Abstract
Faith-based interventions have emerged culturally sensitive way to address mental health issues among African Americans. This systematic review explores the scope and efficacy of faith-based mental health intervention outcomes among African Americans. Extracted data included the study population, setting, study design, intervention, adaptations, outcome measures, data analysis, and results. Methodological quality and sociocultural adaptations were also examined. The literature search identified 627 articles from 9 targeted databases. Five studies met inclusion criteria for this review. All five studies reported improvement in mental health outcomes and most studies sought to increase knowledge of mental health issues. Intervention settings, outcomes of interest, age groups, and attention to sociocultural modifications varied greatly. This review highlights a gap in the literature regarding African Americans and faith-based interventions. We discuss the potential for collaboration between social work professionals and clergy and the need to make faith-based interventions more accessible.
Research has suggested that the course and trajectory of mental illness in general may be more chronic for African Americans than non-Hispanic Whites (Williams et al., 2007). For example, the burden of depression is greater for African Americans compared to non-Hispanic Whites (Williams et al., 2007), and African Americans have more severe, persistent, and disabling depressive episodes compared to Whites (Bailey, Blackmon, & Stevens, 2009; Bailey, Patel, Barker, Ali, & Jabeen, 2011). African Americans who have a mental illness are often at increased risk of poor outcomes due to racial disparities (Office of Minority and National Affairs, 2010), including problems with treatment engagement and retention (Thompson, Bazile, & Akbar, 2004), overdiagnosis of schizophrenia and other psychotic illnesses (Strakowski et al., 2003), overprescription of antipsychotic medications (Lawson & Lawson, 2013), and underutilization of psychiatric services (Dana, 2002; Wang et al., 2005).
Despite the benefits of formal mental health services, African Americans are less likely to seek assistance from formal mental health providers, like social workers, to manage psychological problems (Barrio et al., 2003; Snowden, 2001; Wang et al., 2005). In fact, African Americans with severe mental illness are much less likely than Whites to have received mental health treatment during the previous year (<40% vs. >50%, respectively; Substance Abuse and Mental Health Services Administration, 2010). Several factors contribute to low mental health help-seeking rates among African Americans such as cultural mistrust (Whaley, 2001), lower belief in treatment efficacy (Gonzalez, Alegría, Prihoda, Copeland, & Zeber, 2011), discrimination (Woodward, 2011), and stigma regarding mental health issues (Gary, 2005).
Instead of formal mental health service providers, African Americans tend to rely on informal sources of support, such as church, friends, and family, to manage psychological problems (Matthews, Corrigan, Smith, & Aranda, 2006; Whaley, 2001). In a national sample of African Americans and Caribbean Blacks with a mental disorder, 22% did not receive help, 23% used informal support only, and 14% relied on professional services. The authors concluded that informal social networks may play a strong protective role in the lives of Black Americans (Woodward et al., 2008).
Research has suggested that African Americans generally maintain a more collectivistic worldview than non-Hispanic Whites (Weine & Siddiqui, 2009), and thus family and close peer support may have an effect on the lives of family members challenged by mental health conditions. Specifically, family support has been found to influence individuals’ mental well-being and health (Chatters, Taylor, Lincoln, & Schroepfer, 2002; Kawachi & Berkman, 2001). However, cultural stigma concerning mental health issues and negative attitudes toward treatment seeking among African American families may pose a barrier to improved mental health outcomes for individuals (Gary, 2005).
In addition to family support, African Americans often seek help from church and clergy to manage mental illness (Young, Griffith, & Williams, 2003). Religion has been found to be a protective factor that promotes health (Holt & McClure, 2006), and African Americans are more likely to identify as religious and involved in church life than non-Hispanic Whites (Chatters, Taylor, Bullard, & Jackson, 2009). Moreover, churches play a significant role in the African American community (Lincoln & Mamiya, 1990). Work by Chatters and associates have found that African Americans seek help more often from ministers than from family doctors, psychiatrists, or other mental health professionals (Chatters et al., 2011), and social support from church networks is a protective factor against depressive symptoms and overall psychological distress (Chatters, Taylor, Woodward, & Nicklett, 2014). Thus, clergy and family are significant sources of support for many African Americans experiencing emotional problems.
Significant variation exists with regard to terminology, nature, and spiritual versus secular focus of work in this area. First, faith-based interventions have emerged to respond to racial disparities in mental health while incorporating a preference for informal care and clergy support among African Americans. Terms such as church-based, faith-placed, faith-based, and collaborative programs have all been used to describe interventions with some level of incorporation of religion and spirituality into health promotion efforts (Bopp & Fallon, 2011; Campbell et al., 2007; DeHaven, Hunter, Wilder, Walton, & Berry, 2004). In this article, we use the term faith based, as used by Bopp and Fallon (2011), to refer to interventions that are administered in collaboration with churches and are not faith-placed (i.e., interventions that rely on churches, or its members, only for the sake of subject recruitment or location convenience). In addition to the multiple terms used to describe these interventions, there is variation in the scope and characteristics of faith-based interventions. Faith-based interventions have been developed to target various health conditions including cardiovascular health, cancer screening, weight control, smoking, and general health maintenance (DeHaven et al., 2004). Systematic reviews of CBHP (Church Based Health Promotion) suggest that this mode of intervention can significantly increase knowledge, improve screening, increase behavior change, and reduce risks associated with disease (DeHaven et al., 2004; Peterson, Atwood, & Yates, 2002).
Reviews of faith-based interventions to promote general health have suggested that churches can be a valuable partner and effective venue by which to promote individual health behavior change (Campbell et al., 2007; DeHaven et al., 2004; Peterson et al., 2002). However, the literature concerning faith-based interventions specific to mental health issues is limited and even more so regarding African Americans. In the only systematic review on faith interventions for mental health among African Americans, Hankerson and Weissman (2012) identified eight pertinent studies and concluded that churches are an underutilized resource that can provide individuals and families with valuable knowledge and resources regarding mental health.
Although the review by Hankerson and Weissman (2012) provided essential information about church-based mental health interventions for African Americans, the authors suggested that more empirical investigation is needed to establish the feasibility of churches to provide mental health promotion for African Americans. Further, there were several limitations of their review that should be addressed. First, they focused on interventions that included African American-only samples and excluded interventions that included African Americans in addition to other racial groups. They examined interventions that were implemented only in a church location; this did not take into account faith-based interventions conducted in collaboration with churches but which were not implemented within a church facility. Further, of the eight interventions reviewed, five of them focused on substance abuse issues and not emotional or psychological problems, such as depression. Thus, despite the important information this review provided, there is room for further analysis of the scope and efficacy of faith-based interventions that address psychological conditions among African Americans.
We make here a distinction between the concepts of “faith healing” and “mental health treatment.” Faith healing (or miraculous healing) refers to the belief that healing can be brought about by spiritual means through prayer and/or rituals (communal or individual processes) that are directly attributable to divine intervention (Village, 2005). This theological belief undergirds the emergence of healing ministries as a response to illness, including psychological well-being. Although not necessarily mutually exclusive, mental health treatment addresses psychological outcomes through formal helping channels such as counseling, psychotherapy, psychoeducation, prevention and mental health promotion programs, case management, and the like (Thyer & Wodarski, 2007), regardless of sponsorship under faith-based and/or secular auspices. A discussion of the continuum of beliefs anchored by these two concepts is beyond the scope of the article; nevertheless, this is an important consideration in examining the continuum of beliefs on how spiritual forces are harnessed—or not—in the service of psychological healing.
This systematic review examines the sociobehavioral literature to better understand the scope, efficacy, and sociocultural relevance of faith-based mental health interventions with African Americans. Specifically, the objectives of this review are to (1) review the literature on faith-based mental health interventions that include African Americans, (2) examine the efficacy of interventions by identifying mental health outcomes, (3) describe sociocultural adaptations of interventions by considering characteristics such as race, ethnicity, and religion, (4) evaluate the methodological quality of the interventions, and (5) appraise the evidence regarding the utility of faith-based mental health interventions. This information is necessary to inform community-based social work practice, guide the development of future empirical efforts, and stimulate interest in research on faith-based mental health promotion as a legitimate social work objective.
Methods
Search Strategy
In late 2013, we searched nine electronic databases for peer-reviewed articles published since 1980 that examined faith-based mental health intervention outcomes among African Americans. In addition, we conducted hand searches of relevant reference lists and performed electronic searches for authors who have published research on faith-based mental health interventions. Table 1 outlines the search strategy used in this review. A detailed review protocol was established a priori and included the following inclusion and exclusion criteria for each article in our review.
Description of Search Process and Results by Database Source.
Note: MeSH = medical subject headings.
aResults are not mutually exclusive across sources. Duplicate articles were found across databases and journals searched.
Inclusion Criteria
The study described or discussed a mental health treatment, intervention, or program. This includes interventions that attempted to (1) treat and reduce symptoms of mental illness, (2) improve behaviors or skills to prevent mental illness, (3) increase knowledge of mental health and illness through psychoeducation, or (4) some combination of these aims. Interventions could be formatted for individuals, groups, or both, and focus on individual behavior change or provider-level outcomes (e.g., clergy mental health training).
The study explicitly reported postintervention results for mental health outcomes. This includes psychological symptoms or mental and emotional distress as measured by any standardized clinical assessment or self-report rating. Specific disorders (e.g., depression) or general mental or emotional distress or symptomatology (e.g., psychological well-being, stress reaction) are included. Changes in behavior, knowledge, and skills were also relevant outcomes. Examples include changes in knowledge of mental illness symptoms, diagnoses, risk factors, and skills to prevent illness or promote mental health. This did not include outcomes that focus on substance abuse disorders or physical health problems only but included co-occurring disorders in which a mental or emotional disorder is a main concern.
The study examined faith-based interventions that included at least one of the following characteristics: (1) target and recruit participants from churches and faith settings, (2) train and utilize facilitators or interventionists from faith settings, (3) include curriculum that is spiritually based or adapted, (4) implemented in collaboration with churches or faith groups, or (5) some combination of the aforementioned items. Thus, interventions that involved recruitment in churches for convenience or that were located in a church facility but not spiritually focused were not of interest. Additionally, studies that only examined or measured spiritual or religious variables (e.g., religious coping, religious involvement) were not included.
The intervention was conducted and study sample recruited in the United States. Study results were published in English.
The study reported information on the racial and ethnic composition of the study sample and the study included African American adults (18 years or older).
Study results were published in a peer-reviewed journal between 1980 and 2013.
Exclusion Criteria
Studies that only recruited subjects from churches for sampling convenience or were conveniently located in a church facility.
Samples only included children and adolescents.
Studies that had primary treatment outcomes of physical illnesses, substance abuse, or other nonmental health conditions.
Qualitative studies, case studies, targeted literature reviews, dissertations, meeting abstracts, and reports were excluded given our focus on analyzing evidence for mental health intervention studies.
Procedures
Figure 1 describes the study selection procedures, which included several levels of article screening. Selected electronic databases (PubMed, ProQuest, PsycINFO, Social Work Abstracts, Web of Science, Social Services Abstracts, ATLA Religion Database, CINAHL, and Google Scholar) were searched using the prespecified search terms. Article abstracts were screened for eligibility by the primary author. The full text from studies that appeared to fit eligibility criteria or were in question were then screened by both authors using a form developed to assess for inclusion and exclusion criteria.

Flow of included articles.
Articles that met criteria for inclusion in the review went through a process of data extraction to identify information about the study population, setting, study design, faith-based intervention, adaptations, mental health outcome measures, follow-up, retention, data analysis, and results. The first author (K.H.) completed the data extraction using a table developed by both authors. The second author (M.P.A) then reviewed the selected articles and the extraction table for accuracy.
Level of Evidence
The methodological quality of studies meeting inclusion criteria was appraised using the Methodological Quality Rating Scale (Miller et al., 1995), which has been used in previous systematic reviews to assess the level of evidence of mental health intervention studies (Siantz & Aranda, 2014). This scale rates study quality according to 12 criteria that include group allocation, quality control, follow-up rate, follow-up length, contact, collaterals, objective, drop-outs, attrition, independence, analyses, and multisite (Vasilaki, Hosier, & Cox, 2006). The scale is used to assess whether or not a study reported information on each of the 12 dimensions. For example, a study that used blind interviewers to complete follow-ups would receive 1 point on the independence dimension. Study interventions can receive a score ranging from 0 to 17, with higher scores indicating higher methodological quality. Table 2 lists the methodological quality rating for each study included in this review.
Studies Reporting Mental Health Outcomes of Faith-Based Interventions With African Americans.
Note: DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, Fourth edition; MQRS= Methodological Rating Scale (Miller et al., 1995); PTSD = posttraumatic stress disorder.
Sociocultural Adaptations
Sociocultural adaptations of included interventions were identified using criteria developed by Bernal, Bonilla, and Bellido (1995) who suggested eight dimensions that should be considered when tailoring services to increase cultural sensitivity. These dimensions include language, persons, metaphors, content, concepts, goals, methods, and context which can be used as a theoretical framework to assess adaptations of sociobehavioral interventions (Bernal, Bonilla, & Bellido, 1995; see also Fuentes & Aranda, 2012). Thus, interventions that contain cultural adaptations should have modifications in one or more of these areas to increase the intervention’s cultural appropriateness for the target population. For example, interventions targeting African Americans may employ African American interventionists (persons), include stories about historical African American figures (content and metaphors), or incorporate religion and kinship bonds (context). After data extraction and assessment of the level of evidence, the primary author (KH), in collaboration with the second author (MPA), used the dimensions developed by Bernal et al. (1995) to assess the five selected studies for sociocultural adaptations. Table 3 describes identified adaptations from each article
Sociocultural Adaptations.
Results
The literature search identified 627 articles from the targeted databases that included the selected key terms. Most of the articles were excluded because they were descriptive studies (195), were conducted in other countries or languages (137), studied non–mental health-related health issues (95), were qualitative studies or case studies (51), focused on childhood issues (13), were duplicated (87), or were unrelated to the topic (27). Twenty-nine articles remained for more detailed review, but most were eliminated because they were qualitative (6), had no mental health intervention or outcomes (13), were not faith based (2), were review articles (2), or did not have any African Americans in the sample (1). Five studies were ultimately included in this systematic review and described outcomes of faith-based mental health interventions that included African American participants. Figure 1 details the selection process and Table 2 provides a description of included studies.
Settings and Types of Interventions
Included studies were conducted in various settings, such as a formal mental health treatment agency in St. Louis, a large African American church in Chicago, urban and rural churches in the southeast United States, faith-based and city-operated senior centers in Washington, DC, and churches and public meeting places in New York. One study was a randomized control trial with a follow-up evaluation, three of the studies were quasiexperimental and included pre- and postintervention assessments, and one study assessed only postintervention outcomes.
Bowland, Edmond, and Fallot (2012) evaluated an 11-session manualized psychoeducational, cognitive restructuring, and skills-building approach to addressing spiritual struggles in recovery from trauma to reduce depression, anxiety, and somatic symptoms, and posttraumatic stress disorder (PTSD). The Joy of Living program (Crewe, 2006) focused on increasing knowledge of mental health topic areas among participants in a three workshop series focusing on health and its relationship with mental health, memory loss, and depression. The Fear to Faith program (Suite, Rollin, Bowman, & la Bril, 2007) involved trauma workshops that included a multilingual, spiritually based curriculum designed to increase participants’ knowledge of neuropsychology, diagnostic process, criteria for PTSD, multicultural therapeutic approaches, and symptoms of PTSD. The Pickett-Schenk (2002) study assessed change in participants’ knowledge of the etiology and treatment of mental illness, understanding of the service system, and morale following participation in a monthly church-based support group for family members with a mentally ill loved one. Finally, Brown (2009) described a faith-based mental health education program that included 90-min workshops that focused on decreasing participant stigma regarding mental health issues.
Participant Characteristics
Of the five included studies, sample sizes ranged from 23 to 426 individuals. Two of the interventions (Brown, 2009; Crewe, 2006) did not collect any demographic data on individuals in the interventions but provided observational information. Two of the interventions (Bowland, Edmond, & Fallot, 2012; Crewe, 2006) specifically targeted adults aged 55 or older. Two other studies did not target a particular age-group and had participants with mean ages of 40 and 56 (Pickett, 2002; Suite et al., 2007). Most of the interventions had predominantly female participants, with a range of 64% (Suite et al., 2007) to 100% (Bowland et al., 2012) female samples. Although this systematic review sought to identify interventions that included African Americans, the degree of inclusion varied among interventions. Crewe (2006) and Pickett-Schenk (2002) included only African Americans in their interventions. However, the other interventions did not target a specific racial and ethnic group, and African Americans comprised between 14% (Bowland et al., 2012) and 53% (Suite et al., 2007) of the total samples.
In addition to demographic characteristics such as age, gender, and race, the participants of the included studies described various mental health risk factors and religious affiliations. For example, all participants in the Bowland et al’s. (2012) study had a history of trauma, the participants in the Crewe (2006) study were assumed to be at risk of poor mental health due to their age, low income, and low mental health literacy, Suite, Rollin, Bowman, and la Bril (2007) focused on first responders to crisis events, Pickett-Schenk (2002) recruited individuals who had a family member with severe mental illness, and Brown (2009) assumed church members would have less knowledge about, and high stigma around, mental health issues. All of these characteristics, in one way or another, were suggested to increase the target population’s risk of poor mental health outcomes, which warranted intervention. Further, each intervention focused on individuals with various degrees of religious affiliation. Three of the interventions (Brown, 2009; Pickett-Schenk, 2002; Suite et al., 2007) recruited church members or clergy to engage in their interventions. The other interventions required (Bowland et al., 2012) or assumed (Crewe, 2006) a significant spiritual orientation among participants.
Outcome Measures and Assessment Instruments
Each intervention sought to initiate change on a different mental health outcome. Bowland et al. (2012) focused on reducing depression, anxiety, and somatic symptoms with a secondary outcome of posttraumatic stress. Crewe (2006) sought to increase knowledge of mental health, memory loss, and depression. Suite et al. (2007) attempted to increase knowledge of neuropsychology, diagnostic process and criteria for PTSD, multicultural therapeutic approaches, and symptoms of PTSD among participants. Pickett-Schenk (2002) focused on increasing participants’ knowledge of the etiology and treatment of mental illness, understanding of the service system, and morale. Finally, Brown (2009) strived to increase knowledge and reduce stigma concerning mental illness.
Three of the five interventions used previously validated instruments to measure change in the targeted mental health outcomes (Bowland et al., 2012; Brown, 2009; Pickett-Schenk, 2002). Standardized measures used in the interventions include the Geriatric Depression Scale, Posttraumatic Stress Diagnostic Scale, and the Beck Anxiety Inventory (Bowland et al., 2012), in addition to the Journey of Hope Outcome Survey (Pickett-Schenk, 2002), the Level of Familiarity Scale, and the Attribute Questionnaire-Short Form (Brown, 2009). The authors of the other two interventions created their own measures to assess increased knowledge in topic areas covered by their curricula (Crewe, 2006; Suite et al., 2007).
Intervention Outcomes
All five articles reported positive changes in mental health outcomes as a result of the interventions being tested. Regarding a spiritually focused intervention for older trauma survivors, Bowland et al. (2012) reported that the intervention had a significant effect on outcomes with a decrease in depression, t(41) = –4.86, p < .0001, anxiety t(41) = –2.90, p < .01, somatic symptoms, t(41) = –3.08, p < .01, and PTSD, t(41) = –3.86, p < .001, from pretest to posttest for the intervention group compared to no significant changes in the control group. Further, the authors conducted follow-up evaluations with group members and results suggested that the treatment effects persisted over time. For the Joy of Living program, Crewe (2006) reported that among workshop participants, 80–83% gained knowledge of the signs of poor mental health, where to seek help, causes of depression, and effective treatments from pretest to posttest. Findings indicated a 96–98% improvement in knowledge about memory loss, including detecting signs, identifying causes, dealing with memory loss, and Alzheimer’s disease. Ninety-eight percent of attendees had a better understanding of the relationship between mental and physical health, according to the author.
Suite et al. (2007) examined changes from pretest to posttest for their Fear to Faith intervention. They reported that the average number of correct responses on the 20-item pretest assessing knowledge of neuropsychology, diagnostic process and criteria for PTSD, multicultural therapeutic approaches, and symptoms of PTSD was 6.46 (SD = 4.2). At posttest, they reported a mean number of correct answers of 14.23 (SD = 3.43), which was a statistically significant change, t(1, 134) = 20.7, p < .01. Brown (2009) analyzed pre- and posttest differences for 38 participants and concluded that there was a significant decrease in participants’ total stigma score (pre- to posttest difference = 4.25, p = .04) and that the relationship between a participant’s level of familiarity score and pretest stigma score was significant as well (r = –.451, p = .01), suggesting that the intervention workshop improved attitudes toward mental illness. Pickett-Schenk (2002) concluded that the church-based family group intervention successfully improved target measures at posttest. The author reported that 91% of participants had a significant increase in knowledge of the treatment and causes of mental illness, 91% reported increased knowledge of the treatment service system, and 69% reported significantly improved morale.
Sociocultural Adaptations
Each of the five studies described sociocultural adaptations in varying depth (see Table 3). The treatment dimensions of adaptation described by Bernal et al. (1995) were used to categorize the adaptations employed in each study. Two of the interventions (Brown, 2009; Suite et al., 2007) involved adaptations in language through the use of translators and translated intervention materials to allow Spanish speakers to access the material. Two interventions were adapted by using facilitators who were matched to participants (persons) on race and culture (Crewe, 2006) or gender (Bowland et al., 2012) to ensure that staff members were culturally competent and sensitive to the needs and experiences of the target population. For example, Crewe (2006) intentionally employed African American experts to “ensure that the message was couched in strengths versus pathology framework.” Suite et al. (2007) described the use of culturally appropriate metaphors, integrating biblical stories and characters to demonstrate intervention topics. For example, when discussing the concepts of fear, anxiety, grief, and tragic loss in context of the terrorist attacks of September 11, 2001, the facilitators invited participants to draw comparisons to the experiences of biblical characters such as Job, Joseph, Nehemiah, and Jeremiah. Concepts or content were adapted in two of the interventions to include spiritually focused coping strategies (Bowland et al., 2012; e.g., prayer, music, and visiting a new congregation) and theologically focused curriculum (Suite et al., 2007). Three of the interventions described some form of adaptation in the methods or context. Examples include the utilization of sharing circles as a technique to build on the life experiences of seniors (Crewe, 2006), collaborating with church leaders to recruit participants (Pickett-Schenk, 2002) and conducting interventions at culturally appropriate sites such as churches or senior centers (Brown, 2009; Crewe, 2006; Pickett-Schenk, 2002). Although authors described some form of adaptation, none of the studies focused explicitly on these dimensions nor did they assess the effect of adaptations on intervention outcomes.
Level of Evidence
The methodological quality of the five included studies was evaluated using dimensions of the rating scale presented by Miller et al. (1995). The range of scores was between 3 and 8 with a mean score of 4 of a possible score of 17, indicating overall low methodological rigor among the included studies.
Only one intervention (Bowland et al., 2012) used a randomized design. The other interventions were quasi-experimental with either posttest only or pre–posttests but no between-group comparisons or randomization. Intervention follow-up was reported only by Bowland et al. (2012) who performed a 3-month follow-up to assess whether or not treatment effects were maintained among participants. All of the studies included interventions and assessments performed in person or over the phone, although it was not clear if specialized training for interventionists or assessors was provided. All studies appeared to use appropriate analysis to interpret intervention findings. However, no other criteria were met by any of the interventions, including discussion of attrition, comparisons of multiple sites, blinding, or objective verification.
Discussion and Applications to Social Work
This study reviewed the extant literature on faith-based mental health intervention outcomes. We analyzed five studies that described and evaluated faith-based interventions that targeted changes in mental health outcomes. Based on a thorough review of these five studies, several conclusions can be made regarding the nature, efficacy, and potential effect of faith-based mental health interventions. First, when considering the efficacy of these interventions, none of the studies reported null or negative effects. All five studies reported some improvement in mental health outcomes; however, issues related to publication bias must be considered when evaluating the overall utility of the interventions highlighted in this review.
Second, there has been great variety in the types of settings in which these faith-based interventions are implemented. Other articles and reviews in this area have tended to focus on church-based interventions that emphasize church locations as the setting and intervention catalyst (Hankerson & Weissman, 2012; Peterson et al., 2002). However, several of the interventions reviewed here took a broader approach to addressing mental health issues in the faith context. For example, Crewe (2006) implemented a faith-based intervention in senior centers, Bowland et al. (2012) in a clinic setting, and Suite et al. (2007) in various community meeting places. Thus, the nature and reach of faith-based mental health interventions may go far beyond the traditional church settings that have been the focus of attention in this research area.
Third, participant characteristics varied greatly in the reviewed interventions. Interventions targeted individuals at different levels, including clergy, church members, and mental health consumers. In addition, the target populations varied in terms of the proportion of African Americans included in the study, with no group-specific outcomes cited. For example, Suite et al. (2007) described a diverse sample that included 53% African American participants. However, there was no between-group analysis or description of culturally tailored material for this racial group. Additionally, there was a large span in the age of participants across interventions, yet no study described age or gender effects: the majority of participants were female and none of the reviewed articles addressed the possible impact of age or gender on intervention outcomes including the reasons for the lack of males included in the studies.
Fourth, we found that four of the five studies focused on increasing knowledge of mental health issues. There are several possible reasons for this focus on increasing knowledge as opposed to other mental health outcomes such as symptom management or skills building. A majority of the interventions did not focus on individuals who were already experiencing mental health problems; instead, they sought populations that might be underserved or at risk but not yet identified as ill. Thus, it can be assumed that the focus of these interventions was prevention and increasing awareness instead of treatment. Considering that the target audience of the interventions in this review was religious individuals, including African Americans—who may experience high levels of stigma regarding mental health issues (Gary, 2005; Vogel, Wade, Wester, Larson, & Hackler, 2007) and tend not to seek formal treatment for mental problems (Whaley, 2001)—it may be appropriate for faith-based interventions to start with a focus on psychoeducation and engagement. Once individuals and communities are engaged, they may be more receptive to interventions that are treatment oriented and address symptoms and skill building.
Fifth, it is important to note that religion and faith can have positive and negative effects on well-being. Although often a protective factor, higher levels of religiosity are more often associated with lower levels of mental health help seeking (Abe-Kim, Gong, & Takeuchi, 2004; Moreno & Cardemil, 2013). Qualitative research by Matthews et al. (2006) revealed that religious African Americans are less likely to seek formal help for mental problems and feel that they have been taught to cope through God instead. Additionally, African Americans who seek clergy first for a serious problem are less likely to seek help later from a mental health professional (Neighbors, Musick, & Williams, 1998). In some cases, having a mental illness that requires formal mental health treatment can be seen as weakness or a lack of faith (Mattis et al., 2007). Therefore, religious views of mental health and illness may affect formal help seeking and treatment response among church-involved African Americans. Given the valence of the effect of religion and faith on mental health, it is important to address their respective role in future treatment studies including religious orientations to help seeking may be salubrious or detrimental to well-being.
Finally, despite the use of rigorous search strategies, only five studies reported mental health outcomes of faith-based interventions, and the methodological quality of the studies in this review was low. Several research gaps exist and the low number of included studies highlights that this area of research remains understudied and underdeveloped, although churches and faith-based programs are regarded as key sources of support and responsiveness to emotional and psychological distress (Chatters et al., 2009, 2011; Holt & McClure, 2006; Young et al., 2003; Lincoln & Mamiya, 1990). Several studies did not use a validated measure to assess changes in the target mental health outcome. There was minimal attention given to methodological reporting standards such as full descriptions of subject populations, sample recruitment procedures, dropout and attrition rates, and intervention protocols. Sociocultural adaptations were not detailed or explained, and no information with regard to theoretical underpinnings of the interventions was provided. No information was provided to address the potential for harm or unintended consequences of the interventions.
Implications
Several implications for future social work research and practice emerged from the review. Social work is clearly positioned to contribute to this body of research, given the commitment to bring quality, evidence-based solutions of social or interpersonal importance to underserved populations. Although the limitations and methodological quality of these studies should be considered, these positive results may suggest the potential of faith-based mental health interventions for African Americans. Preliminary evidence regarding the efficacy of these kinds of interventions may be particularly beneficial for underserved groups such as religious African Americans who may face cultural stigma due to their race (Gary, 2005) and religious stigma related to mental illness (Taylor, Ellison, Chatters, Levin, & Lincoln, 2000). Programs that have the potential to increase knowledge, awareness, or skills regarding mental health issues may help African Americans gain access to mental health care which can ultimately reduce racial disparities in mental health.
Social work research in the area of faith-based mental health interventions is clearly needed. Although this review indicated the potential utility of faith-based interventions, the low methodological quality and rigor of the reviewed studies cannot be ignored. Only one article in this review used a randomized controlled study design (Bowland et al, 2012) to determine intervention effects and causality. It is unclear how recruitment procedures of study participants influenced the process and outcomes of the studies included in the review. Drawing inferences from research designs based on samples of participants who mainly self-select to participate calls into question the external validity or generalizeability of the findings. For example, it is unknown how the absence or presence of hard-to-reach groups based on convenience sampling procedures plays a role in the validity of the conclusions. Questions remain regarding if faith-based interventions are reaching hard-to-reach and at-risk populations (stigmatized conditions; age- or gender-specific issues), and if the type of mental health problem implies differential sampling for persons who agree to seek help (e.g., females) and those who do not (younger age groups).
Researchers and practitioners must acknowledge the effect that policy changes such as health care reform will have on individuals’ ability to access mental health care. More interventions must incorporate information on navigating the mental health care system while informing individuals and communities of the availability of mental health resources. Faith-based settings and interventions may be ideal for providing this information.
The review also highlighted several areas for growth in clinical social work practice. These faith-based interventions demonstrated willingness on the part of some clergy, church members, and the general public to address and integrate issues of mental health and religion. Social workers and mental health practitioners would be wise to collaborate with local congregations that serve African Americans, as this may help to bridge the gap between formal and informal support for mental health problems among African Americans, and other groups challenged by mental health disparities.
Social work must also acknowledge that tensions may emerge at the intersection between personal identity and values and professional practice. Although the National Association of Social Workers Code of Ethics (2008) promotes “social justice and social change with and on behalf of clients,” how this principle is applied to research and practice with clients representing diverse religious or spiritual orientations, or lack thereof, is open to interpretation based on diverse worldviews (Aranda, 2008; Hodge, 2012). Several controversial issues emerge regarding how the field of social work approaches faith-based, mental health interventions. What substantive, ethical, and practical issues emerge with respect to the faith healing practices and formal mental health treatment when these are predicated on different—and often conflicting—underlying mechanisms of action? What type of social work provider qualifications and competencies are needed to open mutual exchange and learning between faith-based health and secular mental health treatment practices in general? What would the division of competencies and partnerships between faith-based organizations and social work look like? What special considerations emerge in racial and ethnic communities? What critiques can be offered with respect to potential harm or unintended consequences as a direct result of faith-based interventions (and do we compare these with unintended consequences of non–faith-based interventions)? These questions have not been well addressed in social work practice and research with regard to community-based, mental health interventions.
None of the interventions identified social workers as interventionists or providers. Researchers have suggested several strategies social workers can use to establish partnerships and build collaborations with African American congregations including potential qualifications needed for engagement. First, social workers need to examine their own beliefs regarding religion and spirituality as they relate to mental health treatment (Queener & Martin, 2001; Whaley & Davis, 2007). Social work practitioners may need to seek additional training to increase their knowledge of the religious experience of African Americans, the role of the church in African American communities, and overall cultural competence (Queener & Martin, 2001). This may also help clinicians to become more familiar with the values, cultural norms, practices, and language used by religious clients to better integrate religious coping into mental health treatment programs (Adksion-Bradley et al., 2005). Equally important is increased education on the culture of Black churches and African Americans’ relationship with faith. Social workers must also spend a significant amount of time developing a relationship of trust with local clergy and congregants in order to earn acceptance (Adksion-Bradley, Johnson, Lipford Sanders, Duncan, & Holcomb-mccoy, 2005; Austin & Claiborne, 2011). This might include observing religious services and meetings and offering resources and support to the church. Buy-in and support from African American ministers is also a critical aspect of developing collaborations with African American churches as pastors, other clergy, and key church members serve as key gatekeepers for church communities (Taylor et al., 2000). Orientation and training for church-affiliated providers are equally important and serve to address the bidirectional processes of collaboration and partnerships between stakeholders and social work practitioner networks.
In addition, attempts must be made to make faith-based intervention materials more accessible to clinicians. The articles reviewed here described interventions in varying detail and most did not use a manualized intervention that could be replicated and adapted for future intervention efforts. Promising faith-based mental health programs must be made more widely available to social workers to begin to meet the growing needs in African American communities.
Limitations
The conclusions based on this review have noteworthy limitations. First, small sample sizes, use of different outcome measures, and heterogeneity of study designs and samples limited our ability to determine which intervention components are most effective. Second, we only examined articles published in the United States during a specific time period. Third, possible publication bias of interventions with positive results may have exaggerated the effect of these types of interventions. Fourth, none of the reviewed studies described a consideration of intragroup heterogeneity among African American participants. Research with African Americans has suggested that there are important cultural and behavioral differences concerning mental health between African Americans and Caribbean Blacks that should be explored (Taylor, Forsythe-Brown, Taylor, & Chatters, 2014; Williams et al., 2007). Fifth, our review only focused on studies that included African Americans, the majority of whom had Christian affiliations. However, there may be similar studies that focused on other racial and ethnic groups and religions that could shed light on the effect of faith-based mental health interventions. Future work should consider cross-group reviews.
Conclusion
To our knowledge, this was the first systematic review to examine faith-based mental health interventions that include African Americans. Considering the ongoing racial disparities, low rates of professional service use, and stigma regarding mental health in African American communities, it is critical that interventions such as these be developed to improve mental health outcomes. However, there continues to be very little empirical research in this area and what evidence does exist is of questionable quality. Churches and other religious institutions are often the center for healing, yet this natural community resource remains largely untapped in terms of mental health needs. Reducing disparities in mental health outcomes could be addressed by engaging faith-based programs in rigorous treatment studies with sociocultural adaptations that increase relevance and acceptability of mental health care in the community.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
