Prevalence and Sociodemographic Correlates of Spiritual Healer Use: Findings from the National Survey of American Life

Summary

Objectives

This study investigates sociodemographic and health-related correlates of use of a spiritual healer for medical help. A large national, multiracial-multiethnic data source permits a more comprehensive investigation than was possible in previous studies. It also enables a closer focus on socioeconomic disadvantage and health need as determinants of utilization.

Design and setting

Respondents are from the National Survey of American Life: Coping with Stress in the 21^st Century (NSAL), a nationally representative multi-stage area-probability survey of U.S. adult African Americans, Caribbean Blacks, and non-Hispanic Whites conducted from 2001-2003. The sample contains 6,082 adults aged 18 and over.

Main outcome measures

NSAL respondents were surveyed about lifetime use of alternative providers for medical care or advice. Response categories included two types of spiritual healers: faith healers and psychics. These outcomes were logistically regressed, separately, onto 10 sociodemographic or health-related indicators: race/ethnicity, age, gender, marital status, education, household income, region, medical care use, insurance coverage, and self-rated health.

Results

Lifetime utilization of a faith healer is more prevalent among respondents in good health and less prevalent among Caribbean Blacks and never married persons. Users of a psychic healer are more likely to be educated, residents of the Northeast or West, and previously married, and less likely to report excellent health.

Conclusions

Use a spiritual healer is not due, on average, to poor education, marginal racial/ethnic or socioeconomic status, dire health straits, or lack of other healthcare options. To some extent, the opposite appears to be true. Use of a spiritual healer is not associated with fewer social and personal resources or limitations in health or healthcare.

Introduction

In utilization studies of complementary and alternative medicine (CAM), lifetime or past-year rates of use are typically reported for a list of treatment modalities. One class of modalities above all tends either to be excluded altogether or, if reported, not followed up in depth. Use of spiritual healing, of various types, has been relegated to the margins of health services research on CAM since the earliest days of the field, perhaps because of an understandable desire to dissociate from nonmedical therapies, nonprofessional providers, and treatments without scientifically validated mechanisms of action. Existing data, however, sparse as they are, reveal that the frequency of use of a spiritual healer or reliance on the prayers of others may exceed almost every other CAM therapy.¹ This finding has been validated across racial and ethnic groups in the U.S.^2-5 Further, as with CAM generally the distribution and correlates of this finding “among racial/ethnic groups is complex and nuanced”³ (p. 1236).

The earliest example of note is work by Eisenberg and associates investigating past-year utilization of a list of complementary medical therapies including spiritual healing and prayer.⁶ While 4% of the U.S. population used spiritual healing, 9% reported a visit to a spiritual healer, and users averaged 14 sessions. Further, 25% of the population reported using healing prayer. A followup study⁷ found that spiritual healing by others was now reported by 7.0% of the population and self-prayer for healing by 35.1%, more than twice the rate of any other therapy. Neither modality was subjected to additional analysis. A subsequent study focused on mood disorders, finding that 9.9% of respondents with anxiety attacks and 10.5% of those reporting severe depression utilized spiritual healing by others in the past year; reported rates for energy healing for these conditions were 2.8% and 5.4%, respectively.⁸ Other research has placed the national prevalence of past-year use of “spiritual healing or prayer” at 13.7% and of energy healing at 1.1%.⁹

Research on spiritual factors in personal and population health has become more widely accepted in other medical and sociomedical fields, such as epidemiology, health psychology, and gerontology and geriatrics.¹⁰ In addition, there is a rapidly expanding literature on ethnographic, sociological, clinical, and experimental studies of healers.¹¹ Although not widely publicized, studies of religious correlates of healthcare utilization and inclusion of spiritual variables within health services research studies actually goes back several decades.¹² For spiritual healing, however, we are still lacking very fundamental information as to patterns of use.

Research by anthropologists, sociologists, religious scholars, and others has provided both descriptive and interpretive studies, case histories, and results of small-scale surveys focused on spiritual healing. Most notable have been a few ethnographic or observational studies conducted decades ago^13-17 that provide interesting social-historical analysis, as well as a spate of more recent experimental studies summarized by Benor.¹⁸ But beyond these good descriptions of these phenomena, there is a lack of systematic empirical investigation.

Other CAM modalities (e.g., massage therapy, relaxation techniques, meditation) have gained acceptance within medical and health services research. This is largely because they have provided naturalistic models of explanation for efficacy that are congruent with the physiological, psychophysiological, and pathophysiological theories of mainstream medical discourse.^19-20 The same is true for spiritual or religious therapies such as religious support, pastoral care, chaplaincy visits, and even personal prayer, for which efficacy can be understood in terms of associated psychosocial pathways, mechanisms, and outcomes (e.g, social support, self-esteem, locus of control).²¹

Spiritual healers, in contrast, operate according to explanatory models that are outside the mainstream of naturalistic theories of healing. Whether self-described as faith healers or psychic healers, this class of modalities posits mechanisms that involve superempirical or ostensibly “supernatural” forces (e.g., Holy Spirit, paranormal forces or energies) that are beyond the scope of observational or experimental science. Whether such explanatory models are indeed “real” is a question outside of the present authors' purview and is not suitable subject matter for a medical journal. However, regardless of how unusual or controversial the content of these modalities may appear to mainstream biomedicine, the matter of their utilization is a topic that medicine can and should address. Simply put, clinicians and health services researchers would benefit from identifying who utilizes spiritual healers and the circumstances of their use.

Although there is rudimentary information on patterns of use for spiritual help-seeking, we still lack systematic information on correlates or predictors of use. People who most utilize faith healers or psychics are thought to be, on average, poor, uneducated, Southerners, ethnic minorities, in desperate health straits, and lacking access to medical care and other supportive resources (e.g., family). This is supported by results from a few small-scale, regional studies.^22-24 However, larger-scale empirical investigations of sociodemographic or health-related correlates of the use of spiritual healers or healing have resulted in inconsistent findings^8,25-26 that appear to depend upon the time referent of the utilization question (e.g., lifetime prevalence, past-year prevalence), the specific type of spiritual healing assessed (e.g., use of a spiritual healer, an energy healer, prayer), and study sample characteristics. Moreover, analyses of spiritual healer use within particular sociodemographic subgroups are sparse.^27-29

A different line of thinking suggests that help-seeking directed to faith healers or psychic healers is a part of a larger quest for spirituality and wholeness. Seen in this way, spiritual and psychic healers are utilized out of faithfulness to a particular worldview (e.g., religious) and/or in pursuit of growth and self-actualization, as much as for remission of acute symptoms. For some people, use of healers may be a marker of commitment to transcendental values and not specifically a response that is prompted by an acute health need or conditioned by sociodemographic status. For segments of the population, physical well-being and spirituality are parts of an inseparable whole—pursuit of one goes hand in hand with pursuit of the other. If so, then greater use of spiritual healers may be more prevalent among the opposite sociodemographic categories: wealthier, more educated, married, Northerners, non-minorities, in good health, with access to medical care. The present analysis seeks to address this question by focusing on sociodemographic and health-related correlates of spiritual healer use.

The National Survey of American Life: Coping with Stress in the 21^st Century (NSAL) provides an ideal setting to explore this issue. The NSAL contains a large nationally representative sample, is multiracial and multiethnic, and includes variables assessing use of spiritual healers, as well as items on health status, healthcare use, and some of the sociodemographic characteristics identified above as potential correlates of the use of spiritual healers. A special advantage of using the NSAL to explore this subject is that it builds on prior research using this sample which has documented mental healthcare utilization,³⁰ CAM use,⁵ and patterns and correlates of spirituality in this population.³¹ The present investigation also builds on recent studies of racial and ethnic differences in spiritual help-seeking behavior^32-33 and descriptions of the work of spiritual healers.^34-35

Methods

Sample

The NSAL was collected by the Program for Research on Black Americans (PRBA) at the University of Michigan's Institute for Social Research (ISR).³⁶ The field work for the study was completed by ISR's Survey Research Center, in cooperation with PRBA. A total of 6,082 face-to-face interviews were conducted with persons aged 18 or older, including 3,570 U.S. African Americans, 891 non-Hispanic Whites, and 1,621 Blacks of Caribbean descent. The overall response rate of 72.3% is excellent given that African Americans (especially lower income African Americans) and Caribbean Blacks are more likely to reside in major urban areas which are more difficult and expensive with respect to survey fieldwork and data collection. Final response rates for the NSAL two-phase sample designs were computed using the American Association of Public Opinion Research (AAPOR) guidelines (for Response Rate 3).³⁷

The African-American sample is the core sample of the NSAL and consists of 64 primary sampling units (PSUs). Fifty-six of these primary areas overlap substantially with existing Survey Research Center National Sample primary areas. The remaining eight primary areas were chosen from the South in order for the sample to represent African Americans in the proportion in which they are distributed nationally. The African-American sample is a nationally representative sample of households located in the 48 coterminous states with at least one Black adult 18 years of age or over who did not identify ancestral ties in the Caribbean. Both the African-American and non-Hispanic White samples were selected exclusively from these targeted geographic segments in proportion to the African-American population.

The sample design and analysis weights for this sample were designed to be optimal for comparative analyses in which residential, environmental, and socioeconomic characteristics are controlled in the Black-White statistical contrasts. For all three racial/ethnic samples, the NSAL weights were designed to correct for disproportionate sampling and non-response, and to provide representation across various demographic characteristics in the 48 coterminous states.

The Black Caribbean sample was selected from two area-probability sampling frames: the core NSAL sample, and an area-probability sample of housing units from geographic areas with a relatively high density of persons of Caribbean descent (more than 10% of the population). Of the total Black Caribbean respondents (N = 1,621), 265 were selected from the households in the core sample, while 1,356 were selected from housing units from high-density Caribbean areas (see Heeringa and associates³⁸ for a more detailed description of the sample designs and sampling methods used in the development of the NSAL).

In both the African-American and Black Caribbean samples, it was necessary for respondents to self-identify their race as Black. Those self-identifying as Black were included in the Black Caribbean sample if (1) they answered affirmatively when asked if they were of West Indian or Caribbean descent, (2) they said they were from a country included on a list of Caribbean area countries presented by the interviewer, or (3) they indicated that their parents or grandparents were born in a Caribbean area country (see Jackson and associates³⁹ for a more detailed discussion of the NSAL sample). The interviews were administered face-to-face and conducted within respondents' homes and respondents were compensated for their time. Data collection was conducted from February, 2001, to June, 2003.

Measures

Dependent Variables

Respondents were asked, “Here is a list of people that one might go to for medical help. Please indicate if you have ever gone to any of the following people?” (coded: yes vs. no). They were then presented with a list of alternative providers which included two variants of spiritual healer: (a) “a faith healer” and (b) “a person who practices astrology, reads zodiac signs, or is a psychic” (for simplicity, referred to from now on as a psychic or psychic healer). The question on alternative providers was asked immediately following an item assessing traditional medical care utilization (e.g., doctor, clinic, health center).

Independent Variables

Sociodemographic characteristics include race/ethnicity (African American, Caribbean Black, non-Hispanic White), age (in years), gender, marital status (currently married, cohabit, previously married, never married) and U.S. region (Northeast, North Central, South, West). Socioeconomic status is measured by education and household income. Missing data for household income were imputed for 773 cases (12.7% of the NSAL sample). Missing data for education were imputed for 74 cases. Imputations were done using an iterative regression-based multiple imputation approach incorporating information about age, sex, region, race, employment status, marital status, home ownership, and nativity of household residents. For the bivariate analysis only, education was recoded into four categories (less than high school degree, high school degree, some college, college degree or higher) and income was recoded into tertiles. Use of traditional medical care was assessed by the question, “Is there one place or person you usually go to when you are sick or need medical advice?” Responses to this question included: doctor, clinic, health center, hospital, relative, other, and none. Self-rated health was measured by the question, “How would you rate your overall physical health at the present time? Would you say it is excellent, very good, good, fair or poor?” There was also a dichotomous measure of whether or not the respondent has health insurance.

Data Analysis

Both bivariate and multivariable analyses were conducted. First, patterns of spiritual healer use were stratified by categories of other study variables, separately for faith healers and psychics. Percentages represent weighted proportions based on the distribution of the population. The Rao-Scott χ² represents a complex design-corrected measure of association. Second, multivariable logistic regression was used, regressing use of both types of spiritual healer separately onto all of the independent variables. This enables identification of net effects of respective correlates, adjusting for effects of all other variables in the model. For each of the two logistic regression models (faith healer and psychic healer), we report logistic coefficients (β's), standard errors (se's), and the design-corrected F statistic, as well as odds ratio estimates (OR's) and 95% confidence intervals (CI's). Logistic coefficients and standard errors take into account the complex multistage clustered design of the NSAL sample, unequal probabilities of selection, nonresponse, and poststratification. All analyses were conducted using SAS 9.13, which uses the Taylor expansion approximation technique for calculating the complex design-based estimates of variance.

Results

Among respondents, lifetime prevalence of faith healer use is 3.7%, and 6.1% of respondents reported ever using a psychic for purposes of healing. According to Table 1, gender and self-rated health were significantly associated with the use of a faith healer. Women were more likely than men to use a faith healer for medical care, as were respondents who rated their health as good. Table 1 also presents the bivariate analysis of using a psychic for medical care. Caribbean Blacks and non-Hispanic whites were more likely to use a psychic healer than African Americans. Women were more likely than men and respondents with more years of education were more likely than their counterparts to use a psychic healer. Among the marital status groups, marrieds were the least likely, whereas previously marrieds were the most likely to use a psychic healer, as were respondents who resided in the Northeast and West. Lastly, respondents who usually received their medical care in a hospital were the least likely to use a psychic healer.

Table 1.

Lifetime use of a spiritual healer, by sociodemographic and health-related variables.

Sociodemographic and Health-Related Variables	Faith Healera		Psychic Healerb
	Nc	%c	n	%
Race/Ethnicity
African American	152	4.31	151	4.30
Caribbean Black	71	2.10	117	7.97
Non-Hispanic White	31	3.24	63	7.70
χ2 (df = 2) (p-value)	4.48 (0.10)		8.08 (0.01)
Gender
Male	72	2.76	90	5.02
Female	182	4.49	241	7.07
χ2 (df = 1) (p-value)	12.30 (0.04)		7.27 (0.007)
Age
18-34	76	3.27	128	5.69
35-54	128	4.43	148	7.30
55+	50	2.98	55	4.72
χ2 (df = 2) (p-value)	4.34 (0.11)		4.46 (0.10)
Education
Less than high school	50	3.84	45	3.20
High school degree	77	3.48	97	4.91
Some college	80	4.96	102	7.85
College degree or higher	47	2.50	87	8.71
χ2 (df = 3) (p-value)	3.99 (0.26)		11.20 (0.01)
Household Income
Low	97	4.42	109	6.37
Medium	81	4.15	110	5.55
High	76	2.93	112	6.37
χ2 (df = 2) (p-value)	3.15 (0.20)		0.58 (0.74)
Marital Status
Married	87	3.61	70	4.28
Cohabiting	14	2.81	30	7.81
Previously married	85	5.36	112	8.74
Never married	68	2.51	119	6.01
χ2 (df = 3) (p-value)	6.35 (0.09)		10.87 (0.012)
Region
Northeast	63	3.31	125	9.17
North Central	31	4.93	34	5.24
South	150	3.63	136	3.92
West	10	3.37	303	11.72
χ2 (df = 3) (p-value)	1.86 (0.60)		14.77 (0.002)
Medical care use
Doctor	143	4.11	176	6.46
Clinic	32	2.37	50	4.07
Health center	11	4.43	25	5.59
Hospital	27	2.62	25	3.02
Relative	1	1.20	2	3.07
Other	3	13.80	5	31.85
None	37	3.25	48	7.52
χ2 (df = 6) (p-value)	8.06 (0.23)		24.32 (0.001)
Insurance Coverage
No	54	4.00	56	7.24
Yes	200	3.64	275	5.93
χ2 (df = 1) (p-value)	0.14 (0.70)		1.37 (0.24)
Self-rated Health
Poor	11	2.25	23	7.73
Fair	47	3.93	57	5.33
Good	79	5.16	104	7.15
Very good	73	2.75	97	6.18
Excellent	44	3.27	50	4.22
χ2 (df = 4) (p-value)	9.43 (0.05)		3.90 (0.41)

^aN = 5,885 to 5,888.

^bN = 5,892 to 5,895.

^cData are presented as unweighted frequencies and weighted percentages.

Additional analyses (not reported in Table 1) replicated the above breakouts, this time comparing respondents who used a faith healer only, a psychic only, both, and neither. As only a bit more than 1 percent of the total sample available (71 out of 5,895 respondents) reported ever having used both types of spiritual healers, there were not any substantive differences between the results presented in Table 1 for use of each of these modalities and use “only” of either of these modalities. The “both” group, moreover, was too small to produce stable estimates for this tiny portion of the sample. These additional analyses did validate one substantively interesting observation: that so few respondents reported using both faith healers and psychics provides some evidence against the stereotype that people who seek alternative care are indiscriminate in their use of anything and everything. These two modalities are utilized by largely different people and have somewhat distinct patterns of use.

Results for the weighted multivariable logistic regression analysis are presented in Table 2. Race/ethnicity, marital status, and self-rated health were significantly associated with use of a faith healer. Caribbean Blacks were less likely to use a faith healer than African Americans (β = -0.64, p = .003; OR = 0.52), and never marrieds were less likely to use a faith healer than currently marrieds (β = -0.73, p = .023; OR = 0.47). Respondents who indicated that their health was good were more likely than those with poor health to use a faith healer (β = 0.95, p = .02; OR = 2.61).

Table 2.

Logistic regressions of lifetime use of a spiritual healer on sociodemographic and health-related variables.

Sociodemographic and Health-Related Variables	Faith Healer				Psychic Healer
	βa	se	p-value	OR (95% CI)	βa	se	p-value	OR (95% CI)
Race/Ethnicity
African American				1.0				1.0
Caribbean Black	-0.64	0.22	.003	0.52 (0.34 - 0.81)	.32	.27	.23	1.38 (0.80 - 2.38)
Non-Hispanic White -0.24		0.26	.351	0.78 (0.46 - 1.31)	.39	.24	.10	1.47 (0.91 - 2.38)
Age	-0.01	0.007	.06	0.98 (0.97 - 1.00)	-0.00	0.00	.14	0.99 (0.98 - 1.00)
Gender
Female				1.0				1.0
Male	-0.36	0.24	.13	0.69 (0.43 - 1.12)	-0.24	0.13	0.057	0.78 (0.60 - 1.00)
Marital status
Currently married				1.0				1.0
Never married	-0.73	0.32	.023	0.47 (0.25 - 0.90)	0.07	0.31	.81	1.07 (0.58 - 2.00)
Previously married 0.30		0.25	.23	1.36 (0.82 - 2.25)	0.73	0.31	0.019	2.09 (1.12 - 3.88)
Cohabiting	-0.53	0.55	.33	0.58 (0.19 - 1.72)	0.39	0.45	.37	1.49 (0.61 - 3.62)
Education	-0.07	0.06	.26	0.92 (0.81 - 1.06)	0.12	0.05	0.017	1.13 (1.02 - 1.25)
Household Income	-0.00	0.00	.55	1.00 (1.00 - 1.00)	-0.00	0.00	.99	1.0 (1.00 - 1.00)
Region
South				1.0				1.0
Northeast	0.12	0.24	.59	1.13 (0.70 - 1.82)	0.88	0.37	0.017	2.43 (1.17 - 5.04)
North Central	0.38	0.33	.25	1.46 (0.76 - 2.83)	0.37	0.43	.38	1.45 (0.62 - 3.39)
West	0.07	0.29	0.79	1.08 (0.60 - 1.93)	1.046	0.44	.02	2.84 (1.17 - 6.87)
Medical care use
None				1.0				1.0
Doctor	0.22	0.35	.51	1.25 (0.63 - 2.49)	-0.02	0.35	.94	0.97 (0.49 - 1.94)
Clinic	-0.45	0.40	.25	0.63 (0.28 - 1.39)	-0.63	0.37	.08	0.52 (0.25 - 1.09)
Health center	0.19	0.66	.76	1.22 (0.33 - 4.53)	-0.23	0.36	.51	0.78 (0.38 - 1.62)
Hospital	-0.25	0.44	.57	0.77 (0.32 - 1.85)	-0.87	0.49	.07	0.41 (0.15 - 1.10)
Relative	-0.87	1.06	.40	0.41 (0.05 - 3.32)	-1.03	1.08	.33	0.35 (0.04 - 2.98)
Other	1.83	0.98	.06	6.27 (0.91 - 43.03)	1.85	0.66	.005	6.39 (1.73 - 23.58)
Insurance coverage
No				1.0				1.0
Yes	-0.00	0.31	.97	0.99 (0.53 - 1.83)	-0.36	0.18	.056	0.69 (0.48 - 1.01)
Self-rated health
Poor				1.0				1.0
Fair	0.59	0.44	.18	1.81 (0.75 - 4.37)	-0.57	0.40	.15	0.56 (0.25 - 1.23)
Good	0.95	0.41	.02	2.61 (1.16 - 5.83)	-0.35	0.41	.38	0.70 (0.31 - 1.56)
Very good	0.35	0.45	.44	1.42 (0.58 - 3.51)	-0.64	0.36	.07	0.52 (0.25 - 1.06)
Excellent	0.55	0.50	.26	1.74 (0.65 - 4.68)	-1.02	0.50	.04	0.35 (0.13 - 0.96)
χ2 lr (df)	87.23 (23)				233.30 (23)
p-value	<.0001				<.0001
N	5,883				5,890

^aβ = logistic coefficient.

Marital status, education, region, and self-rated health were significantly associated with seeking assistance from a psychic for medical care. Previously married respondents were more likely to use a psychic healer (β = 0.73, p = .019; OR = 2.09) than married respondents. Education was positively associated with psychic healer use (β = 0.12, p = .017; OR = 1.13). Respondents who resided in the Northeast (β = 0.88, p = .017; OR = 2.43) or in the West (β = 1.046, p = .02; OR = 2.84) were more likely to use a psychic healer than Southerners. Lastly, respondents who rated their health as excellent (β = -1.02, p = .04; OR = 0.35) were less likely to seek medical care from a psychic than those with poor self-rated health. There was also a statistically significant finding indicating greater use of a psychic healer among people usually receiving medical care from some “other” source of care (β = 1.85, p = .005; OR = 6.39), but the absolute frequency was so small (only 5 respondents), and the respective odds ratio's confidence interval so wide (1.73-23.58), that this result may be spurious.

Discussion

Overall, a relatively small proportion of people in this sample have ever sought medical care from a spiritual healer. Contrary to common assumptions, use of such healers is not primarily conditioned by sociodemographic disadvantage, serious health need, or lack of access to mainstream medical care or insurance coverage. These findings indicate that use of a spiritual healer, whether a faith healer or a psychic, is supplementary to traditional care-seeking behavior. Spiritual healing, whatever its merits or faults, is thus an example of care that is both alternative and complementary.

These findings indicate that people who have utilized a spiritual healer are not characterized by lower education, marginal racial/ethnic or socioeconomic status, poor health, or lack of other care options. Efforts to understand this class of healthcare behavior based solely on these sorts of factors are inadequate and do not explain why people utilize such care. In the present study, these variables were either not significant correlates of spiritual healer use or were significantly predictive, but mostly in the opposite direction. Use of a faith healer was associated with reporting good health; and never married persons and Caribbean Blacks were less likely to use a faith healer. Use of a psychic healer was associated with higher education, being previously married, and residence in the Northeast or West. While persons in excellent health were less likely to use a psychic healer, at the net level, there were no significant differences in use across the other response categories for health status.

A few limitations should be noted. First, it would have been helpful to examine whether respondents report self-prayer for healing and if this practice is distinct from use of a spiritual healer. Second, item wording in the NSAL combines use of a psychic for healing with use of an astrologer. Third, the data do not specify why respondents utilized a spiritual healer or when in the process of care-seeking this occurred. Finally, the focus of these particular analyses was solely on sociodemographic and health-related correlates; questions regarding psychosocial or religious determinants of spiritual healer use remain unexplored. Together, these limitations leave much of the “what” and “why” of spiritual healer use unaddressed. These limitations, however, are balanced by the opportunity to begin exploring this topic in a racial/ethnic-comparative context in national probability data with substantial representation of typically undersurveyed respondent populations.

An obvious next step for this subject would be to identify other correlates and predictors of spiritual healer use. Prior research on CAM has identified psychological characteristics including personality traits, primary and secondary control strategies, and perceived social support and strain⁴⁰ as potential determinants of use. Further, religious/spiritual preferences and ideations including particular values, beliefs, and philosophical orientations such as a holistic orientation and personal belief in the power of religion⁴¹ and in “a natural approach to treatment” and the efficacy of “spirituality, religion, or prayer for health reasons”⁴² (pp. 192-193) are potential correlates of spiritual healer use.

Finally, one of the most important issues to address in this literature is conceptual and involves the confounding of the various constructs (e.g., prayer, healing, spirituality, faith healing) that are used in research studies.⁴³ While these are all valid topics for investigation, we should be careful that when we speak of the use of a faith healer, for example, we mean a faith healer and not a bioenergy practitioner, a qi gong master, or private prayer. An insightful discussion of this issue made the important point that “use of the broad term ‘prayer’ in attempts to measure CAM use fails to distinguish between the diverse forms of spiritual healing utilized by practitioners and the common understanding of the word”⁴⁴ (p. 436). The present study did not reference prayer directly and we were able to distinguish between faith healers and healers professing psychic gifts (at least presumably, given the implicit limitations of the wording and binary response format of the survey items). But, still, this leaves a lot of diversity in extant practice traditions and disciplines (e.g., East Asian, Western professional, bioenergy, contemporary metaphysical healing traditions) unexplored.³⁵ For intentional healing, more generally, there is a tremendous diversity of lay, folk, and professional approaches, each presenting unique and sophisticated methodological challenges for researchers.⁴⁵ While health services researchers do not directly involve themselves in the kinds of theoretical issues that make this such a controversial area (e.g., purported mechanisms of action), nonetheless careful attention to conceptual distinctions among spiritual healing modalities is a requirement for future studies. To further our understanding of who uses what types of spiritual therapies, and why they do so, will require a continued systematic approach to focusing on issues of diversity in populations, treatment modalities, and potential covariates and determinants.

Conclusion

The present study provided a unique opportunity to explore the sociodemographic and health-related correlates of use of a spiritual healer for a medical problem in the U.S.. The use of these data provided several advantages: a nationally representative sample that expands upon previous surveys, large ethnically diverse sample of African-origin people, differentiation of faith healers and psychics, and availability of covariates assessing a full range of sociodemographic characteristics. Findings show that people who use a spiritual healer do not do so, on average, out of lack of education, marginal racial/ethnic or socioeconomic status, dire health straits, or lack of other healthcare options. These results encourage a rethinking of tacit assumptions about the determinants of the use of healers, and CAM generally, that have driven much of the research in this field.