Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Dec 1.
Published in final edited form as: J Relig Health. 2020 Dec;59(6):3055–3070. doi: 10.1007/s10943-019-00888-6

Investigating Denominational and Church Attendance Differences in Obesity and Diabetes in Black Christian men and Women

Keisha L Bentley-Edwards 1, Loneke T Blackman Carr 2, Paul A Robbins 3, Eugenia Conde 4, Khaing Zaw 5, William A Darity Jr 6
PMCID: PMC6986996  NIHMSID: NIHMS1535940  PMID: 31359241

INTRODUCTION

Obesity affects nearly one-third of men and women in the United States (U.S.) (Flegal, Kruszon-Moran, Carroll, Fryar, & Ogden, 2016). Recent examination of obesity prevalence from the National Health and Nutrition Examination Survey (NHANES) indicates that African Americans (blacks) are disproportionately affected - almost half are obese (48.4%) (Flegal et al., 2016). Furthermore, compared to other racial and ethnic groups, blacks have the highest prevalence of class 3 (severe) obesity indicated by a body mass index (BMI) ≥ 40 (Flegal et al., 2016).

Obesity increases the risk of developing chronic diseases like cardiovascular disease and diabetes (American Diabetes Association, 2007; Knowler et al., 2003). It is the primary reason for the above average diabetes risk in blacks, attributed to almost 50% of that increased risk (Brancati, Kao, Folsom, Watson, & Szklo, 2000; Chatterjee et al., 2013; Hu et al., 2001). Diabetes is most prevalent among Blacks (Centers for Disease Control and Prevention, 2017a). Nationally diabetes is the seventh leading cause of death (Centers for Disease Control and Prevention, 2017a), but among blacks alone, it is the fifth (Heron, 2018). The current study investigates obesity and diabetes and the relationship with religious denomination among black adults in the United States.

Religion, Obesity & Diabetes

Religion and its practice can shape the overarching moral and more minute choices in daily life. Though spiritual and moral righteousness may be a central facet of religion, physical health is also influenced. Obesity and diabetes represent major health outcomes where religion plays a role. Religious engagement has a varied relationship with BMI in the literature. Religious participation, which includes attendance among other variables, has been associated with greater BMI among a sample of men and women combined (Bruce, 2007). However, when examined by race and gender group, black women emerged as the sole race-gender group where the religious participation-BMI association was significant (Bruce, 2007). Thus gender and race may be important factors when discussing the impact of religion on weight. The distribution of body weight also becomes important in this discussion as abdominally centralized body fat increases the risk of cardiovascular disease (Czernichow, Kengne, Stamatakis, Hamer, & Batty, 2011; Dalton et al., 2003; Zhu et al., 2005), but studies of domestic and international samples indicate no relationship between waist circumference nor waist-to-hip ratio and religious service attendance (Das & Nairn, 2016; Hill, Rote, & Ellison, 2017).

Religious denomination has also been shown to impact weight (Cline & Ferraro, 2006; Ferraro, 1998; Kim, Sobal, & Wethington, 2003; Yeary, Sobal, & Wethington, 2017). In the U.S., states with a greater proportion of individuals who claimed a denomination had higher rate of obesity, this was especially true for those who affiliated with the Baptist faith tradition (Ferraro, 1998). Denominational differences in obesity rates have appeared throughout the literature (Kim et al., 2003; Yeary et al., 2017). Conservative Protestants have presented with a BMI that was 1.1 higher than individuals who do not claim a denomination when only religious denomination was considered (Kim et al., 2003). Although this relationship did not hold when other health behaviors and religious practice variables were included in the model (Kim et al., 2003), denomination may play a role in weight status. Using national data, Cline, et al. investigated the relationship between religion and the prevalence of obesity (Cline & Ferraro, 2006). Across multiple Christian and non-Christian denominations, authors found the prevalence of obesity to be highest among Baptists (Cline & Ferraro, 2006). Further, they observed that blacks were more likely to be obese and more likely to be affiliated with the Baptist denomination than whites (Cline & Ferraro, 2006).

While Cline, et al indicated that being Baptist was related to greater obesity prevalence, and that race is an important factor in that finding, their study does not establish how obesity rates differ across denominations for blacks specifically. A recent systematic review of the literature concerning religion and weight found that Seventh-Day Adventists (Adventists), compared to non-Adventists, had the lowest body weight (Yeary et al., 2017). Studies of health outcomes among Adventists, a Protestant denomination whose doctrine mandates abstinence from alcohol, tobacco, and pork consumption, and prescribes a vegetarian diet, reveal links between religion and health.

A longitudinal study of members of the Seventh-Day Adventist Church examined the relationship between the varied adherence to a vegetarian diet and the development of chronic diseases (Fraser 1999). At baseline, obesity was higher among Adventists who ate meat (semi-vegetarian or non-vegetarian dietary pattern) (Fraser, 1999). The Adventist Health Study-2, which included a sample of black Adventists sufficient to draw statistical conclusions, found obesity to be more prevalent among blacks compared against whites (Montgomery, 2007). This may be due to blacks entry into the Adventist faith later in life than whites, leading to later adoption of the dietary pre- and proscriptions and later intergenerational uptake of Adventist practices (Montgomery, 2007).

Similar to obesity, diabetes also appears to bear a relationship to religious affiliation. A longitudinal study of California Adventists, found diabetes prevalence was almost two times greater in non-vegetarians versus vegetarians (Fraser, 1999). The Adventist Health Study-2 also found diabetes to be more prevalent among blacks relative to whites, again attributing this finding to late uptake of the Adventist faith, similar to obesity outcomes (Montgomery, 2007). However, black Adventist outcomes were better than black non-Adventists (Montgomery, 2007). When considering religious participation, including attendance, studies that used the diabetes biomarker hemoglobin A1c (HbA1c) as an outcome found no relationship to participation (Hill et al., 2017; Hill, Rote, Ellison, & Burdette, 2014), a finding seen in other literature (Das & Nairn, 2016). Much of the work in this area included multiracial and ethnic samples predominantly composed of non-Hispanic whites. Thus providing little insight on the role of religion on obesity and diabetes in black Christians.

Religion & Health in Blacks

Blacks are the most highly religious group in the U.S., being more likely to believe in God, consider religion important, attend church almost weekly, and engage in prayer and scripture reading than other racial and ethnic groups (Pew Research Center, 2014). In black American life, the black church serves many functions, representing a central institution, regarded as “having no challenges in the cultural womb of the Black community” (Lincoln, Mamiya, & Duke University, 1990). The black church provides a site for the practice and determination of worship, faith and morality, but it also engages in the provision of community and social needs, including various social services, education, as well as acting as a political and financial center (Billingsley, 1991). Matters of health also fall under the purview of the black church; the black church has engaged in the delivery of community medicine and primary care and has participated in health promotion and disease prevention efforts (Levin, 1984; Lincoln et al., 1990).

Researchers have long understood the centrality of religion and spirituality in black life, and have sought to reach the black community through the church for the purpose of health promotion (Campbell et al., 2007). The black church has been the location of several health behavior interventions targeting fruit and vegetable intake, reduced diabetes risk through obesity treatment, and increased physical activity (Dodani, Kramer, Williams, Crawford, & Kriska, 2009; McNabb, Quinn, Kerver, Cook, & Karrison, 1997; Resnicow et al., 2004).

While these studies utilized a community-based approach to partner with churches to improve health, they did not distinguish between denominations in their design, nor account for denominational specificity in their findings (Dodani et al., 2009; McNabb et al., 1997; Resnicow et al., 2004). Religious dietary pre- and proscriptions are acknowledged (Campbell et al., 2007), but denomination is not factored into health promotion approaches. While religious denomination might influence the occurrence of obesity and diabetes, it remains an understudied factor in black health.

The Current Study

Considering the role of religion in health and black life, and the consistent involvement of the black church in health promotion, we sought to fill the evidence gap concerning health differences by denomination. Data from the National Survey of American Life (NSAL) (Jackson et al., 2004) presents an opportunity to examine denominational differences in disease outcomes among black Christians. The NSAL includes a nationally representative sample of blacks in the U.S., which facilitates a nuanced examination of black health. Using data from the NSAL, we examine denominational and religious service attendance differences in obesity and diabetes among blacks, while also considering the influence of socioeconomic factors.

METHODS

Data

The National Survey of American Life (NSAL) is a cross-sectional study purposed to understand intra- and inter-group racial and ethnic differences in a wide spectrum of health outcomes. Data on health, religion, and social conditions were collected in the NSAL, which was administered by the Program for Research on Black Americans by the Institute for Social Research at the University of Michigan from 2001-2003. This data set is appropriate for our investigation because the NSAL includes a nationally representative sample of blacks, which enables examination of the role of religion in the lives and health of this sub-population – important given their high religiosity.

In sum, the NSAL data includes 3,570 African Americans, 1,438 Afro-Caribbeans, 891 non-Hispanic Whites (Jackson et al., 2004). Most data were collected through in-home interviews by trained study personnel, with about 14% conducted either entirely or partially by telephone (Jackson et al., 2004). Given the focus of the present study, to investigate religion and health among blacks in the United States, we included only non-Hispanic African Americans (n= 3,570) and Afro-Caribbeans (n=1,438) from the NSAL sample. See Table 1 for detailed demographic information on the participants.

Table 1.

Sociodemographic characteristics of the sample

Total n = 4,344 Women n = 2,852 Men n = 1,492
Variable n (%) n (%) n (%)
Denomination
 Baptist 2,119 (48.8) 1,424 (49.9) 695 (46.6)
 Methodist 276 (6.4) 176 (6.2) 100 (6.7)
 Pentecostal 282 (6.5) 215 (7.5) 67 (4.5)
 Presbyterian 41 (0.9) 31 (1.1) 10 (0.7)
 Protestant (other) 1,121 (25.8) 688 (24.1) 433 (29.0)
 Catholic 505 (11.6) 318 (11.2) 187 (12.5)
Religious Service Attendance
 Never or < once/year 569 (13.1) 195 (6.8) 71 (4.8)
 A few times/year 821 (18.9) 1,183 (41.5) 443 (29.7)
 A few times/month 1,062 (24.5) 705 (24.7) 357 (23.9)
 At least once/week 1,626 (37.4) 461 (16.2) 360 (24.1)
 Nearly every day 266 (6.1) 308 (10.8) 261 (17.5)
Age (mean, SD) 43.5 (16.2) 43.4 (16.2) 43.6 (16.2)
Household Income (mean, SD) 34,258.10 (29,957.2) 30,955.37 (27,624.16) 40,571.33 (33078.14)
Education (years)
 0-11 1,026 (23.6) 674 (23.6) 352 (23.6)
 12 1,548 (35.6) 1,000 (35.1) 548 (36.7)
 13-15 1,054 (24.3) 697 (24.4) 357 (23.9)
 16 or greater 716 (16.5) 481 (16.9) 235 (15.8)
Employment Status
 Employed 2,923 (67.4) 1,857 (65.1) 1,066 (71.5)
 Unemployed 415 (9.6) 300 (10.5) 115 (7.7)
 Not in labor force 997 (23.0) 691 (24.2) 306(20.5)
Ethnicity
 African American 3,134 (72.2) 2,077 (72.8) 1,057 (70.8)
 Afro-Caribbean 1,210 (27.9) 775 (27.2) 435 (29.2)
Geographic Region
 Northeast 1,157 (26.63) 773 (27.1) 384 (25.7)
 Midwest 513 (11.81) 344 (12.1) 169 (11.3)
 South 2,462 (56.68) 1,603 (56.2) 859 (57.6)
 West 212 (4.88) 132 (4.6) 80 (5.4)
Obesity (BMI ≥ 30) 1,395 (34.2) 1,030 (36.1) 365 (24.5)
Diabetes 483 (11.5) 340 (11.9) 1,43 (9.6)
Open in a new tab

SD = Standard deviation

Measures

Religious Denomination

Denomination was identified via self-report. Participants were asked, “What is your current religion?” Investigators who originally administered the NSAL recoded participant responses in fourteen categories, out of which we included only Christians and recoded into six categories: Baptist, Methodist, Pentecostal, Presbyterian, Protestant (other), and Catholic. The Lutheran category was not included in analyses due to the small sample size. Originally, NSAL investigators captured three categories of Catholic traditions (Roman Catholic, Catholic (other), Catholic (no denomination mentioned)). Due to the small sample size of Catholic (other) and the common thread of Catholicism between the three original categories, they were collapsed into one Catholic category. Non-Christian denominations were not distinctly categorized in the original fourteen categories in the NSAL, and may have been grouped in the ‘Other’ response option. We excluded respondents who selected no religion, agnostic/atheist, or no religious preference. These categories did not allow us to distinguish between respondents who were non-denominational or non-organizational, and thus did not align with our central aim to identify differences between denominations.

Religious Service Attendance

The frequency of religious service attendance indicated varying degrees of religious participation among study participants. Response options for the frequency of religious service attendance originally administered the NSAL were: 1) nearly every day (≥4 days/week), 2) at least once a week (1-3 times/week), 3) a few times a month (1-3 times/month), 4) a few times a year, and 5) less than once a year. We included individuals who never attend service and combined them with the original category of ‘less than once a year’. Thus the frequency of religious service covariate had four categories for analysis: 1) never attend - less than once a year, 2) a few times a year, 3)at least once a week (1-3 times/week) - a few times a month (1-3 times/month), and 4) nearly every day (≥4 days/week).

Obesity

Obesity was one outcome of interest for this study. It was determined by BMI, which was calculated by dividing self-reported weight and height. A participant with a BMI ≥ 30 kg/m2 was considered obese in accordance with categories defined by the Centers for Disease Control and Prevention (Centers for Disease Control and Prevention, 2016). In the current study, the obesity variable was dichotomized, and respondents were categorized as either obese (BMI ≥ 30 kg/m2) or non-obese (BMI < 30 kg/m2).

Diabetes

Diabetes, also an outcome variable, was determined by self-report. Participants responded “yes” or “no” when asked if a health care professional ever said they had diabetes.

Covariates

Demographic covariates included race/ancestry (Afro-Caribbean or African American), U.S. born (yes/no), gender (male or female), and age (years). Socioeconomic covariates included household income (thousands), employment status (employed, unemployed, or not in labor force), and education level (high school diploma or equivalent, less than high school diploma, some college, or college degree). The four geographical regions of U.S. included the Northeast, Midwest, South, and West.

Statistical Analyses

Logistic regressions were used to evaluate obesity and diabetes separately. Denomination, race/ethnicity, demographic, socioeconomic, and religious service attendance variables were used as the covariates in each model for obesity and diabetes. Whether born in the U.S. or not was a covariate only in the obesity model. For the diabetes outcome, geographic region was not included. In a final model, we tested the interaction between gender and religious service attendance to determine whether the association between attendance and obesity varied by gender in the diabetes and obesity models.

The statistical analyses were conducted using Stata 15.1. Due to the multi-stage area probability sample designs for household surveys, probability weights were used in the regression analyses. Additionally, standard error estimates corrected for the sample’s complex design (i.e., clustering and stratification). The syntax used in Stata was: svyset SECLUSTR [pweight=NSALWTPN], strata(SESTRAT).

RESULTS

Table 2 displays our obesity results, where no denominational differences were observed when compared to Baptists. The Baptist denomination was the largest in our sample (48.8%), and so was used as the denominational reference group in the logistic regressions. Detailed demographic characteristics of the sample can be found in Table 1. No denominational differences in obesity were observed, but an interaction between gender and frequency of religious service attendance was seen. To more clearly denote the effect of gender in the obesity model, men and women were analyzed separately post hoc. Among black men, those who attended religious service nearly everyday had nearly two times greater odds of obesity compared to black men who never attended (OR=2.99, CI: 1.63,5.48). No differences in the odds of obesity by religious service attendances appeared among black women.

Table 2.

Relationship between denomination and obesity by gender

Male
 Female
Odds Ratio 95% CI Odds Ratio 95% CI
Denominationa
 Methodist 0.59 0.25-1.38 0.78 0.51-1.21
 Pentecostal 1.07 0.46-2.49 1.39 0.92-2.09
 Presbyterian 0.19 0.03-1.38 1.48 0.77-2.84
 Protestant (other) 0.99 0.62-1.57 0.96 0.72-1.28
 Catholic 1.11 0.75-1.65 0.95 0.61-1.48
Religious Service Attendanceb
A few times/year 1.31 0.79-2.19 0.76 0.47-1.23
A few times/month 0.86 0.58-1.27 0.86 0.53-1.39
A few times/week 1.14 0.71-1.85 0.91 0.57-1.44
Nearly everyday 2.99*** 1.63-5.48 1.23 0.73-2.07
Age 1.01 0.99-1.02 1.00 0.99-1.01
Household Income 1.00 1.00-1.00 1.00 1.00-1.00
Educationc
 0-11 years 1.22 0.83-1.79 1.15 0.89-1.48
 13-15 years 1.29 0.87-1.92 0.76* 0.57-1.00
 16 or more years 1.43 0.82-2.48 0.57** 0.38-0.85
Employment Statusd
 Unemployed 0.72 0.34-1.53 0.86 0.61-1.24
 Not in labor force 0.66 0.40-1.09 0.92 0.71-1.19
Ethnicitye
 Afro-Caribbean 0.66 0.34-1.27 1.50 0.98-2.28
Country of birthf
 U.S. born 2.66* 1.03-6.87 3.99*** 2.28-6.97
U.S. Geographic Regiong
 Midwest 0.62 0.33-1.16 1.32* 1.04-1.67
 South 0.70 0.45-1.08 1.01 0.79-1.31
 West 0.39 0.15-1.01 0.62*** 0.48-0.80
Open in a new tab
*

p<0.05,

**

p<0.01,

***

p<0.001

CI = Confidence Interval

a

Reference category: Baptist,

b

Reference category: Never or < 1/year,

c

Reference category: 12 years,

d

Reference category: Employed,

e

Reference category: African American,

f

Reference category: Not U.S. born,

g

Reference category: Northeast

We did observe denominational differences with regards to diabetes. The diabetes model (Table 3) showed that Presbyterians were had 87% lower odds of having diabetes (OR=0.13, CI: 0.03,0.56) and Catholics were 45% lower odds of having diabetes (OR=0.55, CI: 0.35,0.86) in comparison with Baptists in a model that controlled for the effects of race/ethnicity, gender, age, employment status, years of education, and frequency of religious service attendance. We also investigated whether an interaction was present between gender and religious service attendance, but none was observed.

Table 3.

Relationship between denomination and diabetes

Model: Diabetes
Odds Ratio 95% CI
Denominationa
 Methodist 0.61 0.34-1.11
 Pentecostal 1.16 0.75-1.82
 Presbyterian 0.13** 0.03-0.58
 Protestant (other) 1.15 0.81-1.64
 Catholic 0.55** 0.34-0.90
Religious Service Attendanceb
 A few times/year 1.21 0.77-1.89
 A few times/month 1.35 0.87-2.08
 At least once/week 1.73* 1.11-2.70
 Nearly everyday 1.19 0.63-2.25
Age 1.04*** 1.03-1.05
Genderc
 Male 0.81 0.60-1.09
Household Income 1.00 1.00-1.00
Educationd
 0-11 years 1.62** 1.16-2.25
 13-15 years 0.90 0.59-1.36
 16 or more years 0.91 0.60-1.37
Employment Statuse
 Unemployed 0.99 0.56-1.76
 Not in labor force 1.39 0.95-2.04
Ethnicityf
 Afro-Caribbean 1.42 0.89-2.27
Country of birthg
 U.S. born 1.75 0.77-3.96
U.S. Geographic Regionh
 Midwest 1.19 0.74-1.93
 South 0.84 0.56-1.25
 West 1.22 0.77-1.94
Open in a new tab
*

p<0.05,

**

p<0.01,

***

p<0.001

CI = Confidence Interval

a

Reference category: Baptist,

b

Reference category: Never or < 1/year,

c

Reference category: Female,

d

Reference category: 12 years,

e

Reference category: Employed,

f

Reference category: African American,

g

Not U.S. born,

h

Northeast

DISCUSSION

Identifying denominational differences in the cardiovascular disease risk factors of obesity and diabetes in black adults is an important addition to the literature. Presently, the evidence to inform our understanding of these risk factors includes samples predominantly composed of whites, limiting our knowledge in other racial and ethnic groups, namely blacks. However, to improve health promotion and disease prevention approaches, and derive a deeper understanding of the connection between religion and black health, variation among black church denominations must be examined.

In this study, we investigated whether the likelihood of obesity and diabetes differed by denomination in a national sample of blacks. Our analysis of NSAL data found no denominational differences in obesity, but did observe an interaction between gender and the frequency of religious service attendance that greatly increased the likelihood of obesity in men, but not women. Thus church engagement is an important factor in obesity for black men. Concerning diabetes, lower odds emerged among Presbyterians and Catholics vis-a-vis Baptists. No interaction between gender and religious service attendance was observed for this outcome. This finding indicates that the odds of diabetes does vary between denominations within the black church, and communicates the importance of considering denomination in health promotion and prevention efforts that target Blacks.

While we observed no denominational differences in obesity, we found an interaction between male gender and frequent service attendance where black men presented with a high odds of obesity. In contrast to our results, previous investigations found no relationship between attendance and obesity among men(Cline & Ferraro, 2006; Godbolt, Vaghela, Burdette, & Hill, 2017). When Cline and Ferraro evaluated the role of gender among Baptists, the subset of their sample with the highest obesity prevalence, no attendance-obesity relationship was observed in men (Cline & Ferraro, 2006). Similar findings were observed in a more recent study that compared church attendance and BMI in blacks and whites, where no relationship was observed among men regardless of race (Godbolt et al., 2017).

The present study’s findings among black men may reflect social networks differences between men who are and are not highly engaged in church, which can impact obesity development. Social networks reflect a “web” of relationships among individuals (Glanz, Rimer, & Viswanath, 2015). A longitudinal study of the spread of obesity by Christakis and Fowlers demonstrated that over thirty-two years that obesity can be transferred among individuals (Christakis & Fowler, 2007). Developing obesity was seen among siblings and spouses if one person became obese . Among friends, if one became obese, the other had a 57% increased chance of also becoming obese . Male same-gender friendships showed an increased chance of becoming obese when one friend enters that weight group, but same-gender obesity spread was not significant for women . The influence of same-gender connections also seemed important for siblings (Christakis & Fowler, 2007). Black men attending church nearly daily, and interacting with other black men, may facilitate the transfer of obesity through the norm-setting and other functions of social networks (Glanz et al., 2015) present in the church.

Although social network spread of obesity may be relevant for black men in our study, other factors may explain the absence of an obesity-attendance relationship among black women. Similar to our results, Cline and Ferraro found no attendance-obesity relationship among black women, but did find an inverse relationship between church attendance and obesity for white women (Cline & Ferraro, 2006). Contrasting our findings and previous work, no attendance-BMI relationship was seen in white women, but a positive direct relationship was seen in black women (Godbolt et al., 2017). This indicated a detrimental effect of more frequent church attendance (Godbolt et al., 2017). Given the mixed evidence base, further study into the gendered aspects of church attendance and its impact on obesity is warranted to deepen our understanding of the relationship across genders.

The absence of an attendance-obesity relationship among black women in our study may relate to the lack of a significant effect of same-sex social network influence on the spread of obesity among women that was seen in men, specifically among friends (Christakis & Fowler, 2007). Perhaps the same-gender networks, or the mechanism of obesity’s spread operates differently by gender. Study of the religious social network and obesity may inform the gender differences. Further, national obesity rates in black women (57.2%) supersede that of all other race and gender groups (Flegal et al., 2016). Whether black women are or are not engaged in the church daily, the majority live with obesity. Our results may reflect the present state of obesity for black women in the U.S., especially considering that the majority of black women, not black men, attend religious services weekly (Cox, 2018). In this study, where nearly daily versus almost no engagement differentiated black men in terms of obesity, and black men have been characterized as a highly religious group (Cox, 2018), they may be a unique subset of church attendees. Gender requires further examination to expand upon our findings by understanding church social networks, their contexts, and weight-related behaviors.

Gendered roles within the church social context may influence obesity in black men and women. Leadership in the Black Church has historically been male (Barnes, 2006; Lincoln et al., 1990). Distinct church roles may also encompass distinct, taxing responsibilities for black men and women within and outside of the church. Perhaps a factor in our findings is the superwoman role, a liability and asset-laden phenomenon describing black women’s stress-coping response and its connection to adverse health outcomes like obesity (Woods-Giscombé, 2010). In the superwoman role, the challenge of being both black and a woman can lead to overwhelm, stress-related behaviors and deprioritization of self-care. Religion and spirituality are inherent in the superwoman role as black women draw strength and resilience from religion and spirituality - a central facet of their lives (Woods-Giscombé, 2010). In our sample, the black men frequently engaged in the church with obesity may embody a “superman role”, experiencing overwhelm from church and other social contexts that undermine self-care, promote obesity and other adverse health outcomes. Despite potentially similar outcomes, the characteristics and mechanisms of these roles may differ by gender. Investigating the role of gendered church context on dietary pattern, eating and physical activity behaviors, and sedentary time may inform our understanding of obesity and diabetes in black adults. Exploration of self-efficacy, social support, motivation and other psychosocial variables related to these behaviors may provide depth to our comprehension of these health outcomes.

We observed a lower likelihood of diabetes among Presbyterians and Catholics in comparison with Baptists in all models. Prior evidence comparing the presence of diabetes across Christian denominations’ is sparse, thus limiting our ability to make comparisons with existing findings.

However, the identification of only Presbyterians and Catholics was unexpected, given that obesity is a primary risk factor for diabetes (Centers for Disease Control and Prevention, 2017b), and we found no denominational differences in the likelihood of obesity. While high odds of obesity were seen among black men with frequent attendance, regardless of denomination, no sex-attendance relationship was observed for the diabetes outcome. Lack of an attendance-diabetes relationship has been reported in the extant literature even when independently-assessed HbA1c is used as an outcome, instead of self-reported diabetes (Das & Nairn, 2016; Hill et al., 2017; Hill et al., 2014). Prospective study designs and HbAlc measurement may advance our understanding of diabetes and attendance among black Christians in future research.

In addition to attendance, age may partially explain the reduced odds of diabetes in Presbyterians and Catholics. Older age is a risk factor for diabetes, specifically being 45 years old and above, according to the Centers for Disease Control and Prevention (Centers for Disease Control and Prevention, 2017b). Perhaps the mean age of 43.5±16.2 years old in our sample contributed to our observations of lower diabetes risk. Other risk factors for diabetes, including the presence of prediabetes, having a parent or sibling with diabetes, a history of gestational diabetes, and being active less than three times per week were not examined in this study. Future investigations may seek to examine the role of these risk factors to understand more completely the denominational differences in diabetes we identified.

Differences in denominational orientations towards health also may inform our diabetes results. Among Presbyterians, Ellison, et al conducted a survey in a predominantly white sample regarding their beliefs and exercise, and found an orientation towards treating the body as a temple (Ellison, Lee, Krause, Hill, & Marcum, 2009). Miller, et al reached a similar conclusion in their discussion of weight and health promotion programs, and they add that the body is also treated as a vessel through which members serve God (Miller, 2017). Catholics may also share a similar perspective between their faith and health.

In a qualitative study that included only white faith leaders, almost half of the Catholic participants held a holistic view of health that included spiritual, mental, social and physical wellness (Webb, Bopp, & Fallon, 2013). While the finding reflects the view of Catholic leadership, these beliefs may be communicated to or also represent that of their parishioners. The literature on religion and health has revealed that denominations have prescriptions that influence health behaviors such as diet, or proscriptions to avoid alcohol, drugs and other “sinful” behaviors (Brathwaite, Fraser, Modeste, Broome, & King, 2003; Faries, McClendon, & Jones, 2017; Fraser, 1999; Holt & McClure, 2006; Miller, 2017). The beliefs specific to denominations may guide individual health behaviors. More research is necessary to understand our diabetes findings among Presbyterians and Catholics, and particularly among blacks in these and other denominations.

Strengths & Limitations

Results from this study should be considered in the context of its limitations and strengths. The secondary data used in this investigation included only self-reported measures of height, weight and diabetes status. The type of diabetes, type 1 or 2, could not be determined in this data set. This distinction is important since type 2 diabetes, the type present in 90-95% of cases (Centers for Disease Control and Prevention, 2017a), can be managed behaviorally through dietary and physical activity changes. Also, the cross-sectional nature of the NSAL data set means conclusions about the causality between denomination and obesity and diabetes cannot be drawn. Considering the type of data collected, we cannot infer that frequent attendance causes obesity in black men, or if black men with obesity attend daily. We also cannot disentangle causality between diabetes and denomination in our sample. Longitudinal data is necessary to understand the influence of denomination and religious engagement through attendance on health across the lifespan.

While our aim was to understand disease differences between Christian denominations, identifying any differences among non-Christian denominations or non-religious persons (agnostic/atheist) in the black population would be an important addition to the literature on religion and health. Non-Christian denominations may be examined to fully assess disease risk across the full range of faith groups to which blacks belong. Importantly, variation in the definition and collection of religious denominations often varies across studies, which limits the possibility for effective between-study comparison. We were limited in our own classifications, considering the secondary data set already defined categories of religious denomination.

Despite these limitations, our study contributes to the understudied area of race, religion and health. The NSAL data set provided a national sample of blacks in the U.S., thus allowing us to draw statistical conclusions. Though the black church is often approached as a monolithic religious entity in health promotion, we provide evidence that sometimes denomination matters in the health of black adults.

Future Directions

Further study is necessary to advance the understanding of denominational differences in obesity and diabetes among blacks in the U.S. This area of research is understudied, but requires in-depth investigations to understand the association between religion and health in blacks who experience higher rates of obesity and diabetes. Given the incongruence between our obesity and diabetes findings, subsequent research may need to compare diabetes risk factors, including, but not limited to, obesity, across denominations to understand drivers of the increased risk. Objective, independently-assessed outcomes like HbA1c should be included to strengthen study conclusions. Comparison of church health doctrine by denomination may also inform opportunities to address diabetes.

The present study applied a quantitative approach, and considering the novelty of the question at hand, more mixed methods research will be required to add breadth and depth to our knowledge of this topic. Analysis of longitudinal data and conducting prospective studies are stronger, more conclusive study designs that may infer causality. Qualitative studies are equally essential as they can provide context for quantitative findings. A mixed methods approach can serve as a mechanism to comprehend denominational differences in the health behaviors, psychosocial processes and other factors related to our observations.

Conclusions

The difference in the likelihood of diabetes between denominations we observed can guide future health promotion and intervention activities that may target blacks through the Black 5 Church. A singular faith-based or faith-placed approach applied to blacks across denominations ) may not be appropriate for research or intervention protocols. Denomination tailored ) strategies may be more appropriate, and should be investigated to determine their effectiveness in reducing diabetes among blacks in the United States.

Suggested reviewers:

  1. Terrence D. Hill, Associate Professor, University of Arizona, tdhill@email.arizona.edu

  2. Gary E. Fraser, MD, PhD, Loma Linda University, gfraser@llu.edu

  3. Karen Hye-cheon Kim Yeary, PhD, Associate Professor, University of Arkansas for Medical Sciences, khk@uams.edu

Acknowledgments

Funding:

This study was funded by the National Institute on Minority Health and Health Disparities (R01 MD011606-01 A1).

Autobiographical paragraph for first author:

Keisha L. Bentley-Edwards, PhD is an Assistant Professor of General Internal Medicine at Duke University School of Medicine. She is also Associate Director of the Samuel DuBois Cook Center on Social Equity, and leads the Health Equity Working Group. Her expertise is in developmental psychology where she uses a cultural lens to understand the human experience and health outcomes. Her current research focuses on the influence of religion and spirituality on African American cardiovascular disease outcomes.

Footnotes

Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.

Conflict of Interest:

The authors have no conflicts of interest to declare.

REFERENCES

  1. American Diabetes Association. (2007). Reduction in weight and cardiovascular disease risk factors in individuals with type 2 diabetes: one-year results of the look AHEAD trial. Diabetes care, 30(6), 1374–1383. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Barnes SL (2006). Whosoever will let her come: Social activism and gender inclusivity in the Black church. Journal for the Scientific Study of Religion, 45(3), 371–387. [Google Scholar]
  3. Billingsley AHC, Cleopatra (1991). The Church, the Family, and the School in the African American Community. The Journal of Negro Education, 60(3), 427–440. [Google Scholar]
  4. Brancati FL, Kao W, Folsom AR, Watson RL, & Szklo M (2000). Incident type 2 diabetes mellitus in african american and white adults: The atherosclerosis risk in communities study. JAMA, 283(17), 2253–2259. doi: 10.1001/jama.283.17.2253 [DOI] [PubMed] [Google Scholar]
  5. Brathwaite N, Fraser HS, Modeste N, Broome H, & King R (2003). Obesity, diabetes, hypertension, and vegetarian status among Seventh-Day Adventists in Barbados: preliminary results. Ethnicity and Disease, 13(1), 34–39. [PubMed] [Google Scholar]
  6. Bruce MA, Sims M, Miller S, Elliott V, Ladipo M (2007). One size fits all? Race, gender and body mass index among US adults. Journal of the National Medical Association, 99(10), 1152. [PMC free article] [PubMed] [Google Scholar]
  7. Campbell MK, Hudson MA, Resnicow K, Blakeney N, Paxton A, & Baskin M (2007). Church-Based Health Promotion Interventions: Evidence and Lessons Learned. Annual Review of Public Health, 28(1), 213–234. doi: 10.1146/annurev.publhealth.28.021406.144016 [DOI] [PubMed] [Google Scholar]
  8. Centers for Disease Control and Prevention. (2016). Defining Adult Overweight and Obesity. Retrieved from https://www.cdc.gov/obesity/adult/defining.html
  9. Centers for Disease Control and Prevention. (2017a). National Diabetes Statistics Report, 2017 Centers for Disease Control and Prevention, U.S. Dept of Health and Human Services. [Google Scholar]
  10. Centers for Disease Control and Prevention. (2017b). Who’s at risk? Retrieved from https://www.cdc.gov/diabetes/basics/risk-factors.html
  11. Chatterjee R, Brancati FL, Shafi T, Edelman D, Pankow JS, Mosley TH, … Yeh HC (2013). Non-traditional risk factors are important contributors to the racial disparity in diabetes risk: the atherosclerosis risk in communities study. J Gen Intern Med, 29(2), 290–297. doi: 10.1007/s11606-013-2569-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Christakis NA, & Fowler JH (2007). The spread of obesity in a large social network over 32 years. New England Journal of Medicine, 357(4), 370–379. [DOI] [PubMed] [Google Scholar]
  13. Cline KM, & Ferraro KF (2006). Does religion increase the prevalence and incidence of obesity in adulthood? Journal for the Scientific Study of Religion, 45(2), 269–281. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Cox KD, Jeff. (2018). Black men are less religious than black women, but more religious than white women and men. Retrieved from https://www.pewresearch.org/fact-tank/2018/09/26/black-men-are-less-religious-than-black-women-but-more-religious-than-white-women-and-men/
  15. Czernichow S, Kengne A-P, Stamatakis E, Hamer M, & Batty GD (2011). Body mass index, waist circumference and waist-hip ratio: which is the better discriminator of cardiovascular disease mortality risk? Evidence from an individual-participant meta-analysis of 82 864 participants from nine cohort studies. Obesity Reviews, 12(9), 680–687. doi: 10.11ll/j.1467-789X.2011.00879.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Dalton M, Cameron AJ, Zimmet PZ, Shaw JE, Jolley D, Dunstan DW, … committee, O. b. o. t. A. s. (2003). Waist circumference, waist-hip ratio and body mass index and their correlation with cardiovascular disease risk factors in Australian adults. Journal of Internal Medicine, 254(6), 555–563. doi: 10.1111/j.1365-2796.2003.01229.x [DOI] [PubMed] [Google Scholar]
  17. Das A, & Nairn S (2016). Religious Attendance and Physiological Problems in Late Life. The Journals of Gerontology: Series B, 71(2), 291–308. doi: 10.1093/geronb/gbu089 [DOI] [PubMed] [Google Scholar]
  18. Dodani S, Kramer ΜK, Williams L, Crawford S, & Kriska A (2009). Fit body and soul: a church-based behavioral lifestyle program for diabetes prevention in African Americans. ETHNICITY & DISEASE, 19(2), 135. [PubMed] [Google Scholar]
  19. Ellison CG, Lee J, Krause N, Hill T, & Marcum J (2009). Faith and fitness: Religious beliefs congregational support, and exercise in a nationwide sample of Presbyterians. Fiath and well-being in late life: Linking theories with evidence in an interdisciplinary inquiry. Nova Science. [Google Scholar]
  20. Faries MD, McClendon M, & Jones EJ (2017). Destroying God’s Temple? Physical Inactivity, Poor Diet, Obesity, and Other “Sin” Behaviors. Journal of Religion and Health. doi: 10.1007/s10943-017-0368-2 [DOI] [PubMed] [Google Scholar]
  21. Ferraro KF (1998). Firm believers? Religion, body weight, and well-being. Review of Religious Research, 224–244. [Google Scholar]
  22. Flegal KM, Kruszon-Moran D, Carroll MD, Fryar CD, & Ogden CL (2016). Trends in Obesity Among Adults in the United States, 2005 to 2014. JAMA, 315(21), 2284–2291. doi: 10.1001/jama.2016.6458 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Fraser GE (1999). Associations between diet and cancer, ischemic heart disease, and all-cause mortality in non-Hispanic white California Seventh-day Adventists. The American Journal of Clinical Nutrition, 70(3), 532s–538s. doi: 10.1093/ajcn/70.3.532s [DOI] [PubMed] [Google Scholar]
  24. Glanz K, Rimer BK, & Viswanath K (2015). Health behavior: theory, research, and practice: San Francisco, CA : Jossey-Bass: & Pfeiffer Imprints, Wiley, [2015]. [Google Scholar]
  25. Godbolt D, Vaghela P, Burdette AM, & Hill TD (2017). Religious attendance and body mass: an examination of variations by race and gender. Journal of Religion and Health, 1–13. [DOI] [PubMed] [Google Scholar]
  26. Heron M (2018). Deaths: Leading Causes for 2016 National Vital Statistics Reports. Retrieved from Hyattsville, MD: [PubMed] [Google Scholar]
  27. Hill TD, Rote SM, & Ellison CG (2017). Religious Participation and Biological Functioning in Mexico. Journal of Aging and Health, 29(6), 951–972. doi: 10.1177/0898264317716244 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Hill TD, Rote SM, Ellison CG, & Burdette AM (2014). Religious Attendance and Biological Functioning:A Multiple Specification Approach. Journal of Aging and Health, 26(5), 766–785. doi: 10.1177/0898264314529333 [DOI] [PubMed] [Google Scholar]
  29. Holt CL, & McClure SM (2006). Perceptions of the religion-health connection among African American church members. Qualitative health research, 16(2), 268–281. [DOI] [PubMed] [Google Scholar]
  30. Hu FB, Manson JE, Stampfer MJ, Colditz G, Liu S, Solomon CG, & Willett WC. (2001). Diet, Lifestyle, and the Risk of Type 2 Diabetes Mellitus in Women. New England Journal of Medicine, 345(11), 790–797. doi: 10.1056/NEJMoa010492 [DOI] [PubMed] [Google Scholar]
  31. Jackson JS, Torres M, Caldwell CH, Neighbors HW, Nesse RM, Taylor RJ, … Williams DR (2004). The National Survey of American Life: a study of racial, ethnic and cultural influences on mental disorders and mental health. International Journal of Methods in Psychiatric Research, 13(4), 196–207. doi: 10.1002/mpr.177 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kim KH, Sobal J, & Wethington E (2003). Religion and body weight. International Journal of Obesity & Related Metabolic Disorders, 27(4), 469. doi: 10.1038/sj.ijo.0802220 [DOI] [PubMed] [Google Scholar]
  33. Knowler W, Barrett-Connor E, Fowler S, Hamman R, Lachin J, Walker E, & Nathan D (2003). Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. Scandinavian Journal of Medicine & Science in Sports, 13(3), 208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Levin JS (1984). The role of the black church in community medicine. Journal of the National Medical Association, 76(5), 477. [PMC free article] [PubMed] [Google Scholar]
  35. Lincoln CE, Mamiya LH, & Duke University P (1990). The Black church in the African-American experience: Durham: : Duke University Press, 1990. [Google Scholar]
  36. McNabb W, Quinn M, Kerver J, Cook S, & Karrison T (1997). The PATHWAYS church-based weight loss program for urban African-American women at risk for diabetes. Diabetes care, 20(10), 1518–1523. [DOI] [PubMed] [Google Scholar]
  37. Miller DG (2017). Considering Weight Loss Programs and Public Health Partnerships in American Evangelical Protestant Churches. Journal of Religion and Health, 57(3), 901–914. [DOI] [PubMed] [Google Scholar]
  38. Montgomery S, Herring P, Yancey A, Beeson L, Butler T, Knutsen S, et al. . (2007). Comparing Self-reported Disease Outcomes, Diet, and Lifestyles in a National Cohort of Black and White Seventh-day Adventists. Preventing Chronic Disease, 4(3). [PMC free article] [PubMed] [Google Scholar]
  39. Pew Research Center. (2014). Religious Landscape Study. Racial and ethnic composition. Retrieved from http://www.pewforum.org/religious-landscape-study/racial-and-ethnic-composition/#
  40. Resnicow K, Campbell ΜK, Carr C, McCarty F, Wang T, Periasamy S, … Stables G (2004). Body and soul: a dietary intervention conducted through African-American churches. American journal of preventive medicine, 27(2), 97–105. [DOI] [PubMed] [Google Scholar]
  41. Webb B, Bopp M, & Fallon EA (2013). A qualitative study of faith leaders’ perceptions of health and wellness. Journal of Religion and Health, 52(1), 235–246. [DOI] [PubMed] [Google Scholar]
  42. Woods-Giscombé CL (2010). Superwoman schema: African American women’s views on stress, strength, and health. Qualitative health research, 20(5), 668–683. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Yeary KH c K, Sobal J, & Wethington E. (2017). Religion and body weight: a review of quantitative studies. Obesity Reviews, 18(10), 1210–1222. doi: 10.1111/obr.12569 [DOI] [PubMed] [Google Scholar]
  44. Zhu S, Heymsfield SB, Toyoshima H, Wang Z, Pietrobelli A, & Heshka S (2005). Race-ethnicity–specific waist circumference cutoffs for identifying cardiovascular disease risk factors. The American Journal of Clinical Nutrition, 81(2), 409–415. doi: 10.1093/ajcn.81.2.409 [DOI] [PubMed] [Google Scholar]

RESOURCES