	15.6, [6.6, 36.6]	76 (91.7)	18.1, [7.4, 44.3]
Non-Evangelical Protestant	54 (24.1)	2.2, [1.1, 4.6]	73 (32.3)	3.5, [1.8, 7.0]	148 (74.2)	4.8, [2.7, 8.3]	153 (74.1)	4.7, [2.7, 8.4]
Other religion	8 (18.9)	1.3, [0.5, 3.9]	9 (22.0)	2.6, [0.8, 7.9]	24 (61.6)	2.8, [1.2, 6.4]	24 (62.5)	2.9, [1.2, 7.0]
Psychiatrists^c
No affiliation		Referent		Referent	18 (39.1)	1.0 referent	16 (34.8)	1.0 referent
Hindu		NS^§		NS	13 (54.2)	1.7, [0.6, 5.0]	13 (54.2)	2.2, [0.8, 6.1]
Jewish		NS		NS	13 (33.3)	0.8, [0.3, 2.1]	8 (22.2)	0.5, [0.2, 1.5]
Muslim		NS		NS	3 (42.9)	1.5, [0.3, 7.8]	5 (62.5)	3.1, [0.7, 14.7]
Catholic		NS		NS	43 (64.2)	2.7, [1.2, 5.9]	39 (59.1)	2.7, [1.2, 5.8]
Evangelical Protestant		NS		NS	12 (63.2)	2.3, [0.7, 7.1]	15 (83.3)	9.2, [2.3, 37.0]
Non-Evangelical Protestant		NS		NS	43 (60.6)	2.3, [1.03, 5.0]	36 (51.4)	1.9, [0.9, 4.2]
Other religion		NS		NS	10 (47.6)	1.5, [0.5, 4.4]	10 (41.7)	1.3, [0.5, 3.7]

OR: odds ratio; CI: confidence interval.

Percentages are adjusted for survey design and estimate the percentage of US primary care physicians and psychiatrists.

Multivariable ORs are adjusted for whether the patient is Christian or Jewish, and whether the patient regularly or rarely attends religious services.

Primary care physicians and psychiatrists were analyzed in separate models.

NS means variable is not significantly different than the referent category.

Christian PCPs were also more likely than unaffiliated physicians to refer patients to a religious mental health provider. Following the depression vignette, 39.0% (referent) of unaffiliated physicians were likely to refer, while 62.9% of Catholics (OR = 2.8, 95% CI = [1.6, 4.9]), 90.3% of Evangelical Protestants (OR = 15.6, 95% CI = [6.6, 36.6]) and 74.2% of Non-Evangelical Protestants (OR = 4.8, 95% CI = [2.7, 8.3]) would refer. Following the anxiety vignette, 38.0% (referent) of unaffiliated physicians were likely to refer to a religious provider, while 68.5% of Catholics (OR = 3.6, 95% CI = [2.0, 6.4]), 91.7% of Evangelical Protestants (OR = 18.1, 95% CI = [7.4, 44.3]) and 74.1% of Non-Evangelical Protestants (OR = 4.7, 95% CI = [2.7, 8.4]) would refer (Table 4).

Psychiatrists’ religious affiliation did not affect their knowledge of local religious mental health professionals. There was some indication that Christian psychiatrists were more likely than unaffiliated psychiatrists to refer to a religious provider. Following the depression vignette, 39.1% (referent) of unaffiliated psychiatrists would refer to a religious provider, while 64.2% of Catholics (OR = 2.7, 95% CI = [1.2, 5.9]), 60.6% of Non-Evangelical Protestants (OR = 2.3, 95% CI = [1.03, 5.0]) and 63.2% of Evangelical Protestants (not significant) would refer. Following the anxiety vignette, 34.8% (referent) of unaffiliated psychiatrists were likely to refer to a religious provider, while 59.1% of Catholics (OR = 2.7, 95% CI = [1.2, 5.8]), 83.3% of Evangelical Protestants (OR = 9.2, 95% CI = [2.3, 37.0]) and 51.4% of Non-Evangelical Protestants (not significant) would refer (Table 4).

Knowledge of religious providers as a predictor for referrals

In a post hoc analysis adjusting for the patient’s religiosity (religious affiliation and service attendance) and the physician’s religiosity (religious affiliation and frequency of attending services), we found that physicians who knew a religious provider in the area were more likely to refer patients there, compared with physicians who did not know any religious providers in the area.

For PCPs who did not know of a local religious provider, 59.6% would consider referring a depressed patient there, and 55.8% would consider referring an anxious patient there. However, among PCPs who knew of a local religious provider, 74.0% would refer a depressed patient (OR = 1.7, 95% CI = [1.1, 2.6]), and 84.1% would refer an anxious patient (OR = 3.1, 95% CI = [1.9, 5.1]).

For psychiatrists who did not know of a local religious provider, 48.7% would consider referring a depressed patient there, and 40.7% would consider referring an anxious patient. However, among psychiatrists who knew of a local religious provider, 61.0% would refer a depressed patient there (OR = 2.2, 95% CI = [1.2, 3.8]), and 61.0% would refer an anxious patient there (OR = 2.4, 95% CI = [1.3, 4.2]).

Discussion

In this national survey, we found that one-third of PCPs and half of psychiatrists know of local mental health providers who practice according to Christian teachings, while fewer than one in five know of local mental health providers who practice according to Jewish teachings. Approximately two-thirds of PCPs and half of psychiatrists were somewhat or very likely to refer a depressed or anxious religious patient to these religiously based practitioners; among PCPs in particular, those who were more religious were much more likely to make such referrals.

The proportion of physicians who were somewhat or very likely to refer a depressed or anxious patient to a religiously based practitioner was lower than the proportion who would refer a patient for faith-based alcohol treatment. Elsewhere in this survey, we provided a vignette of an alcohol-addicted patient and found that 79% of PCPs and 71% of psychiatrists were somewhat or very likely to refer to an explicitly faith-based alcohol treatment program (not including Alcoholics Anonymous) (Lawrence, Rasinski, Yoon, Koenig, et al., 2012). This provides some evidence that for treating mental health issues, physicians value religious resources differently depending on the diagnosis and the alternative treatment options.

Both the patient’s and the physician’s religious characteristics were associated with physicians’ knowledge of, and willingness to refer to, a religious mental health provider. The fact that physician religiosity frequently shapes referrals is not surprising. Past research (among psychologists) showed that religious therapists are more likely to utilize religious interventions such as prayer, religious language and religious/spiritual books, and to recommend participation in religion (Shafranske & Malony, 1990). A physician’s assessment of such practices is likely influenced by his or her religious background. However, these findings raise more complicated questions about the degree to which physicians are responsive to patients’ religious preferences, and the degree to which physicians allow their own personal biases (for or against religious treatments) to shape clinical recommendations.

Considering the findings more closely, we found support for our hypotheses among PCPs. Physicians who frequently attend religious services were more likely to know about religious mental health providers, and more likely to refer patients there. Christian affiliation was an especially strong predictor of knowing about and referring to religious providers. Perhaps religious mental health providers network with local religious organizations, so doctors who frequent those religious organizations are more familiar with religious providers than doctors who do not frequent religious organizations. Importantly, the data indicate a significant knowledge gap, with nonreligious physicians being significantly less likely to know about religious mental health providers in their area. Improving this knowledge could benefit patients who would prefer to receive treatment from a religious mental health provider.

Data from the psychiatrist sample only partially supported our hypotheses. There was no difference between religious and nonreligious psychiatrists’ knowledge of local religious mental health providers, but there was limited evidence that psychiatrists who frequently attend services and those who are Christian are more likely to refer a religious patient to a religious provider. We can only speculate about why these trends are weaker among psychiatrists than PCPs. Psychiatrists may have a strong professional interest in knowing the other mental health providers nearby, which outweighs any effect from religion. Psychiatrists may also be less likely in general to refer to another mental health provider, preferring to treat the patient themselves. Psychiatrists might have in mind treating with medications and/or evidence-based psychotherapies, which can be provided regardless of the practitioner’s religious orientation. The smaller sample size (relative to the PCPs) could also be a factor.

Christian PCPs were significantly more likely than unaffiliated physicians to know about and refer to a religious mental health provider. Evangelical Christians were especially likely. The high ORs are accompanied by somewhat broad CIs, which limit their interpretation, but they still provide compelling indicators that Evangelical Protestants have a particularly strong interest in religious-oriented mental health care. Some writers within this community have claimed that Christians should only seek treatment from other Christians who base their treatments on biblical principles (Adams, 1970; Lambert, 2012; Macarthur & Mack, 1994; Powlison, 2010). Our data may also reflect efforts by the Christian Counseling and Christian Psychology community to market themselves among Christian organizations to a greater degree than among minority religious groups. There is also a fear, within some Christian groups, that mainstream psychiatry and psychology are inherently antireligious (Farber, 1999; Macarthur & Mack, 1994), which could lead to a strong preference for religious mental health counseling (evidence suggests, however, that this fear is not widespread among physicians) (Lawrence, Rasinski et al., 2013a).

This study has limitations. The survey addressed religious mental health care in general, yet the training and theoretical orientation of practitioners within the field is diverse (McMinn, Staley, Webb, & Seegobin, 2010). Many physicians did not know of religious mental health providers, and our data cannot distinguish whether this is because physicians are unaware of local providers, or because no religious providers are located in the area. Another potential confounding factor is that physicians might know of a local religious provider but dislike that person’s clinical approach and consequently avoid referring patients there because of that provider’s idiosyncrasies, rather than bias against religious mental health counseling in general. (This concern is tempered by our finding that physicians who know of local religious providers were more likely to refer patients there.) We surveyed PCPs and psychiatrists, but recognize that opinions might differ in other medical specialties. Vignettes cannot fully reproduce clinical encounters, and it is unclear how far these findings extend to depression-in-general or anxiety-in-general. We measured beliefs about Christian and Jewish providers and cannot draw firm conclusions about providers from other religions. Some analyses used ‘no religious affiliation’ as the referent group, yet persons in this group are not necessarily uniform in their views on religion. Some might be antireligion, while others could be pro-religion but choose not to affiliate with a particular organized religion. Survey nonresponders might differ from responders in ways that bias the results.

In conclusion, we found that one-third of PCPs and half of psychiatrists know of local Christian mental health providers, and one in five knows of local Jewish providers. More than half of PCPs and psychiatrists would consider referring a religious patient to such providers. Religious physicians were significantly more likely than their less religious colleagues to make such referrals. Given the variability regarding how many physicians know of religious mental health providers, there is room to increase collaboration, especially between PCPs and religious mental health providers. It is positive for patients that so many physicians and psychiatrists would consider referring to a religious mental health professional, since this allows patients greater opportunities to seek care consistent with their values. Yet, this openness also highlights the need for further research to discern which interventions are helpful to which patient groups.

Footnotes

Funding

This study was funded by the John Templeton Foundation and the National Center for Complementary and Alternative Medicine (1 K23 AT002749 to Farr Curlin).

References

Adams

J. E.

(1970). Competent to counsel: Introduction to nouthetic counseling. Grand Rapids, MI: Zondervan.

American Association of Christian Counselors. (n.d.). Available from http://www.aacc.net/

American Psychological Association. (n.d.). Retrieved from http://www.apa.org/ed/accreditation/programs/index.aspx

Association of Theological Schools. (n.d.). Retrieved from http://www.ats.edu/MemberSchools/Pages/DegreeSchools.aspx?ik=8

Azusa Pacific University. (n.d.). Retrieved from http://www.apu.edu/bas/graduatepsychology/psyd/

Belaire

C. J.

Young

J. S.

Elder

(2005). Inclusion of religious behaviors and attitudes in counseling: Expectations of conservative Christians. Counseling and Values, 49, 82–94.

Bergin

A. E.

Jensen

J. P.

(1990). Religiosity of psychotherapists: A national survey. Psychotherapy, 27(1), 3–7.

Biola University Rosemead School of Psychology. (n.d.). Retrieved from http://www.rosemead.edu/about/

Council for Christian Colleges and Universities. (n.d.). Retrieved from http://www.cccu.org/members_and_affiliates

10.

Curlin

F. A.

Lawrence

R. E.

Odell

Chin

M. H.

Lantos

J. D.

Koenig

H. G.

Meador

K. G.

(2007). Religion, spirituality, and medicine: Psychiatrists’ and other physicians’ differing observations, interpretations, and clinical approaches. American Journal of Psychiatry, 164, 1825–1831.

11.

Curlin

F. A.

Odell

S. V.

Chin

M. H.

Lantos

J. D.

Meador

K. G.

Koenig

H. G.

(2007). The relationship between psychiatry and religion among U.S. physicians. Psychiatric Services, 58, 1193–1198.

12.

Domino

(1990). Clergy’s knowledge of psychopathology. Journal of Psychology & Theology, 18, 32–39.

13.

Farber

(1999). Unholy madness: The church’s surrender to psychiatry. Downers Grove, IL: InterVarsity Press.

14.

Franzblau

A.N.

D’Agostino

Draper

Egan

M.H.

Grosch

W.N.

Honig

E.M.

Rizzuto

A.M.

(1975). Psychiatrists’ viewpoints on religion and their services to religious institutions and the ministry (Task Force Report No. 10). Washington, DC: American Psychiatric Association Task Force on Religion and Psychiatry.

15.

Freud

(1964). Moses and Monotheism: Standard edition of the complete psychological works of Sigmund Freud. London, England: Hogarth Press, p. 23.

16.

Genia

(1994). Secular psychotherapists and religious clients: Professional considerations and recommendations. Journal of Counseling & Development, 72, 395–398.

17.

George Fox University. (n.d.). Retrieved from http://www.georgefox.edu/psyd/index.html

18.

Kelly

E. W.

Aridi

(1996). Muslims in the United States: An exploratory study of universal and mental health values. Counseling and Values, 40, 206–218.

19.

Koenig

H. G.

(2009). Research on religion, spirituality, and mental health: A review. Canadian Journal of Psychiatry, 54, 283–291.

20.

Koenig

H. G.

Larson

D. B.

Weaver

A. J.

(1998). Research on religion and serious mental illness. New Directions for Mental Health Services, 80, 81–95.

21.

Lambert

(2012). The biblical counseling movement after Adams. Wheaton, IL, Crossway.

22.

Lauderdale

D. S.

(2006). Birth outcomes for Arabic-named women in California before and after September 11. Demography, 43, 185–201.

23.

Lauderdale

D. S.

Kestenbaum

(2000). Asian American ethnic identification by surname. Population Research and Policy Review, 19, 283–300.

24.

Lawrence

R. E.

Rasinski

K. A.

Yoon

J. D.

Curlin

F. A.

(2013a). Religion and anxiety treatments in primary care patients. Anxiety, Stress, & Coping, 26(5), 526–538.

25.

Lawrence

R. E.

Rasinski

K. A.

Yoon

J. D.

Curlin

F. A.

(2013b). Religion and beliefs about treating medically unexplained symptoms: A survey of primary care physicians and psychiatrists. International Journal of Psychiatry in Medicine, 45(1), 31–44.

26.

Lawrence

R. E.

Rasinski

K. A.

Yoon

J. D.

Koenig

H. G.

Meador

K. G.

Curlin

F. A.

(2012). Physicians’ beliefs about faith-based treatments for alcoholism. Psychiatric Services, 63, 597–604.

27.

Lawrence

R. E.

Rasinski

K. A.

Yoon

J. D.

Meador

K. G.

Koenig

H. G.

Curlin

F. A.

(2012). Primary care physicians’ and psychiatrists’ approaches to treating mild depression. Acta Psychiatrica Scandinavica, 126(5), 385–392.

28.

Leavey

Loewenthal

King

(2007). Challenges to sanctuary: The clergy as a resource for mental health care in the community. Social Science & Medicine, 65, 548–559.

29.

Macarthur

J. F.

Mack

W. A.

(1994). Introduction to biblical counseling. Nashville, TN: W Publishing Group.

30.

Marx

J. H.

Spray

S. L.

(1969). Religious biographies and professional characteristics of psychotherapists. Journal of Health and Social Behavior, 10, 275–288.

31.

McMinn

M. R.

Staley

R. C.

Webb

K. C.

Seegobin

(2010). Just what is Christian counseling anyway? Professional Psychology: Research and Practice, 41, 391–397.

32.

Mitchell

J. R.

Baker

M. C.

(2000). Religious commitment and the construal of sources of help for emotional problems. British Journal of Medical Psychology, 73, 289–301.

33.

Powlison

(2010). A biblical counseling view. In Johnson

E. L.

(Ed.), Psychology & Christianity: Five views (2nd ed., pp. 245–273) Downers Grove, IL: InterVarsity Press.

34.

Regent University. (n.d.). Retrieved from http://www.regent.edu/acad/schcou/aboutus/index.htm

35.

Shafranske

E. P.

Malony

H. N.

(1990). Clinical psychologists’ religious and spiritual orientations and their practice of psychotherapy. Psychotherapy: Theory, Research, Practice, Training, 27(1), 72–78.

36.

Sheskin

I. M.

(1998). A methodology for examining the changing size and spatial distribution of a Jewish population: A Miami case study. Shofar: An Interdisciplinary Journal of Jewish Studies, 17, 97–114.

37.

Walker

D. F.

Worthington

E. L.

Gartner

A. L.

Gorsuch

R. L.

Hanshew

E. R.

(2011). Religious commitment and expectations about psychotherapy among Christian clients. Psychology of Religion and Spirituality, 3, 98–114.

38.

Wheaton College. (n.d.). Retrieved from http://www.wheaton.edu/Academics/Departments/Psychology/Graduate-Programs/Programs/PsyD-in-Clinical-Psychology

39.

Young

Cashwell

Wiggins-Frame

Belaire

(2002). Spiritual and religious competencies: A national survey of CACREP-accredited programs. Counseling and Values, 47, 22–33.