|Home | About | Journals | Submit | Contact Us | Français|
To present nationally representative data on the part played by clergy in providing treatment to people with mental disorders in the United States.
The National Comorbidity Survey (NCS), a nationally representative general population survey of 8,098 respondents ages 15–54.
A modified version of the Composite International Diagnostic Interview was used to assess DSM-III-R mental disorders. Reports were obtained on age of onset of disorders, age of first seeking treatment, and treatment in the 12 months before interview with each of six types of professionals (clergy, general medical physicians, psychiatrists, other mental health specialists, human services providers, and alternative treatment providers).
One-quarter of those who ever sought treatment for mental disorders did so from a clergy member. Although there has been a decline in this proportion between the 1950s (31.3 percent) and the early 1990s (23.5 percent), the clergy continue to be contacted by higher proportions than psychiatrists (16.7 percent) or general medical doctors (16.7 percent). Nearly one-quarter of those seeking help from clergy in a given year have the most seriously impairing mental disorders. The majority of these people are seen exclusively by the clergy, and not by a physician or mental health professional.
The clergy continue to play a crucial role in the U.S. mental health care delivery system. However, interventions appear to be needed to ensure that clergy members recognize the presence and severity of disorders, deliver therapies of sufficient intensity and quality, and collaborate appropriately with health care professionals.
Despite the enormous burdens imposed by mental illnesses and the availability of effective treatments, unmet need for mental health care continues to be an enormous public health problem (Kessler, Olfson, and Berglund 1998; Olfson et al. 1998; Wang, Berglund, and Kessler 2000; Wang et al. 2002). To begin addressing this, it is crucial to shed light on the help-seeking processes of those with mental disorders, including to whom they go for help and what types of services they receive (Rogler and Cortes 1993; Gallo et al. 1995).
Important unanswered questions remain concerning seeking help for mental disorders from the clergy. In one of the few earlier epidemiologic studies conducted among community samples, Gurin and colleagues (1960) observed in the middle of the last century that 42 percent of those seeking help for emotional problems sought it from a clergy member, a considerably larger proportion than those contacting physicians, mental health specialists, or any other professional. In a follow-up study 25 years later (Veroff, Kulka, and Douvan 1981), this proportion had fallen but was still high in absolute terms (34 percent). The Epidemiologic Catchment Area (ECA) study in the early 1980s found that only 20 percent of those who sought treatment for mental disorders contacted clergy and other human services professionals (Regier et al. 1993); however, seeking help from the clergy continued to be as common as seeking care from mental health professionals (Larson et al. 1988). Unfortunately, there is little empirical data shedding light on the more recent extent of seeking mental health care from clergy members. In addition, it is also unclear what types of patients now seek help from clergy and for what reasons; earlier investigations suggest that a variety of demographic (e.g., age, gender, race, region of the country, social class) and clinical characteristics (e.g., type and seriousness of mental disorders) may all be important factors (Larson et al. 1988; Ray, Raciti, and MacLean 1992; Sorgaard et al. 1996; Mitchell and Baker 2000).
It is also critically important to understand what services the clergy provide to those with mental disorders, especially given the growing literature (Katon et al. 1995; 1996; Lehman and Steinwachs 1998a; Wells et al. 2000) suggesting that mental health treatments must conform with evidence-based guidelines (Agency for Health Care Policy and Research 1993; American Psychiatric Association 1994; 1997; 1998; 2000; Lehman and Steinwachs 1998b) regarding modality, intensity, and duration of treatment in order to be effective. Earlier surveys of clergy members have revealed that many are insufficiently trained in recognizing the presence and severity of psychopathology as well as in providing pastoral counseling (Wylie 1984; Rupert and Rogers 1985; Domino 1990; Weaver 1995). Perhaps for these reasons as well as having competing demands on their time, clergy members have been found to spend less than 10 percent of their time providing pastoral counseling (McCann 1962; Virkler 1979; Mollica et al. 1986). Earlier studies have also found that clergy refer fewer than 10 percent of those with emotional problems to other mental health care providers (McCann 1962; Piedmont 1968; Hong and Wiehe 1974; Virkler 1979; Mollica et al. 1986). More recent efforts to improve the mental health services delivered by clergy have had a positive impact on the quality of pastoral care and cooperation with health care professionals remains unknown (O'Connor and Meakes 1998).
The specific aims of the current study were two-fold. First, we sought to describe the recent extent and correlates of seeking mental health care from clergy in the National Comorbidity Survey (NCS), a large general population survey of the United States conducted in the early 1990s. Second, we sought to shed light on the functions that clergy play in the mental health care delivery system by identifying characteristics of individuals they see, their frequency of visits with potential counselees, and the extent to which they work alone versus in conjunction with other professionals.
The NCS data from 8,098 respondents were collected between September 1990 and February 1992 from a stratified, multistage area probability sample of persons ages 15–54 in the 48 coterminous United States (response rate 82.4 percent). The data were weighted for differential probabilities of selection and differential nonresponse as well as to adjust the sample to approximate the cross-classification of the population distribution on a range of sociodemographic characteristics. These methods are described in more detail elsewhere (Kessler et al. 1994; Kessler, Little, and Groves 1995).
Psychiatric diagnoses were assessed with a modified version of the Composite International Diagnostic Interview (World Health Organization 1990) and included DSM-III-R (American Psychiatric Association 1987), major depressive episode and dysthymia, panic disorder, generalized anxiety disorder, phobias (either simple, social, or agoraphobia with or without panic), nonaffective psychotic disorders (schizophrenia, schizoaffective disorder, schizophreniform disorder, delusional disorder, brief psychotic disorder, and psychotic disorder not otherwise specified), and addictive disorders (alcohol or drug abuse or dependence). The WHO field trials (Wittchen 1994) and NCS clinical reappraisal studies (Wittchen et al. 1995; 1996; Kessler et al. 1998) have documented acceptable reliability and validity of all these diagnoses.
We created an operational definition of serious mental illness (SMI) consistent with Public Law 102-321, which defines SMI as having any DSM mental disorder, substance use disorders, and developmental disorders that lead to “substantial interference” with “one or more major life activities.” Respondents who met criteria for 12-month DSM-III-R disorders were defined as having SMI if their disorder was associated with (a) vocational incapacity (as indicated either by inability to hold a job or frequent work absence due to mental health problems), (b) serious interpersonal difficulties (as indicated either by social isolation or frequent interpersonal difficulties), (c) a suicide plan or attempt within the past 12 months, or (d) if their disorder met criteria for a “Severe Mental Illness” as operationalized by the National Advisory Mental Health Council of the National Institute of Mental Health (1993). More detailed discussions of this operationalization of SMI are presented elsewhere (Wang, Demler, and Kessler 2002; Kessler et al. 1996; 1999). We also created an ordinal variable representing the severity of mental disorders. Respondents with no mental or substance disorders received a score of 1; those with non-SMI mental or substance abuse disorders received a score of 2; and those qualifying for SMI or substance dependence received a score of 3.
The NCS respondents were asked at what age they first saw each of six types of professionals for problems with their emotions, nerves, mental health, or alcohol and substance use: clergy (including ministers, priests, or rabbis); psychiatrists; general medical doctors other than psychiatrists (including general practitioners, family physicians, and physician specialists such as cardiologists or gynecologists); mental health specialists other than psychiatrists (including psychologists, mental health counselors, social workers other than in a social services agency, or providers in outpatient mental health or alcohol or drug clinics); human services providers (including counselors or social workers specifically in a social service agency or department); and alternative treatment providers (including spiritualists, herbalists, natural therapists, faith healers, or self-help groups). All respondents were also asked how many times they had contacted these six types of professionals for mental health care within the year prior to survey.
Sociodemographic variables used as predictors of treatment-seeking included: cohort (categorized as born in 1936–1945, 1946–1955, 1956–1965, or 1966–1975); sex; race (non-Hispanic white, non-Hispanic black, Hispanic, other); marital status (never married, married, previously married); educational attainment (less than high school, high school graduate, some college, college graduate); urbanicity (metropolitan area, other urbanized area, rural); and U.S. region (northeast, Midwest, south, west). Clinical variables used as predictors included: type of mental disorder (depression/dysthymia, panic disorder, generalized anxiety disorder, substance/alcohol abuse or dependence, nonaffective psychoses, and phobias); number of disorders; and presence of suicidal thoughts, suicidal plans, or making suicide attempts. In the analyses that focused on the prediction of any treatment with a clergy member, we also included variables on past treatment with each of the other five types of professionals as predictors.
We constructed Kaplan-Meier (Kaplan and Meier 1958) curves among all respondents with mental and substance disorders to calculate cumulative lifetime probabilities of contacting any, as well as specific types of, providers, among cohorts defined by age at interview (i.e., 15–24, 25–34, 35–44, and 45–54). Among respondents with mental disorders who made any treatment contact, we calculated the probabilities of making first treatment contact with clergy and other specific types of professionals in different calendar time periods (prior to 1960, 1960–1969, 1970–1979, 1980–1985, and 1986–1991). We then calculated the proportions contacting clergy and other types of providers in the year prior to survey, among all respondents and among strata defined by the seriousness of mental and substance disorders. We identified the mean number of visits to providers in the prior year, among those making contact with specific provider types. Among respondents who made contact with a clergy member in the prior year, we calculated the frequencies of being seen alone or with other types of providers.
We constructed a discrete-time survival model (Efron 1988) among all respondents with lifetime mental disorders, to study predictors of the probability of making treatment contact with clergy after first onset of a mental or substance disorder. Sex, race, urbanicity, and region of the country were treated as time-invariant predictors. Age at interview (cohort), education, marital status, and prior history of contact with specific types of professionals were treated as time-varying covariates. Reported ages of onset were used to assign time-varying values of the type and number of mental disorders, as well as the presence of suicidal thoughts, plans, and attempts. Time-varying covariates representing prior treatment contact with other specific professionals were assigned from the reported ages of first contacting providers. Finally, we also constructed a logistic regression model of first contacting clergy among respondents with lifetime mental disorders who sought any treatment, to identify predictors of seeking one's first treatment from clergy specifically versus other types of providers. Predictors included all those described above, except variables representing prior contacts with other specific professionals. Standard errors of parameter estimates in models were estimated using the Taylor Series Linear Approximation technique in SUDAAN statistical software version 7.5.6 (SAS Institute, Cary, NC). These corrections adjust for the fact that the NCS is based on a weighted dataset with geographic clustering of observations.
Figure 1 presents Kaplan-Meier curves of contacting any provider by the 46.9 percent (standard error 1.0) of NCS respondents with lifetime mental and substance disorders, separately for subsamples defined by their age at interview. The cumulative lifetime probability of making any treatment contact has significantly increased in more recent cohorts (χ32=45.2; p<.001). Figure 2 presents similar curves of contacting clergy specifically. While the cumulative lifetime probability of contacting clergy has also increased in more recent cohorts (χ32=26.0; p<.001), this increase does not appear to be as great as that seen for contacting any provider.
Table 1 presents probabilities of making one's first treatment contact with clergy or other specific types of professionals in different time periods, among respondents with mental disorders who made any treatment contact. The proportion contacting clergy was highest prior to 1960 (31.3 percent), then declined during the 1960s and 1970s, but appears to have stabilized during the 1980s and early 1990s at approximately 25 percent. The proportion first contacting psychiatrists declined even more dramatically over this time period (40.4 to 16.7 percent). On the other hand, the proportions first contacting non-psychiatrist mental health specialists (9.2 to 43.1 percent) and alternative treatment providers (5.3 to 25.9 percent) rose substantially.
The first set of columns in Table 2 present the net effects of sociodemographic and clinical variables on the probability of contacting clergy. Because this model was constructed among all respondents with mental or substance disorders, whether or not they sought treatment, odds ratios and 95 percent CIs estimate effects of predictors on the overall process of seeking care from clergy, including effects on steps such as recognizing one has a problem, deciding the problem requires care, and seeking that care from a clergy member. Significant predictors included-being in younger age cohorts; being of nonblack race; being a student with less than a high school education; having suicidal ideation, plans, or attempts; having panic disorder, depressive disorders, and PTSD relative to phobias; having had prior contact with psychiatrists, general medical physicians, or human services professionals; and having fewer numbers of mental disorders. The second set of columns in Table 2 presents the effects of predictors on first contacting religious professionals, but only among respondents with disorders who sought care from any type of provider. Because of this restriction, odds ratios and 95 percent CIs only estimate effects of predictors on the step of seeking care from clergy specifically versus other types of providers (but not earlier steps in the help-seeking process such as recognizing one has a problem or deciding the problem requires care). Significant predictors of contacting clergy versus other providers included: residing in the south relative to the west; being married relative to never married; and not having substance disorders or panic disorder.
The distribution of 12-month mental and substance disorders among NCS respondents is shown in Table 3. Table 4 presents the probabilities of those with active mental and substance disorders to have contacted providers in the year prior to survey. Among the 12.3 percent contacting any provider, 21.1 percent saw a clergy member (see first column). The probability of contacting clergy was greater than for contacting psychiatrists or human services providers, but less than for contacting other medical physicians, mental health specialists, or alternative treatment providers. Respondents with more serious mental disorders were more likely to contact all types of providers in a generally monotonic fashion across the three severity levels (displayed in the second set of columns). While those with more serious substance disorders were more likely to contact most types of professionals, this pattern was not consistently seen for contacting clergy and nonpsychiatrist medical doctors (displayed in the third set of columns). The fourth set of columns shows that those with more serious mental and substance disorders combined, were more likely to contact all types of providers, again in a generally monotonic fashion across the three severity levels.
Table 5 presents the mean number of visits to providers in the year prior to survey, among respondents who made any visit to that type of provider. As shown in the first column, the mean number of visits to clergy (5.7) was lower than that to any other type of provider. As shown in the second set of columns, respondents with more serious mental disorders made more visits to mental health specialists and human services providers; however, this pattern was not seen for clergy, psychiatrists, and other medical physicians. Those with more serious substance disorders made more visits to only nonpsychiatrist physicians, but not to any other type of provider (see the third set of columns). Those with more serious mental and substance disorders combined made more visits to general medical physicians, mental health specialists, and human services professionals but not psychiatrists, clergy, or alternative treatment providers.
Among respondents contacting clergy in the prior year, the majority (56 percent) saw clergy members alone without other providers (see Table 6); only 38.9 percent saw clergy together with a psychiatrist, other physician, or mental health specialist. Table 7 presents the types and severities of mental and substance disorders among those contacting clergy. While 48.2 percent had no discernible mental or substance disorder, 23.5 percent had the most serious and impairing forms of these illnesses (see first column). As shown in the last three sets of columns, substantial proportions of those with the most serious disorders seen by clergy were not treated in conjunction with health care professionals.
Table 8 presents the mean number of visits to providers in the year prior to survey among respondents who made any visit to a clergy member. Respondents with more severe mental and substance disorders were not observed to make more frequent visits to clergy, whether or not the respondent was also treated by other professionals.
Results of this study shed light on the important role that the clergy continues to play in the U.S. mental health services delivery system. Our results confirm earlier findings (Gurin, Veroff, and Feld 1960; Veroff, Kulka, and Douvan 1981; Regier et al. 1993) suggesting that use of clergy declined during the 1960s–1970s. However, we found that use of clergy stabilized in the 1980s and early 1990s, with substantial proportions (approximately one-quarter) of those with disorders who sought mental health care continuing to first contact clergy members. Reasons for this recent stabilization after decades of “secularization” are unclear but may be related to a resurgence in religious beliefs and behavior among Americans as well as growing interest in spiritually oriented healing modalities (Bezilla 1993; Bergin, Payne, and Richards 1996; Princeton Religion Research Center 1996). Our results are also consistent with earlier research in showing a sustained decline in contacts with psychiatrists, and a dramatic rise in the use of nonpsychiatrist mental health specialists as well as alternative, self-help, and nontraditional forms of mental health care (Eisenberg et al. 1993; 1998; Regier et al. 1993; Kessler et al. 2001).
These results should be interpreted with the following methodological limitations kept in mind. First is the long and uncertain period of recall associated with dating disorder onset and initial treatment, as well as the absence of confirmatory data from providers. Many self-help groups and recovery movements are religiously based or sponsored (Galantar 1997); to the extent that this is the case, we may have underestimated the role of religion in the service delivery system of patients with mental disorders. In addition, the nature and quality of the treatments delivered after contact are unclear. Nor could we study the outcomes of treatment, due to the nonrandom nature of contacts with clergy in this observational study. Furthermore, predictors were limited to a small number of patient demographic and clinical variables. The cross-sectional nature of the survey data makes it impossible to conclude that factors associated with treatment contacts were related causally. Finally, the data employed in this study were collected in the early 1990s. Many changes have occurred in treatments for mental disorders (e.g., introduction of new medications with potentially greater tolerability) and in mental health care delivery systems (e.g., greater proportions receiving mental health treatment under managed care) since that time. The National Comorbidity Survey Replication (NCS-R), which is currently in progress in the United States, will provide data on any temporal changes in the use of clergy that may have occurred in the past decade (Kessler, Olfson, and Berglund 1998).
In spite of these potential limitations, several observations in this study raise concerns about the mental health services that clergy provide. First, we found the frequency of visits was lower among respondents contacting clergy than among those using any other type of provider, confirming earlier surveys (Virkler 1979; Mollica et al. 1986) showing that the counseling treatments offered by clergy members are often of low intensity (e.g., four sessions on average in one study [McCann 1962]). It is worth noting that the mean number of visits to clergy (5.7 in the prior year) that we observed falls below the number of sessions required in clinical trials of time-limited psychotherapies with documented efficacy (Agency for Health Care Policy and Research 1993; American Psychiatric Association 1998; 2000). While we confirmed earlier findings (Larson et al. 1988) that clergy frequently contact individuals with the most serious and impairing mental and substance disorders, we did not find evidence that the clergy titrated the intensity of their visits to the presence and severity of disorders. Whether the poor recognition of the presence and severity of mental disorders or lack of training in pastoral counseling observed previously (Wylie 1984; Rupert and Rogers 1985; Domino 1990; Weaver 1995) also underlie our findings, needs to be explored further.
Because it is unclear to what extent clergy members provide mental health treatments with proven efficacy (Agency for Health Care Policy and Research 1993; American Psychiatric Association 1994; 1997; 1998; 2000; Lehman and Steinwachs 1998b), it is concerning that large proportions of those with disorders were seen by clergy in the absence of contact with psychiatrists, other mental health specialists, or other physicians. The low levels of cooperation between clergy and health care providers observed in earlier studies have been explained by lack of training in assessing needs for referral and transitioning patients to other providers, lack of knowledge of referral sources in the community, lack of feedback to clergy from other treaters, financial considerations, lack of perceived or common values, and outright negative attitudes toward clergy held by many health care professionals (McCann 1962; Piedmont 1968; Hong and Wiehe 1974; Virkler 1979; Mollica et al. 1986; Neeleman and Persaud 1995).
It is also of concern that suicidal thoughts and behaviors increased contacts with clergy and that suicidal individuals who sought treatment were as likely to contact the clergy as other providers. Clergy members have been found in prior studies (Holmes and Howard 1980) to be relatively unprepared to assess suicidality. Even if individuals at high risk are eventually recognized and referred to health care providers, as is mandated in most jurisdictions, delays in the receipt of potentially life-saving interventions may occur.
Recent cohorts were more likely to contact clergy although not to a greater extent than other providers; this may reflect general effects of recent public education programs focused on mental disorders (Regier et al. 1988; Ross 1993) and liberalization of public attitudes toward mental illness and its treatment (Bhugra 1989). African Americans were less likely to contact clergy although not to a lesser degree than other providers, confirming earlier findings that blacks are generally less likely to seek all forms of mental health care (Temkin-Greener and Clark 1988; Padgett et al. 1994; Larson et al. 1988). Students were also less likely to contact clergy but not to a lesser extent than other professionals, perhaps because students lack the resources to seek treatment from all providers (Leaf et al. 1988; Wells et al. 1988; Wang et al. 2000). Mass screening efforts may have led to more treatment-seeking for depressive disorders, both from clergy as well as other professionals (Jacobs 1995). The prominent somatic symptoms that are often part of anxiety disorders such as panic disorder have been suggested to lead to more treatment-seeking in general, although to a larger extent from established health care professionals (Katon, Von Korff, and Lin 1992; Katerndahl and Realini 1995). Prior contact with psychiatrists, general medical physicians, or human services professionals may be a marker of those with greater needs or lower thresholds to seek help from all providers; on the other hand, individuals with a greater number of disorders who have not already sought care, may be a marker of individuals who are more resistant to seeking care in general. Married individuals who sought mental health care preferentially contacted clergy versus other providers, supporting earlier findings that marital and family problems are the most common presenting complaints brought to the clergy (Gurin, Veroff, and Feld 1960). Residents of the south who sought help also turned preferentially to the clergy, confirming earlier observations (Koenig, George, and Siegler 1988) that religion is especially important in the “Bible Belt” of the United States for those coping with emotional problems.
In summary, while the clergy continue to be a frequent point of contact in the U.S. mental health care delivery system, additional efforts may be needed to optimize their role. First, more research is needed, especially on the quality and outcomes of mental health care delivered by the clergy. In addition, interventions are needed to ensure that the clergy are trained in recognizing patients with mental and substance disorders, especially severe disorders in need of urgent care (Weaver 1992; 1995). Improved training in care planning (Weiss 1991) and the provision of pastoral counseling are also critical to insure that the care that is delivered is of sufficient intensity and quality (O’Connor and Meakes 1998). While some formal mental health care training programs for clergy are available (e.g., Stone and Clements 1991; Wicks and Parsons 1993; Childs 1990; Stone 1994), data on the effectiveness of such training programs is extremely limited (Gartner, Larson, and Allen 1991). Finally, interventions that facilitate timely referrals and greater collaboration with health care professionals are also crucial (Larson et al. 1988). Studies of smaller programs have shown that such efforts can be effective at increasing the rate and appropriateness of referrals (Kaseman and Anderson 1977); however, instances of large-scale or systematic efforts in this regard are lacking.
Clearly the success of any future efforts to improve the quality and outcomes of mental health care delivered by the clergy will depend in large part on the motivation of clergy to participate in such efforts. Prior reports indicate that many clergy are likely to possess such motivation. For example, many clergy consider the ability to counsel an essential skill and approximately half reported attending seminars or workshops on mental health in the prior year (Lau and Steele 1990). On the other hand, others have also written of a hesitancy by some clergy toward greater “professionalization” of their delivery of mental health services, in part out of concern that this could blur traditional clergy roles and cause actual or perceived deviation from religious principles (Jorjorian 1972; Bruder 1965). Understanding and accommodating such concerns will be crucial to ensure that patients who contact clergy for mental and substance disorders receive effective treatment and ultimately experience improvements in their health outcomes.