Religious Involvement and DSM-IV Anxiety Disorders among African Americans

Abstract

This study examined the relationship between eight measures of religious involvement and five anxiety disorders among a nationally representative sample of African Americans (N=3,403). The DSM-IV World Mental Health Composite International Diagnostic Interview was used to assess 12-month and lifetime prevalence for each disorder. Logistic regression indicated weekly service attendance was inversely associated with 12-month and lifetime panic disorder, lifetime agoraphobia, and 12-month and lifetime PTSD. Prayer was inversely associated with 12-month agoraphobia, 12-month social phobia, and lifetime PTSD. Listening to religious radio and looking to God for strength were also inversely related to, respectively, 12-month and lifetime panic disorder. Conversely, reading religious materials was positively associated with 12-month panic disorder, 12-month agoraphobia, lifetime PTSD, and lifetime generalized anxiety disorder. The results are discussed in light conceptual models that specify multiple and sometimes divergent pathways through which religion impacts health, and suggestions for clinicians addressing anxiety disorders are delineated.

Anxiety disorders are the most common mental health condition in the United States (Rosmarin & Leidi, 2020). It is estimated that approximately 41% of Americans will have a diagnosable anxiety disorder over their life span (Kessler et al., 2012). A wide variety of disorders exist, with panic disorder, agoraphobia, social phobia, post-traumatic stress disorder, and generalized anxiety disorder among the most prevalent. These disorders are induced by different stimuli but are all characterized by excessive anxiety and the behaviors that stem from this condition. The impaired functioning associated with such disorders results in substantial societal and individual costs including, for example, decreased personal earnings (Kessler et al., 2008) and increased suicide ideation and attempts (Joe et al., 2006).

One potential predictor of clinical anxiety is religious involvement or engagement (Hopkins & Shook, 2017). The relationship between religion and anxiety may be particularly salient for African Americans, given the elevated levels of religious involvement among members of this population (Nguyen, 2020). In theory, religious resources may function as a protective factor and/or therapeutic or coping factor (Ellison & Levin, 1998). However, comparatively little research has examined the relationship between religious engagement and anxiety disorders (Bonelli & Koenig, 2013), with the paucity of research being especially pronounced among African Americans (Hopkins & Shook, 2017). The present study sought to address this gap in the literature by examining the relationship between eight measures of religious involvement and the five anxiety disorders mentioned above among African Americans. We begin by briefly summarizing the literature on religion and anxiety, followed by a discussion of religion and anxiety among African Americans, and conclude the section by describing two fundamental models that shape our expectations regarding the study findings.

Religion and Anxiety

The extant research on religious involvement and anxiety is decidedly mixed (Rosmarin & Leidi, 2020). Some studies report a positive relationship between the two constructs (Ellison et al., 2014), others an inverse relationship (Abdel-Khalek et al., 2019), and still others no relationship (Rasic et al., 2011). This research differs from the relatively consistent, salutary relationships that have been observed with some other health outcomes (Koenig & Al Shohaib, 2019; Oman & Lukoff, 2018; Oman & Syme, 2018).

To account for the disparate findings in the area of anxiety, some observers have highlighted the multidimensional nature of religion (Rosmarin & Leidi, 2020). In other words, the relationship with anxiety may differ depending upon the dimension of religion measured (Ellison et al., 2014). For example, organized religious involvement (service attendance), non-organized involvement (reading religious materials, prayer), subjective religious engagement, and religious coping may have different relationships with anxiety (Rosmarin & Leidi, 2020). Other factors that may help account for different outcomes are the population featured in a given study (Abdel-Khalek et al., 2019) as well as the fact the most studies examine broad symptoms of anxiety as opposed to DSM measured clinical disorders (Bonelli & Koenig, 2013).

Religion and Anxiety among African Americans

African Americans tend to have lower prevalence rates of anxiety disorders than European Americans (Hopkins & Shook, 2017). Although when anxiety disorders do occur, they tend to be relatively more chronic, functionally impairing, and unresponsive to treatment (Hopkins & Shook, 2017). For example, Himle et al. (2009) found that African Americans with anxiety disorders were more likely than European Americans to experience severe disorders. Similarly, African Americans reported comparatively greater levels of functional impairment.

Another key construct in terms of understanding anxiety among African Americans may be religion. As noted above, African Americans are highly engaged in religious activities relative to European Americans (Chatters et al., 2009; Mohamed et al., 2021; Pew Research Center, 2008; 2014; Taylor et al., 2003). Consistent with the broader literature (Koenig et al., 2012), African American religious involvement is associated with many positive psychological outcomes (Hays & Aranda, 2016; Nguyen, 2020). This suggests that religion may have a similarly salutary effect on anxiety. Yet, relatively few studies have examined the relationship between religion and anxiety among African Americans, which may contribute to the mixed outcomes in the literature (Hopkins & Shook, 2017).

For instance, service attendance was unrelated to anxiety symptoms among a sample of Chicago-based African Americans (Sternthal et al., 2012). Religious coping also failed to exhibit a protective effect on anxiety symptoms among a sample of African American college students (Chapman & Steger, 2010), a finding largely replicated among a community sample of Black adults (Richards et al., 2016). Conversely, in a study of African American veterans with post-traumatic stress disorder (PTSD), religious involvement was inversely associated with PTSD symptoms, a finding that suggests that religious involvement may play a protective role (Koenig et al., 2018). Similarly, meaning and interpersonal forgiveness were inversely related to anxiety symptoms among the above-mentioned Chicago-based sample (Sternthal et al., 2012).

In considering the extant literature, at least two considerations should be noted. First, the above studies use different, sometimes incompatible measures of religion and feature diverse samples of varying size and quality. As a result, it is difficult to make comparisons across studies or to generalize the findings to the larger population of African Americans (Babbie, 2016). Another consideration is the use of different dependent measures (e.g., overall anxiety symptoms, PTSD symptoms). Given the substantial impact of clinical anxiety on American Americans, more research is needed that features clinical measures of anxiety (Bonelli & Koenig, 2013).

The present study addresses both issues. To enhance the utility of the results, we measure multiple dimensions of religious involvement, including organized religious engagement (i.e., service attendance), non-organized involvement (e.g., reading religious materials), subjective religious engagement, and participation in religious coping strategies, in tandem with the use of a nationally representative sample of community-dwelling African Americans. In addition, the study uses a structured diagnostic interview to assess the common anxiety disorders within two timeframes: the past 12-months and lifetime. In the following section, key models are described that guide our expectations regarding the study findings.

Conceptual Models and Expectations

Several models have been proposed to explain the multiple and sometimes divergent effects of religion on health (Ellison & Levin, 1998). The Prevention Model specifies that religious beliefs, experiences and behaviors reduce the risk of problems in many areas in which life challenges are commonly encountered. In other words, religious involvement acts as a protective factor, functioning to mitigate risk.

This framework suggests an inverse relationship exists between specific measures of religious involvement and the likelihood of having an anxiety disorder. Drawing on religious capital theory, it is plausible that organized religious involvement, manifested in consistent weekly service attendance, may be inversely related to the probability of having an anxiety disorder (Scott et al., 2018). Regular service attendance positions attendees to build up stores of religious capital—such as a network of caring individuals who share a divine perspective on life—that in turn may help mitigate the possibility of developing an anxiety disorder (Chatters et al., 2018).

The Resource Mobilization or Stressor Response Model represents another critical perspective on the relationship between religion and health (Nguyen, 2020). This model posits that major life problems elicit a religious response. In other words, religious resources are mobilized to deal with challenges or stressors, such as an anxiety disorder (Pargament et al., 2013).

This perspective suggests that some measures of religious involvement may be associated with a greater likelihood of having an anxiety disorder. Given their more flexible nature, it is conceivable that some forms of non-organized religious involvement might fall under this rubric. For example, African Americans diagnosed with an anxiety disorder might be more likely to read religious material to understand and deal with the challenges associated with the disorder from a transcendent perspective. In the following section, the methods used to operationalize the study are described.

Method

Data

The study data were drawn from the National Survey of American Life: Coping with Stress in the 21st Century (NSAL). The NSAL was administered by the Program for Research on Black Americans at the Institute for Social Research at the University of Michigan. Fieldwork for the NSAL was completed by the Institute for Social Research’s Survey Research Center in conjunction with the Program for Research on Black Americans. The NSAL sample has a national multi-stage probability design which consists of 64 primary sampling units (PSUs). Fifty-six of these PSUs overlap substantially with the existing Survey Research Center’s National Sample primary areas. The remaining eight PSUs were chosen from the South, so the sample represents African Americans in the proportion they are distributed in the United States.

Data collection occurred from February 2001 to June 2003. Some 86% of the interviews were administered face-to-face in respondents’ homes, with the remainder being conducted over the phone. Respondents were compensated for their time. A total of 6,082 face-to-face interviews were administered with individuals aged 18 or older, including 3,570 African Americans, 1,621 Blacks of Caribbean descent, and 891 non-Hispanic Whites. Given cultural differences between these three groups, and our desire to focus on African Americans, the present study uses the African American sub-sample. The overall response rate was 72.3%, and the response rate was 70.7% for African Americans. 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 samples) (AAPOR, 2006). Respondents provided written informed consent regarding their participation in the study, and all survey procedures were approved by the institutional review board at the University of Michigan. Further information about the NSAL is available elsewhere (Jackson et al., 2004).

Measures

Independent Variables

Religious Involvement.

The eight measures of religious involvement investigated in this study included measures of organizational, nonorganizational, and subjective religious engagement, as well as religious coping. The measure of organizational religious participation is frequency of service attendance. This variable is measured by combining two items—one that indicates the frequency of attendance and one that identifies respondents who have not attended services since the age of 18. The resulting categories for service attendance are: attend nearly everyday, at least once a week, a few times a month, a few times a year, less than once a year, and (except for weddings and funerals) never attended services since the age of 18.

Four measures of nonorganizational religious participation are used in this analysis: 1) reading religious books or other religious materials, 2) watching religious television programs, 3) listening to religious radio programs on the radio, and 4) praying. Respondents were asked the frequency with which they engaged in these activities (i.e., nearly everyday, at least once a week, a few times a month, at least once a month, a few times a year or never). The range of each item was 6 for nearly everyday to 1 for never.

Subjective religious involvement is measured by an item assessing respondents’ self-rating of religiosity: “How religious would you say you are?” This item had four categories ranging from 4 (very religious) to 1 (not religious at all). Two measures of religious coping were also included in the study. The first provides an assessment of the significance of prayer in difficult circumstances and asks: “How important is prayer when you deal with stressful situations?” very important (4), fairly important (3), not too important (2), or not important at all (1). The second question reflects an overall orientation toward God as a resource and asks respondents’ level of agreement with the following statement: “I look to God for strength, support, and guidance.” Respondents indicate whether they: strongly agree (4), somewhat agree (3), somewhat disagree (2), or strongly disagree (1) with this statement.

Dependent Variables

Five different anxiety disorders were included as dependent variables: Panic Disorder (PD), Agoraphobia without panic (AGO), Social Phobia, formerly referred to as Social Anxiety Disorder (SAD), PTSD, and Generalized Anxiety Disorder (GAD). The DSM-IV World Mental Health Composite International Diagnostic Interview (WMH-CIDI)—a fully structured diagnostic interview—was used to assess each anxiety disorder’s 12-month and lifetime prevalence (Kessler & Ustun, 2004). Twelve-month anxiety disorder denotes having an anxiety disorder within the past year, while lifetime anxiety disorder refers to having an anxiety disorder at any time during a respondent’s life.

Control Variables.

The demographic variables included in the analysis were age, gender, marital status, education, household income, geographic region, and self-rated health. Age is represented by three categories: 18–34, 35–54 and 55 and older. Gender is coded as male = 0 and female = 1. Marital status is represented by two categories: married or cohabiting, and non-married. Education has four categories: 0–11 years, 12 years, 13–15 years, and 16 and more years of formal education. Household income has four categories (0–14,999, 15,000–27,999, 28,000–46,999, ≥47,000). Geographic region is represented by two categories (South and Other Regions). Self-rated physical health was assessed with the following 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?” The distribution of the study variables is presented in Table 1.

Table 1.

Demographic Characteristics of the Sample and Distribution of Study Variables

	%	N	Mean	S.D.	Range
12-Month Panic Disorder	2.32	93
Lifetime Panic Disorder	3.45	132
12-Month Agoraphobia	1.09	44
Lifetime Agoraphobia	2.08	77
12-Month Social Phobia	4.55	163
Lifetime Social Phobia	7.58	258
12-Month PTSD	3.80	135
Lifetime PTSD	9.10	313
12-Month GAD	2.49	85
Lifetime GAD	4.47	159
Organized Religious Involvement
Church Attendance
Never	8.56	267
Less that Once Per Year	9.63	312
Few Times Per Year	19.53	670
Few Times Per Month	24.34	891
At Least Once a Week	32.60	1226
Nearly Everyday	5.32	204
Non-organized Religious Involvement
Reading Religious Materials		3569	4.18	1.44	1–6
Watch Religious Television		3569	3.77	1.55	1–6
Listen to Religious Radio		3569	3.70	1.75	1–6
Prayer		3567	5.60	0.95	1–6
Subjective Religious Involvement
Self-Rated Religiosity		3556	3.13	0.68	1–4
Religious Coping
Importance of Prayer in Stressful Situations		3562	3.86	0.43	1–4
Look to God for Strength		3434	3.85	0.44	1–4
Age
18–34	35.73	1232
35–54	42.65	1501
≥55	21.62	837
Gender
Male	44.03	1271
Female	55.97	2299
Marital Status
Married/Partner	41.65	1220
Not Married	58.35	2333
Education
0–11	24.19	920
12	37.86	1362
13–15	23.83	809
≥16	14.12	479
Annual Household Income
0–14,999	24.70	1054
15,000–27,999	23.76	925
28,000–46,999	25.50	866
≥47,000	26.05	725
Region
South	56.24	2330
Other Regions	43.76	1240
Self-rated health		3437	3.42	0.95	1–5

Note. Percents and N are presented for categorical variables and Means and Standard Deviations are presented for continuous variables. Percents are weighted; frequencies are unweighted.

Analytic Strategy

All frequencies are unweighted, while all percentages are weighted based on the distribution of African Americans in the population. For the multivariate analyses, logistic regression was used. Odds ratio estimates and 95% confidence intervals are presented. All analyses were conducted using SAS, which uses the Taylor expansion approximation technique for calculating the complex design-based estimates of variance. All statistical analyses accounted for the complex multi-stage clustered design of the NSAL sample, unequal probabilities of selection, nonresponse, and poststratification to calculate weighted, nationally representative population estimates and standard errors.

Results

Table 2 presents the results of the logistic regression analysis of religious involvement and DSM-IV 12-month and lifetime panic, agoraphobia and social phobia disorders. Religious service attendance, reading religious materials, and listening to religious radio programs were significantly associated with having a panic disorder within the past 12-months. African Americans who attended religious services at least once a week and a few times a month were less likely to have 12-month panic disorder compared to the reference group, those who attended services less than once per year. Reading religious materials was positively associated with 12-month panic disorder, whereas listening to religious radio programs was negatively associated with panic disorder. Regarding lifetime disorders, African Americans who never attended religious services and those who attended once per week had a lower likelihood of having a panic disorder during their lifetimes. Also, African Americans who looked to God for strength had a lower likelihood of having a lifetime panic disorder.

Table 2.

Logistic Regressions of Religious Involvement and 12-Month and Lifetime Panic Disorder, Agoraphobia and Social Phobia

	Panic Disorder (N = 3,403)		Agoraphobia/without Panic (N = 3,403)		Social Phobia (N = 3,412)
	12-Month	Lifetime	12-Month	Lifetime	12-Month	Lifetime
	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)
Organized Religious Involvement
Service Attendance
Never	0.24 (0.06–1.02)	0.24 (0.06–0.93)*	1.03 (0.21–5.00)	1.25 (0.42–3.71)	3.17 (1.20–8.42)*	1.36 (0.68–2.69)
Less than Once Per Yeara	1.0	1.0	1.0	1.0	1.0	1.0
Few Times Per Year	1.05 (0.60–1.85)	0.97 (0.55–1.72)	0.61 (0.15–2.59)	0.56 (0.20–1.56)	3.39 (1.06–10.78)*	1.74 (0.94–3.20)
Few Times Per Month	0.49 (0.28–0.86)*	0.70 (0.37–1.32)	0.46 (0.11–1.97)	0.45 (0.13–1.52)	2.93 (1.02–8.38)*	1.33 (0.79–2.24)
At Least Once a Week	0.20 (0.09–0.49)†	0.26 (0.13–0.55)†	0.26 (0.05–1.52)	0.36 (0.13–0.96)*	1.98 (0.59–6.68)	1.27 (0.74–2.18)
Nearly Everyday	0.44 (0.12–1.66)	0.48 (0.16–1.44)	0.38 (0.05–3.07)	0.40 (0.09–1.77)	2.09 (0.46–9.41)	1.24 (0.57–2.69)
Non-organized Religious Involvement
Reading Religious Materials	1.27 (1.04–1.54)*	1.20 (0.97–1.48)	1.37 (1.07–1.75)*	1.11 (1.00–1.24)	1.08 (0.93–1.27)	1.03 (0.90–1.17)
Watch Religious TV Programs	1.07 (0.83–1.37)	1.01 (0.82–1.24)	1.41 (1.02–1.95)*	1.06 (0.86–1.31)	1.05 (0.92–1.20)	1.02 (0.94–1.12)
Listen to Religious Radio Programs	0.89 (0.79–1.00)*	0.99 (0.87–1.11)	1.03 (0.85–1.27)	1.15 (0.98–1.36)	0.97 (0.88–1.07)	1.01 (0.93–1.10)
Prayer	1.10 (0.78–1.56)	1.14 (0.88–1.47)	0.62 (0.41–0.94)*	0.77 (0.61–0.98)*	0.79 (0.65–0.96)*	0.87 (0.72–1.04)
Subjective Religious Involvement
Self-rated Religiosity	1.04 (0.72–1.50)	1.17 (0.87–1.56)	1.06 (0.64–1.76)	1.36 (0.89–2.08)	1.04 (0.69–1.58)	0.93 (0.70–1.24)
Religious Coping
Imp. of Prayer in Stressful Situations	1.38 (0.68–2.79)	1.29 (0.78–2.11)	>999 (>999->999)1	1.55 (0.42–5.81)	1.05 (0.65–1.71)	1.16 (0.78–1.73)
Look to God for Strength	0.62 (0.35–1.09)	0.60 (0.39–0.92)*	1.05 (0.37–2.97)	1.58 (0.55–4.57)	1.20 (0.78–1.83)	1.05 (0.73–1.53)

OR=Odds Ratio, CI=Confidence Interval, Analysis controls for age, gender, marital status, education, household income, region and self-rated health.

^aReference Category.

^*p < 0.05;

^#p < 0.01;

^†p < 0.001

¹Estimates are unstable and not discussed in the paper.

Concerning agoraphobia, the frequency of reading religious materials and watching religious television programming was positively associated with 12-month agoraphobia, whereas the frequency of prayer was negatively associated. As to lifetime agoraphobia, African Americans who attended religious services at least once per week were less likely to have been diagnosed during their lives. In addition, the frequency of prayer was negatively associated with lifetime agoraphobia.

In terms of social phobia, religious service attendance and the frequency of prayer were significantly associated with a 12-month social phobia. African Americans who attended religious services a few times a month, a few times a year, and never had greater odds of having 12-month social phobia, compared to the reference group. The frequency of prayer was negatively associated with 12-month social phobia. None of the religious involvement variables were significantly related to lifetime social phobia.

Table 3 depicts the results for PTSD and generalized anxiety disorder. Regarding the former, only service attendance was related to 12-month PTSD; respondents who attended religious services at least once per week were less likely to have 12-month PTSD relative to members of the reference group. Service attendance, reading religious materials, prayer, and self-rated religiosity were associated with lifetime PTSD. Respondents who attended religious services at least once per week and a few times a month were less likely to have lifetime PTSD than those who attended less than once per year. Reading religious materials was positively associated with lifetime PTSD whereas, prayer, and self-rated religiosity were negatively associated with lifetime PTSD.

Table 3.

Logistic Regressions of Religious Involvement and 12-Month and Lifetime PTSD and Generalized Anxiety Disorder

	PTSD (N = 3,391)		Generalized Anxiety Disorder (N = 3,402)
	12-Month	Lifetime	12-Month	Lifetime
	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)
Organized Religious Involvement
Service Attendance
Never	1.18 (0.42–3.30)	0.86 (0.43–1.72)	0.61 (0.15–2.47)	1.43 (0.51–4.00)
Less than Once Per Yeara	1.0	1.0	1.0	1.0
Few Times Per Year	1.19 (0.56–2.52)	0.85 (0.52–1.37)	1.06 (0.35–3.20)	1.33 (0.55–3.21)
Few Times Per Month	0.53 (0.21–1.33)	0.56 (0.32–0.98)*	0.49 (0.20–1.19)	0.82 (0.41–1.65)
At Least Once a Week	0.36 (0.13–0.99)*	0.48 (0.30–0.76)#	0.46 (0.15–1.40)	0.60 (0.26–1.42)
Nearly Everyday	0.75 (0.23–2.40)	0.77 (0.41–1.43)	0.69 (0.15–3.16)	1.05 (0.38–2.92)
Non-Organized Religious Involvement
Reading Religious Materials	1.18 (0.87–1.61)	1.21 (1.05–1.39)*	1.23 (0.93–1.63)	1.22 (1.00–1.48)*
Watch Religious Television Programs	1.02 (0.86–1.21)	1.01 (0.90–1.13)	1.02 (0.80–1.31)	1.01 (0.85–1.19)
Listen to Religious Radio Programs	1.08 (0.95–1.24)	1.05 (0.95–1.15)	0.94 (0.80–1.09)	1.02 (0.90–1.16)
Prayer	0.93 (0.67–1.29)	0.84 (0.70–1.00)*	0.84 (0.58–1.22)	0.93 (0.71–1.22)
Subjective Religious Involvement
Self-rated Religiosity	0.83 (0.63–1.10)	0.77 (0.64–0.93)#	0.84 (0.59–1.19)	0.80 (0.63–1.03)
Religious Coping
Importance of Prayer in Stressful Situations	1.62 (0.85–3.10)	1.47 (0.96–2.25)	1.47 (0.59–3.67)	1.17 (0.63–2.17)
Look to God for Strength	0.85 (0.51–1.43)	1.10 (0.78–1.55)	2.08 (0.57–7.51)	2.18 (0.92–5.12)

OR=Odds Ratio, CI=Confidence Interval, Analysis controls for age, gender, marital status, education, household income, region and self-rated health.

^aReference Category.

^*p < 0.05;

^#p < 0.01;

^†p < 0.001

Lastly, reading religious materials was positively associated with lifetime generalized anxiety disorder. No other religious involvement variables were significantly associated with either 12-month of lifetime generalized anxiety disorder.

Discussion

This study examined the relationship between religious involvement and anxiety disorders among African Americans. To the best of our knowledge, it is the first investigation of the effects of multiple, conceptually distinct measures of religious involvement on frequently occurring DSM-IV anxiety disorders among a national sample of African Americans. Given the debilitating effects of clinical anxiety disorders on African Americans and the paucity of research on this topic, this study makes a critical contribution to the literature (Hopkins & Shook, 2017).

In keeping with our expectations, weekly service attendance was associated with a lower likelihood of 12-month and lifetime panic disorder, lifetime agoraphobia, and 12-month and lifetime PTSD. Of the measures of non-organized religious involvement, prayer was associated with a lower probability of 12-month agoraphobia, 12-month social phobia, and lifetime PTSD. Likewise, listening to religious radio programs was inversely related to 12-month panic order. Regarding coping strategies, looking to God for strength was inversely related to lifetime panic disorder.

These findings are consistent with theory and prior, related research examining the effects of religion on mental health among African Americans. A growing body of research has linked African American religious involvement with positive health outcomes (Hays & Aranda, 2016; Nguyen, 2020). For example, previous research using NSAL data recorded an inverse relationship between service attendance and 12-month and lifetime major depressive disorder (Taylor et al., 2012), as well as suicide ideation and attempts (Taylor et al., 2011). Similarly, researchers using the Add Health Surveys documented an inverse relationship between service attendance and substance use (Hodge et al., 2021).

These findings are in concert with the Prevention Model. As noted in the introduction, this model specifies that religious involvement acts as a protective factor by, for example, engendering positive cognitive scripts, mitigating involvement in detrimental behaviors, and fostering the development of salutary social networks (Ellison & Levin, 1998). The results of the present study suggest this model is also appliable to anxiety disorders.

The results also suggest the protective effects of religious involvement are most acute for those who attend religious services at least once a week. While weekly attendance was often inversely related to anxiety disorders, other attendance patterns exhibited inconsistent relationships with the dependent measures. Longitudinal research examining the effects of different religious attendance patterns on substance use has found that consistent, weekly attendance exhibits the most pronounced protective effect. This dynamic has been observed among samples drawn from the general population (Guo & Metcalfe, 2019; Zhang et al., in press), as well as among African Americans (Hodge et al., 2021). These results may be explained by religious capital theory, which builds on Putnam’s (2000) social capital work. Individuals who attend services consistently over time build up stores of religious capital (e.g., salutary scripts about the nature of life) that can be operationalized to deal with life challenges.

This dynamic may be especially relevant with certain types of anxiety. For instance, it is plausible that African Americans coping with panic disorder, agoraphobia, and social phobia may experience psychological distress that impacts the frequency with which they attend services. Individuals coping with panic disorder may avoid being in a location, such as a congregational setting, where they might be likely to have another panic attack while those dealing with agoraphobia may be less likely to leave their home to attend services. Similarly, those with social phobia may avoid social circumstances, such as religious services, where they may encounter and be expected to interact with other people. This may help explain why infrequent attendance was associated with greater odds of having 12-month social phobia. Being an unfamiliar environment where one does not have supportive social connections may facilitate or exacerbate symptoms.

Conversely, attending services weekly may serve to plug African Americans into supportive horizontal and vertical social networks that help mitigate anxiety (Chatters et al., 2018). Regular interactions with other empathetic individuals in a context where transcendent reality is emphasized may help alleviate distress and foster more salutary perspectives (Scott et al., 2018). In addition to horizontal support from caring others, service participation may reinforce African Americans’ relationship with God. Focusing on a transcendent perspective that is personally meaningful may help alleviate symptoms of distress (Steffen et al., 2017). The horizontal and vertical support may have a synergistic effect on symptoms, but only in the context of regular attendance. In other words, the positive effects may only manifest among those who have accumulated sufficient stores of religious capital via consistent, engaged participation. This perspective may explain the observed relationship between weekly service attendance and the lower likelihood of 12-month and lifetime panic disorder, and lifetime agoraphobia, as well as 12-month and lifetime PTSD.

It should also be mentioned one measure of non-organized religious involvement exhibited a positive relationship with anxiety disorders. Reading religious materials was associated with a greater probability of having 12-month panic disorder, 12-month agoraphobia, lifetime PTSD, and lifetime generalized anxiety disorder. These findings are consonant with the Resource Mobilization or Stressor Response Model (Ellison & Levin, 1998). Having an anxiety disorder mobilizes a religious response—in this case reading religious content—as a pathway for understanding and coping with the disorder. Religious assets are marshaled to deal with the anxiety disorder (Taylor et al., 2011).

This perspective is congruent with prior research, including an investigation of how veterans cope with anxiety symptoms. This study found that veterans frequently engaged in self-management strategies—such as scripture reading—to deal with stress and anxiety (Shepardson et al., 2017). Subsequent research suggests that scripture reading helps buffer the effects of stress (Krause & Pargament, 2018). It is possible that African Americans use religious reading as a proactive strategy to: hasten remission from current episodes of anxiety, heal from past episodes, and prevent future episodes from occurring (Taylor et al., 2012). Longitudinal studies, however, are needed to clarify the role that religious reading may play in dealing with anxiety disorders. Several implications stem from the findings for professionals who work with African Americans dealing with anxiety disorders, although the cross-sectional nature underscores the need for caution when considering the following suggestions.

Implications

Religious involvement was related to all five anxiety disorders examined in this study. Consequently, clinicians working with African Americans should conduct a spiritual assessment to understand the role of religion in the client’s life and its intersection with the presenting problem (Hodge, 2015). When clinically warranted, clinicians might consider interventions that integrate clients’ religious beliefs and behaviors into therapeutic strategies when clients are interested in and supportive of such approaches. A small but growing body of outcome research indicates that such interventions are at least as effective as standard interventions (Gonçalves et al., 2015). Religiously adapted interventions have been shown to be effective with many outcomes, including anxiety among African Americans (Hays & Aranda, 2016).

The assessment might also explore the existence of currently employed religious self-management strategies and other strategies that clients might be interested in implementing. In addition to religious practices that clients might implement outside an organized religious context, clinicians might explore potential resources that might exist in churches and other organized religious settings (Chatters et al., 2018). Clients overwhelmed with anxiety may overlook resources that might be leveraged to deal with anxiety symptoms. However, it is important to note that the study implications, in conjunction with the findings, should be understood within the context of the study’s limitations.

Limitations

Although the study employed a nationally representative sample, the results cannot be generalized to all African Americans. Certain subgroups, such as those who were homeless or institutionalized, were not included in the sampling frame (Babbie, 2016). It likely that some African Americans with anxiety disorders are members of the subgroups that fell outside the study’s sampling parameters.

Another factor that limits generalizability is the age of the dataset. The NSAL was collected in the 2001–2003 timeframe using DSM-IV criteria. It is possible that the associations observed in the present study may not be generalizable to the current population of African Americans using DSM-V measures of anxiety. Concurrently, it is important to note that methodologically rigorous ground-breaking datasets, such as the NSAL, continue to be widely used in current research. The NSAL is part of the Collaborative Psychiatric Epidemiology Surveys (CPES). The CPES consists of three nationally representative surveys: the NSAL, the National Comorbidity Survey Replication (NCS-R), and the National Latino and Asian American Study (NLAAS). All three CPES studies shared identical mental health questions with regards to various psychiatric disorders, functional impairments, and help-seeking. They also were administered during the same period (2001–2003) by the same survey research organization (i.e., the Institute for Social Research at the University of Michigan).

Data from these surveys continues to be widely used. For instance, the NCS-R has been used in recent studies (Chasson et al., 2022; Legg & Turner, 2021; Mueller et al., 2022; Stickley et al., 2021), including those featuring anxiety disorders (Hazzard et al., in press; Stapp et al., 2022). Similarly, the NSAL also continues to be used in contemporary research (Nicholson & Wheeler, 2021; Sullivan et al., 2021; Whaley, 2021) including studies incorporating anxiety disorders (Jones et al., 2020; Oh et al., 2021; Oh et al., in press) psychotic experiences (Oh & Anglin, in press) and depression (Robinson et al., in press).

It might also be noted that the NSAL may be the only dataset based on a national probability sample that has a large number of African Americans, a large number of religion items, and DSM disorders. Indeed, to the best of our knowledge as scholars who work in this general area, this is the first nationally representative study of the effects of multiple, conceptually distinct measures of religious involvement on clinical anxiety disorders among African Americans. Moreover, there is little reason to believe the underlying mechanisms linking religious involvement to anxiety have changed substantially over the intervening years.

Another issue that should mentioned concerns ceiling effects. Consistent with the findings in this study, African Americans have long exhibited some of the highest levels of religious engagement of any population (Gallup & Castelli, 1989; Hodge et al., 2021; Mohamed et al., 2021). The lack of variability on some measures of religious involvement may attenuate the observed associations.

The cross-sectional design precludes any definitive assessment regarding causality between religion and anxiety disorders. Although the results are consistent with relevant theory and research, it is impossible to rule out alternative explanations of the findings. As noted above, further research that employs longitudinal designs is needed to clarify the relationship between religious involvement and anxiety disorders among American Americans.

Conclusion

This study contributes to our understanding of the relationship between religion and anxiety among African Americans. Key strengths of the study include the use of multiple, conceptually distinct measures of religious involvement, clinically valid measurement of the most commonly occurring anxiety disorders, and the use of a nationally representative sample. As such, this study may provide our first clear understanding of the relationship between various forms of religious engagement and five different anxiety disorders.

Additional research is needed to clarify the specific pathways, but it is clear certain measures of religious involvement are associated with various anxiety disorders. Given the salience of religion among African Americans, the findings underscore the need for clinicians to explore the role that religion plays in their clients’ lives, especially as it intersects anxiety symptoms. This study highlights potential religious resources might be leveraged to assist African Americans in preventing, recovering from, and coping with anxiety.