This item is only available as the following downloads:
1 RELATIONSHIP BETWEEN RELIGIOSITY/SPIRITUALITY AND HEALTHY EATING BEHAVIORS AMONG BLACK MEN By LILLIAN BOYNTON KAYE A DISSERTATION PRESENTED TO THE GRADUATE SCHO OL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2012
2 2012 Lillian Boynton Kaye
3 To my mother, Nan Boynton, who did not leave this earth without first inst illing in me a efficacy needed t o pursue endeavors such as this
4 ACKNOWLEDGMENTS I would like to thank all the people who have supported me, especially throughout my six years of graduate education I have appreciated the encouragement, understanding, and patience. Maintaining balance has been challenging, and although I have tried to demonstrate the high priority of my close relationships, I have often not been able to spend as much time as I would like to with the people I care about most. I would like to thank my family, particularly my father, Freddy Kaye, Ph.D., my stepmother, Susan Evans Kaye, and my grandmother, Sue C. Boynton, for their support and encouragement. I would also like to thank m y closest friends, who I value very highly and who have often provided grounding when I felt overwhelmed. I am particularly grateful to Travis, for having a great deal of patience with the stress and demands of graduate school, and for helping me to take a break from it when I could. I would especially like to thank Dr. Carolyn Tucker for being my mentor, teacher, and advisor for over six years, and for personally caring about me and her other students in a way that goes far above and beyond what would be e xpected. I have learned a great deal from her and value her high degree of dedication to helping underserved individuals. I would also like to thank my dissertation committee members, Dr. Chun Chung Choi, Dr. Scott Miller, Dr. Steven Pokorny, and Dr. Shar i Robinson, for making the time and effort to assist me in accomplishing this major professional goal. Additionally, I would like to thank my supervisors at the University of Florida Counseling and Wellness Center, Dr. Natasha Maynard Pemba, Dr. Ann Jarons ki, and Dr. Paul Schauble, for their genuine care and concern, and for supporting my continued dissertation efforts while on internship.
5 Lastly, I would like to acknowledge Dereck G. Chiu, whose life on this earth ended much too soon but not before he coul d touch the lives of many people. He helps me to remember to be thankful for the opportunities I have and that obtaining a Ph.D. in Counseling Psychology is not only an accomplishment but also a privilege.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURES ................................ ................................ ................................ ........ 10 ABSTRACT ................................ ................................ ................................ ................... 11 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 13 Health Disparities Among African Americans/Black Americans .............................. 13 African American vs. Black: A Note on Terminology ................................ ............... 13 Health Disparities among African American/Black Men ................................ .......... 14 The Importance of Healthy Eating in Reducing Health Disparities .......................... 15 Churches as Health Promotion/Intervention Sites ................................ .................. 16 Religiosity/Spirituality and Health ................................ ................................ ............ 16 God Locus of Health Control as a Potential Predictor of Health Behaviors ............ 18 2 LITE RATURE REVIEW ................................ ................................ .......................... 21 Health Disparities ................................ ................................ ................................ .... 21 Prominent Health Disparities among African American/Black Adults ...................... 22 High Prevalence of Cardiovascular Disease s among African American/Black Adults ................................ ................................ .................. 22 High Prevalence of Cancer among African American/Black Adults .................. 23 High Prevalence of Diabetes among African American/Black Adults ............... 24 Prominent Health Disparities among African American/Black Men ......................... 24 High Prevalence of Cardiovascular Disease s among African American/Black Men ................................ ................................ ..................... 24 High Prevalence of Cancer among African America n/Black Men ..................... 25 High Prevalence of Diabetes among African American/Black Men .................. 25 Influence of Obesity and Diet on Leading Causes o f Death ................................ .... 26 Churches as Health Promotion/Intervention Sites ................................ .................. 28 Links Between Religiosity/Spirituality and Health ................................ ................... 30 Defining Religiosity and Spirituality in Health Related Research ...................... 30 Defining Health ................................ ................................ ................................ 32 Determining the Population of Focus ................................ ............................... 34 Religiosity/Spirituality and Health: Positive Associations ................................ .. 35 Religiosity/Spiri tuality and Health: No Associations or Negative Associations 38 Mechanisms Involved in the Associations Between Religiosity/Spirituality and Health ................................ ................................ ................................ ..... 42 Religiosity/Spirituality and Health Behaviors: Mixed Results ............................ 45
7 God Locus of Health Control ................................ ................................ ................... 50 Health Fat alism Beliefs as a Predictor of Health Behaviors .............................. 51 God Locus of Health Control: Development and Definition .............................. 53 God Locus of Heal th Control as a Predictor of Health Behaviors ..................... 53 The Present Study ................................ ................................ ................................ .. 58 Research Questions ................................ ................................ ......................... 59 Hypotheses ................................ ................................ ................................ ...... 59 3 RESEARCH METHODOLOGY ................................ ................................ ............... 60 Participants ................................ ................................ ................................ ............. 60 Measures ................................ ................................ ................................ ................ 62 Demographic Data Questionnaire ................................ ................................ .... 62 God Locus of Health Control Scale ................................ ................................ .. 63 Brief Multidimensional Measure of Religiousness/Spirituality ........................... 64 Measures of Healthy Eating ................................ ................................ ............. 66 The Health Promoting Lifestyle Profile II ................................ .................... 66 Eating Behavior Patterns Questionnaire ................................ .................... 67 Procedure ................................ ................................ ................................ ............... 69 4 RESULTS ................................ ................................ ................................ ............... 72 Preliminary Analyses ................................ ................................ .............................. 72 Bivariate Correlations ................................ ................................ .............................. 75 GLHC Scores ................................ ................................ ................................ .......... 76 Results for the Research Questions ................................ ................................ ....... 78 Research Question 1 ................................ ................................ ........................ 78 Research Question 2 ................................ ................................ ........................ 78 Research Question 3 ................................ ................................ ........................ 79 Results for the Hypotheses ................................ ................................ ..................... 80 Hypothesis 1 ................................ ................................ ................................ ..... 80 Hypothesis 2 ................................ ................................ ................................ ..... 81 Measurement model ................................ ................................ .................. 84 Structural model ................................ ................................ ......................... 84 Summary of results for hypothesis 2 ................................ .......................... 87 5 DISCUSSION ................................ ................................ ................................ ......... 99 Summary and Interpretation of Results ................................ ................................ 100 Research Question 1: Demographic Associations with Healthy Eating Behaviors ................................ ................................ ................................ .... 101 Research Question 2: Demographic Associations with God Locus of Health Control ................................ ................................ ................................ ......... 101 Research Question 3: The Relationship Between Relig iosity/Spirituality and Healthy Eating Behaviors ................................ ................................ ............ 103 Hypothesis 1: The Relationship Between GLHC and Healthy Eating ............. 103
8 Hypothes is 2: GLHC as a Moderator of Religiosity/Spirituality and Healthy Eating ................................ ................................ ................................ .......... 105 Limitations of the Present Study ................................ ................................ ........... 106 Future Directi ons for Research and Interventions ................................ ................. 109 Implications for Counseling Psychologists ................................ ............................ 113 Conclusions ................................ ................................ ................................ .......... 115 APPENDIX A DEMOGRAPHIC DATA QUESTIONNAIRE ................................ .......................... 117 B GOD LOCUS OF HEALTH CONTROL SCALE ................................ .................... 123 C BRI EF MULTIDIMENSIONAL MEASURE OF RELIGIOUSNESS/SPIRITUALITY (SELECTED SUBSCALES) ........................... 124 D HEALTH PROMOTING LIFESTYLE PROFILE II: NUTRITION SUBSCALE (REVISED) ................................ ................................ ................................ ............ 127 E EATING BEHAVIOR PATTERNS QUESTIONNAIRE: LOW FAT EATING SUBSCALE (REVISED) ................................ ................................ ........................ 128 LIST OF REFERENCES ................................ ................................ ............................. 129 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 140
9 LIST OF TABLES Table page 4 1 Descriptive statistics and bivariate correlations among the variables of interest. ................................ ................................ ................................ ............... 89 4 2 Relations of demographic variables with healthy eating measures. ................... 90 4 3 Relations of demographic variables with God locus of heal th control. ................ 90 4 4 Relations of religiosity/spirituality dimensions and healthy eating. ...................... 91 4 5 Relations of God locus of health control and healthy eating. .............................. 92 4 6 Loadings for the measurement model. ................................ ............................... 93 4 7 Loadings onto healthy eating in the hypothesized s tructural model. ................... 93
10 LIST OF FIGURES Figure page 3 1 Levels of spirituality and religiosity endorsed by study participants. ................... 71 4 1 Structural model of relations between variables of interest and healthy eating.. ................................ ................................ ................................ ................ 94 4 2 Linear relationships between religious coping and healthy eating at the upper and lower quartiles of endorsement of G od locus of health control ................... 95 4 3 Linear relationships between God locus of health control and healthy eating behaviors at two level s of engagement in religious coping. ................................ 96 4 4 Linear relationships between private religious practices and healthy eating at the upper and lower quartiles of endorsement of Go d locus of health c ontrol ... 97 4 5 Linear relationships between God locus of health control and healthy eating behaviors at two levels of engagement in private religious practices. ................ 98
11 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy REL ATIONSHIP BETWEEN RELIGIOSITY/SPIRITUALITY AND HEALTHY EATING BEHAVIORS AMONG BLACK MEN By Lillian Boynton Kaye August 2012 Chair: Carolyn M. Tucker Major: Counseling Psychology A number of major health conditions disproportionately impact Black men, i ncluding cardiovascular diseases, cancers, type 2 diabetes, and obesity. Good nutrition is an important factor in prevent ing and reducing these conditions. In community based health promotion and intervention efforts targeting these health problems, religi ous institutions have received increased attention as implementation settings, particularly within African American/Black communities. Given this prominent trend and given the centrality of faith within mainstream African American culture, it is important to understand how religiosity may relate to African American/Black engagemen t in behaviors such as healthy eating. The pr esent study examined relationships among multiple dimensions of health (God locus of health control; GLHC), and engagement in healthy eating in a sample of 140 Black men. GLHC was examined as a potential m oderator of the relationship s between religiosity/spirituality dimensions and healthy eating.
12 Results revealed that healthy eating was positively associated with daily spiritual experiences and negatively associated with GLHC. Additionally, GLHC moderated the relationships between religiosity/spirituality and healthy eating for two of the four examined dimensions (reli gious coping and private religious practices). Specifically, healthy eating was positively associated with religious coping and private religious practices among individuals with high endorsement of GLHC; however, among individuals with low endorsement of GLHC, healthy eating was negatively associated with religious coping and not associated with private religious practices. health may have an overall negative association w ith engagement in healthy eating among Black men, high religiosity may be protective against this effect. Additionally, spirituality demonstrated a more consistent positive relationship with healthy eating than did religious practices. Findings suggest tha t the relationship between religiosity and healthy eating is complex and inconsistent across varying dimensions of religiosity/spirituality. Given the need for effective interventions to target chronic diseases among African American/Black individuals, and given the trend of housing such interventions in faith based institutions, it is important that program planners understand associations between religious variables and engagement in health behaviors, and consider how these variables might be used to incr ease the efficacy of such interventions.
13 CHAPTER 1 INTRODUCTION Health Disparities Among African Americans/Black Americans violence, or opportunities to achieve optimal health experiences by socially for Disease Control and Prevention [CDC], National Center for Chronic Disease Prevention and Health Promotion, Division of Adult and Commu nity Health 2011). Poverty, environmental threats, inadequate access to health care, individual and behavioral factors, and educational inequalities are some of the numerous factors that may contribute to health disparities (CDC, National Center for Chron ic Disease Prevention and Health Promotion, 2011 ). A few examples of health disparities faced by African Americans/Black Americans are as follows: compared to White Americans African Americans /Black Americans are significantly more likely to rate their he alth as fair or poor ( i.e., 8.7% vs. 14.2%, respectively; National Center for Heal th Statistics [NCHS], 2011 ). O verall mortality, as well as mortality due to stroke, heart disease, cancer, and diabetes, is higher for Black Americans than for White American s (NCHS, 2010). A dditionally although overweight and obesity are staggeringly high among the American population in general, rates of overweight and obesity are higher among African American adults than among any other racial/ethnic group in the U.S. ( NCH S 2010). African American vs. Black : A Note on Terminology Both the terms Black and African American are used throughout this paper. It is important to note that although these terms may seem to be used interchangeably, they
14 are not synonymous. Some ind ividuals identify as Black but do not identify as African American (e.g., Black Caribbeans); some individuals identify as African American, and some individuals identify as both African American (ethnically) and Black (racially). Black is typically a broad er descriptor (i.e., may include both African American individuals as well as other Black individuals). Within the research literature, some studies identify their participants as Black, and some studies identify their participants as African American, whi le other studies may use a longer (but potentially more accurate) term of African American/Black (meaning African American or Black). In the literature review of this paper, when race/ethnicity descriptors are used to describe participants in a published s tudy, the descriptors will be the same as those used by the authors of that study; otherwise, the term African American/Black will be used. Health Disparities among African American/Black Men In addition to health disparities that relate to the African Ame rican/Black population in general, there are significant disparities that relate specifically to African American/Black men. For instance, estimated life expectancy at birth is significantly lower for Black males (70.0 years) as compared with both White ma les (75.9 years) and Black females (76.8 years; NCHS 2010). It is also the case that African American men are 30% more likely to die from heart disease and 60% more likely to die from stroke than non Hispanic White men ( U.S. Department of Health and Human Services [ USDHHS ] Office of Minority Health [ OMH ] 2009). Compared with African American women, African American men also have higher cancer incidence rates overall (USDHHS, National Institutes of Health [NIH], National Cancer Institute [NCI], 2010) and higher death rates due to heart disease (Xu, Kochanek, Murphy, & Tejada Vera, 2010). Yet, not surprisingly, health promotion
15 research is typically more commonly conducted with African American women than with African American men. For example, an automated search within the widely used PubMed database for journal articles (published within the past 10 years and written in produced only 588 articles (search conducted by the author in June 2011). The Importance of Healthy Eating in Reducing Health Disparities National he alth authorities state that the most effective approach to address ing the leading causes of death in the U.S. is to reduce and prevent underlying health risk factors, including poor nutrition physical inactivity, tobacco use, and excessive alcohol use. Ac cordingly, the objectives of Healthy People 2020 include numerous nutrition related goals, including increased consumption of fruits, vegetables, and whole grains, as well as decreased consumption of calories from fat, added sugars, saturated fat, and sodi um (USDHHS, 2011). Good nutrition can help lower the risk for many diseases, including heart disease, stroke, diabetes, and some cancers (USDHHS, National Prevention, Health Promotion, and Public Health Council, 2010); thus, the prevalence of four of the f ive leading causes of death among African Americans (i.e., heart disease, cancer, cardiovascular disease, and diabetes; CDC, NCHS 2011) could be decreased through engagement in recommended dietary behaviors (e.g., eating a diet that is low in fat, added s ugars, and sodium, and eating recommended amounts of fruits, vegetables, and whole grains)
16 Churches as Health Promotion/Intervention Sites Churches and other faith based organizations have recently received increased attention from health promotion resea rchers as useful setting s for implementing community based health promotion and intervention programs for the purpose s o f decreas ing obesity and other chronic diseases and reducing health disparities Several studies have reported implementing health relat ed intervention programs within faith based organizations in African American/Black communities (e.g., Kannan, Sparks, DeWitt, Krishnakumar, & Lumen g 2010; Resnicow et al., 2004; Tuck er et al., 201 2 ; Wilcox et al., 2010 ). Such programs have typically focu sed on decreasing obesity and othe r diet related diseases (e.g., t ype II diabetes) through improving diet and increasing physical activity ; however, some of these programs have targeted o ther health promoting behaviors such as engagement in cancer screenin gs. Given this prominent trend of using churches as intervention sites, and given the centrality of faith and religion within mainstream African American culture, it is important t in health promoting behaviors such as healthy eating. However, such information is currently limited. Ellison and colleagues (2010) point out the need for further research on the intersection of race, religion, and health behaviors; additionally, these autho rs indicate that of the limited research that has been conducted in this area, alcohol and tobacco use has received particular attention (Ellison, Hummer, Burdette, & Benjamins, 2010). Religiosity/Spirituality and Health Research on religiosity/spiritualit A mong community samples of U.S. adults, research suggests that frequent religious practice is associated with living
17 longer, even after controlling for confounding variabl es (Larson, Larson, & Koenig, 2002). A uthors of a systematic review of literature on the linkages between religiosity/spirituality and physical health reported the following wide ranging conclusions: (a) weekly church/service attendance is associated with improved physical and mental health behaviors, with the association being stronger for women than for men, (b) religiosity/spirituality protects against cardiovascular disease possibly due in part to promotion of a healthy lifestyle, (c) the association b etween religiosity/spirituality and cancer mortality is confounded by prior health status (i.e., people who become sick are more likely to become increasingly religious/spiritual), (d) there is a lack of evidence for the general hypothesis that depth of re ligiousness is predictive of mortality, (e) there is inadequate support for religion/spirituality being protective against disability, (f) there is little evidence that religiosity/spirituality slows the progression of cancer, (g) it is inconclusive whethe r use of religious coping results in a longer life, (h) there is a lack of support for religiosity/spirituality improving recovery from acute illness and (i) some evidence suggests that religiosity/spirituality impedes recovery (Powell, Shahabi, & Thorese n, 2003). T he overall conclusion of the authors of the review that a relationship between religion or spirituality and physical health does exist but that it may Shahabi, & Th oresen, 2003, p. 50). The potential pathways (mechanisms) by which religiosity or spirituality influences health are varied and unclear. Park (2007) review ed a multitude of potential pathways, including meaning in life, social support, health locus of con trol, body sanctification, health behaviors, positive and negative affect and stress moderation, treatment
18 adherence, and coping. Park concluded that research is particularly needed that examines the process by which religiosity and spirituality may influe lifestyle choices and ultimately influence their health. Although some studies have found religiosity to be associated with healthy lifestyle choices such as choosing not to smoke cigarettes (e.g., Koenig et al., 1998) and engaging in pre ventative health care behaviors such as cancer or cholesterol screenings (e.g., Reindl Benjamin & Brown, 2004), there is a lack of research that examines potential associations between religiosity/spirituality and healthy eating behaviors in particular. In terestingly, interviews with African American church members suggest that the relationship between religiosity and health may be stronger for avoiding health risk behaviors (e.g., alcohol and drug use) than for increasing health promoting behaviors (e.g., healthy eating; Holt & McClure, 2006). In an effort to address the need for research on the relationship between religiosity and health behaviors, the present study examines God locus of health control as a pathway by which religiosity/spirituality may be associated with healthy eating beh aviors. God Locus of Health Control as a Potential Predictor of Health Behaviors God locus of health control (GLHC ) is the extent to which an individual believes that God control s ough God locus of health control may be perceived as a form of external (vs. internal) locus of control, research has demonstrated the validity of GLHC as a construct that is distinct from other dimensions of health locus of control (e.g., powerful others; Chaplin et al., 2001). It is also important to note that the GLHC construct shares some similarities with the concept of fatalistic health beliefs control) a potentially more familiar construct The i dea that one cannot control the
19 in research as a potential barrier to engaging in preventative health practices; for example, fatalistic beliefs about cancer have been found to be negatively correlated with exercis ing regularly, abstaining from smoking, and eating recommended amounts of fruits and vegetables (Niederdeppe & Levy, 2007). The body of research on God locus of health control as a potential predictor of health or engagement in health behaviors is rather l imited. However, it is interesting to note that much of the research that has been conducted on the potential association between God locus of control and health behaviors has focused specifically on African American/Black participant samples Some finding s from the few studies that ha ve been conducted in this area suggest that endorsing a high God locus of health control is engagement in prostate cancer screening (Leaks, 200 9 ) and br east cancer screening (Kinney, Emery, Dudley, & Croyle, 2002). However, other research has suggested a lack of association of God locus of health control with nutrition behaviors and physical activity behaviors (e.g., Gonnerman, Lutz, Yehieli, & Meisinger, 2008). Given that religiosity is traditionally a central component of African American culture, it is important that religiosity related constructs be examined in relation to health behaviors and health care practices (Polzer & Miles, 2005). It is notewo rthy that some differences in beliefs about the roles of God in relation to health and illness have been linked to race/ethnicity For example, authors of a qualitative study involving older rural African Americans and Whites found that the White older adu lts presented limited God health beliefs in comparison to the African American older adults; specifically, African
20 American participants in the study identified various roles of God in relation to health, including comforter of the sick, arbiter, guardian, health communicator, evaluator, savior, and miracle maker, whereas White participants only identified the concept of God as a miracle maker (and to a much lesser degree than did African American participants; McAuley Pecchioni, & Grant, 2000). Overall, t he amount of research that exists on potential associations between God locus of health control and engagement in health promoting behaviors is highly limited, making it challenging to draw conclusions about this topic until further research has been condu cted
21 CHAPTER 2 LITERATURE REVIEW Health Disparities Numerous definitions exist for the term health disparity (Carter Pokras & Baquet, of disease, injury and viole nce, or opportunities to achieve optimal health experiences by socially disadvantaged racial, ethnic, and other populations, groups, and Centers for Disease Control and Prevention [ CDC ] National Center for Chronic Disease Prevention and Heal th Promotion, Division of Adult and Community Health gender, race or ethnicity, education or income, disability, living in rural localities, or US Dep artment of Health and Human Services [ USDHHS ] 2000). For a review of various definitions of health disparity, see Carter Pokras and Baquet (2002). Multiple factors contribute to health disparities, including poverty, environmental threats, inadequate acc ess to health care, individual and behavioral factors, and educational inequalities (CDC, 2011), as well as lack of access to safe and affordable housing, fresh and affordable fruits and vegetables, transportation, and culturally appropriate interventions and services (World Health Organization [WHO], 2008). Thus, although this paper specifically focuses on personal/psychological factors and on a specific health promoting behavior (i.e., healthy eating), the author recognizes that there are numerous and var ied contributing causes to health disparities, many of which are socioenvironmental (WHO, 2008), and are beyond the scope of this literature review. Additionally, although the topic of proposed study involves African American /Black men
22 as the primary focus the author of this study recognizes that health disparities occur among and wi thin a multitude of demographic groups Prominent Health Disparities among African American/Black Adults There are many examples of health disparities that have a disproportion ately negative impact on African American and Black adults. For example, in 2007, overall mortality was 25% higher for Black Americans than for White Americans, and in 2006, age adjusted death rates of Black Americans exceeded those of White Americans by 4 8% for death caused by stroke, 31% for death caused by heart disease, 21% for death caused by cancer, and 113% for death caused by diabetes ( National Center for Health Statistics [ NCHS ] 2010). Additionally, although obesity is staggeringly high among the American population in general, obesity rates are higher among African Americans than any other racial/ethnic group in the U.S. (NCHS, 2010). Heart disease, malignant neoplasms (cancer), cerebrovascular disease (stroke), and diabetes are among the five lea ding causes of death for African Americans (NCHS, 2011). High Prevalence of Cardiovascular Disease s among African American/Black Adults Heart disease and stroke are two types of cardiovascular diseases (i.e., diseases of the heart and blood vessels). Spec ifically, a stroke is a serious medical condition that occurs when flow of oxygen rich blood is blocked to part of the brain (USDHHS, NHLBI, 2011). Coronary heart disease (i.e., heart disease) is a disorder of the blood vessels of the heart that can lead t o heart attack (USDHHS, NHLBI, n.d.). Heart disease is the leading cause of death for all Americans (NCHS, 2011). Among other major risk factors (e.g., older age, smoking tobacco) that contribute to heart disease are being of the male
23 sex, being African Am erican, having excess body fat, having high blood pressure (i.e., hypertension), and having diabetes (American Heart Association [AHA], 2011). African American adults are more likely than White adults to be diagnosed with heart disease and to die from hea rt disease. Specifically, age adjusted heart disease death rates have been estimated to be 251.9 per 100,000 population for African Americans and 191.4 per 100,000 population for non Hispanic White Americans (Xu, Kochanek, Murphy, & Tejada Vera, 2010). Si milarly, the National Stroke Association estimates that African Americans are twice as likely to die from stroke as Caucasian Americans and more likely to experience stroke earlier in life as compared with Caucasians. Risk factors that appear to contribute to this health disparity include diabetes, sickle cell anemia, smoking, obesity, and high blood pressure (National Stroke Association, 2012). At 44% prevalence, African American adults have among the highest rates of hypertension in the world (Roger et al ., 2011). High Prevalence of Cancer among African American/Black Adults Cancer is the second leading cause of death, both among African Americans and among the overall U.S. population (CDC, NCHS, 2011). However, for most cancers, African Americans experie nce the highest death rates of any racial and ethnic group in the United States. Overall, cancer death rates are estimated to be 32% higher among African American men than White men and 16% higher among African American women than in White women, despite a n overall decline in racial disparities for cancer death rates. Among African Americans, the most common cancer related causes of death are from (1) lung and bronchus cancers, (2) prostate cancer (for men) and breast cancer (for women), (3) colon and rectu m cancer, and (4) pancreatic cancer (American Cancer Society, 2011).
24 High Prevalence of Diabetes among African American/Black Adults Diabetes is a group of diseases marked by high levels of blood glucose resulting from defects in insulin production and/or insulin action (CDC, 2011). Diabetes is a serious health condition that puts one at risk for a variety of other serious health complications, including blindness, heart disease, stroke, amputations, and nerve damage (USDHHS, National Institute of Diabetes and Digestive and Kidney Diseases [NIDDK], National Diabetes Information Clearinghouse, 2008). Diabetes is also the leading cause of kidney failure. Recent research estimates that 4.9 million, or 18.7% of all non Hispanic Black adults aged 20 years or olde r, have (diagnosed or undiagnosed) diabetes. The risk of diagnosed diabetes is estimated to be 77% higher among non Hispanic Black Americans as compared to non Hispanic White Americans (CDC, National Center for Chronic Disease Prevention and Health Promoti on, Division of Diabetes Translation, 2011). In fact, although diabetes has been designated the seventh leading cause of death in the U.S., it is the fifth leading cause of death among African Americans (Xu et al., 2010). Prominent Health Disparities among African American/Black Men High Prevalence of Cardiovascular Disease s among African American/Black Men Estimated death rates due to heart disease are substantially higher among African American men (311 deaths per 100,000 population) as compared with both non Hispanic White men (239 deaths per 100,000 population) and African American women (208 deaths per 100,000 population; Xu et al., 2010). Rates of hypertension are also more prominent among Black men than Black women; specifically, among non Hispanic Bl ack adults who are 20 years of age and over, the age adjusted percentage of individuals with high blood pressure (a major risk factor for heart disease; AHA, 2011)
25 was 23.9 for women and 26.5 for men (from 2003 2006; NCHS, 2010). With regard to mortality d ue to cardiovascular disease s African American men are 30% more likely to die from heart disease and are 60% more likely to die from stroke than non Hispanic White men (USDHHS, OMH, 2009). High Prevalence of Cancer among African American/Black Men Estimat ed overall cancer incidence rates are higher among African American men than African American women (i.e., 88,860 for men vs. 80,040 for women). Similarly, although overall cancer death rates have been decreasing at a faster rate among African American mal es than among African American females, African American males continue to experience significantly higher mortality caused by cancer as compared with African American females (American Cancer Society, 2011). The impact of cancer health disparities on Afri can American/Black men is also evident when comparing these men to men from other racial/ethnic backgrounds. For example, past research has suggested that African American men were 1.3 times more likely to be diagnosed with lung cancer and 1.4 times more l ikely to be diagnosed with prostate cancer, as compared to non Hispanic White men. Additionally, it has been found that African American men are nearly twice as likely as non Hispanic White men to be diagnosed with stomach cancer and 2.5 times as likely to die from prostate cancer (CDC, NCHS, 2011). High Prevalence of Diabetes among African American/Black Men The Centers for Disease Control and Prevention reports that the age adjusted prevalence of diagnosed diabetes is 6.3% for White men, 7.9% for Hispani c men, and 8.5% for African American men. In contrast to the previously mentioned diseases, diabetes rates are actually higher among African American women than among African
26 American men (CDC, 2010). African American men have been estimated to be 2.1 time s more likely than non Hispanic White men to start treatment for end stage renal disease related to diabetes and to be twice as likely to die from diabetes (CDC, National Center for Chronic Disease Prevention and Health Promotion, Division of Diabetes Tran slation, 2010 ). Influence of Obesity and Diet on Leading Causes of Death National health authorities clearly state that the most effective approach to addressing the leading causes of death in the U.S. (e.g., heart disease, cancer) is to reduce and preven t the underlying risk factors of those diseases including poor nutrition (i.e., as well as physical inactivity, tobacco use, and excessive alcohol use; USDHHS, 2011). As such, the objectives of Healthy People 2020 include numerous nutrition related goals, including increased consumption of fruits, vegetables, and whole grains, as well as decreased consumption of added sugars, saturated fat, sodium, and calories from fat (USDHHS, 2011). Good nutrition can help lower the risk for many diseases, including hea rt disease, stroke, diabetes, and some cancers (USDHHS, National Prevention, Health Promotion, and Public Health Council, 2010); thus, one reasonable approach to decreasing four of the five leading causes of death among African Americans (i.e., heart disea se stroke cancer, and diabetes; CDC, NCHS 2011) is through targeting dietary behaviors. O besity is associated with a number of health problems, including each of the above mentioned leading causes of death People who are overweight are more likely to experience high blood pressure, hyperlipidemia, and high LDL cholesterol, all of Although the prevalence of obesity is staggeringly high among the American population
27 a s a whole, obesity rates are higher among African Americans than among any other racial/ethnic group in the U.S.; in fact, the Centers for Disease Control and Prevention has estimated that the proportion of non Hispanic Black m ales who are obese is 35% ( CD C, NCHS 2010). Although cancer incidence is associated with various risk factors, previous research has estimated that 14% of deaths from cancer in men are due to overweight and obesity ( Calle, Rodriguez, Walker Thurmond, & Thun, 2003 ) and recent researc h has estimated that overweight and obesity cause approximately 20% of all cancer cases (Wolin, Carson, & Colditz, 2010). Furthermore, a recent review of research states that approximately 85,000 new cancer cases in the U.S. per year are related to obesity and that as body mass index increases by 5 kg/m 2 risk of death from cancer increases by 10% (Basen Engquist & Chang, 2011). National health authorities state that b ehavioral factors such as quitting smoking, eating a healthy diet, and exercising regular ly can decrease the risk of some types of cancers ( e.g., USDHHS, NIH, NCI, 2011a) ; however, i t is unknown whether or not eating a healthy diet can decrease the risk of certain cancers such as prostate cancer ( USDHHS, NIH, NCI, 2011b) Even so, public heal th/healthcare organizations such as the Mayo Clinic and the Prostate Cancer Foundation recommend decreasing prostate cancer risk through engaging in healthy eating behaviors, such as eating a low fat diet rich in fruits, vegetables, and fish, and maintaini ng a healthy weight (Mayo Clinic, 2009; Prostate Cancer Foundation, 2011). In addition to heart disease, cancer, and stroke, another leading cause of death among African Americans is diabetes. The majority (i.e., 90 95%) of Americans who are diabetic have type 2 diabetes. In addition to risk factors such as family history, older
28 age, physical inactivity, and being of certain racial/ethnic backgrounds ( e.g., African American), type 2 diabetes is associated with being overweight. In fact, researchers have es timated that approximately 80% of people with type 2 diabetes are overweight (USDHHS, NIDDK, N ational Diabetes Information Clearinghouse 2008 ) Thus, one reasonable strategy for preventing diabetes (as well as cardiovascular disease, cancer, and stroke) i s decreasing overweight/obesity, in part through the promotion of engagement in healthy eating behaviors. Churches as Health Promotion/Intervention Sites Research suggests that 10% of U.S. congregations report having at least one formal health related prog ram, and 18% of attendees report being part of a congregation that offers a health related program (Trinitapoli, Ellison, & Boardman, 2009). Churches and other faith based organizations have become increasingly popular sites for community based health prom otion/intervention programs and research studies (Campbell et al., 2007). This may be the case for a number of reasons, such as the availability of the needed space and participants for conducting health promotion interventions and research. However, use o f faith based organizations for these purposes may also present some challenges (e.g., overcoming mistrust of research from pastors and congregants, respecting organizational values and norms, deciding whether or not to incorporate religious beliefs to mak e the program faith based or to simply make it faith placed ; for further information, see Campbell et al., 2007). Churches have received particular attention as settings for implementing health promotion/intervention programs within African American/Black communities. Recent examples of such interventions include the Body & Soul Program (Resnicow et al., 2004), Healthy Eating and Harambee (Kannan, Sparks, DeWitt, Krishnakumar, &
29 Lumeng, 2010), the Faith, Activity, and Nutrition (FAN) Program (Wilcox et al. 2010), The WORD (Wholeness, Oneness, Righteousness, Deliverance; Kim et al., 2008), and the Health Smart Chu rch Program (Tucker et al., 201 2 ). Such programs have typically focused on decreasing obesity and other diet related diseases (e.g., type II diabe tes) through improving diet and increasing physical activity, although some programs have targeted other health promoting behaviors, such as engaging in preventative cancer screenings. Given the evidence that church based health promotion programs have mu ch potential for promoting healthy behaviors (Campbell et al., 2007; Newlin, Dyess, Allard, Chase, & Melkus, 2011), it is important to understand if and how religious beliefs may hy eating. Additionally, given the centrality of faith and religion within mainstream African American culture, as well as the movement to use faith based organizations within African American/Black communities as sites for health intervention programs, it is particularly important understand if/how faith engagement in the health promoting behaviors. Holt and McClure (2006) note that (a) few church based health promotion programs directly integrate religiosity in to the content of their interventions and (b) it is unclear how integrating religious beliefs into health programs might affect risk behaviors versus prevention behaviors. It is challenging to determine what portion of health programs that are housed in fa ith based organizations actually incorporate a direct religious component, as reviews of such programs often do not include examination of the potential impact of religiosity on the intervention.
30 Links Between Religiosity/Spirituality and Health Given the great deal of research that has been conducted on the very broad topic of the relationship between religiosity/spirituality (R/S) and health, it may be helpful to first offer a sample conceptualization of the multiple viewpoints on this topic. Seybold (200 7) conceptualizes researchers who study the association between R/S and health (i.e., mental health and physical health) as typically being in one of two camps. One camp holds the view that R/S has a positive association with physical and mental health, as indicated by findings such as positive correlations between R/S and health habits and negative correlations between R/S and mortality, heart disease, stroke, cholesterol levels, depression, suicide, substance abuse, and risky sexual behaviors (e.g., Koeni g et al., 2001, Larson and Larson, 2003, as cited by Seybold, 2007). The second camp denies that there is significant empirical evidence of a link between R/S and positive health outcomes (e.g., Sloan & Bagiella, 2002, as cited by Seybold, 2007). Seybold a dded that there are other researchers who are not quite in either camp (e.g., Thoresen & Harris, 2002, as cited by Seybold, 2007), and these researchers uphold a dismissed by cri 304). Defining Religiosity and Spirituality in Health Related Research Historically, when health related studies have included a measurement of religiosity and/or spirituality, researc hers have examined religiosity or spirituality as (frequently single item) and imprecise global indices, such as frequency of church attendance, denominational affiliatio n, or self
31 Pargament, 2003, p. 65). For instance, in a review by Koenig and colleagues (2001) of studies that examined the association between religiosity/spirituality (R/S) and mortality, almost half of the re viewed studies measured religious affiliation (only), without incorporating any other measures of religiosity or spirituality. Hill and Pargament (2003) question the reliability of such brief and uni dimensional measures and also suggest that such measures may miss potential harmful effects of religion and spirituality on health (i.e., negative associations between R/S and health). Additionally, it is interesting to note that over the past few decades, studies in psychology and other behavioral sciences hav e demonstrated a trend toward assessing spirituality as well as religion (i.e., as opposed to the prior norm of assessing only religion) in research addressing potential connections between these variables and health (Weaver, Pargament, Flannelly, & Oppenh eimer, 2006). The debate over how to define and measure religiosity and spirituality is vast and goes beyond the scope of this paper; however, a number of thorough and thought provoking writings are available for more information on this subject (e.g., Ch iu, Emblen, Hofwegen, Sawatzky, & Meyerhoff, 2004; Hall, Meador, & Koenig, 2008; Hill & Pargament, 2003; Idler et al., 2003; Moberg, 2002). Idler and colleagues (2003), who created a multidimensional measure of religiousness and spirituality, contend that the lack of reliable and validated measures of religiosity and spirituality has been one of the greatest barriers to research in this area. Seybold and Hill (2001) also explain that multidimensional measures of religion and spirituality were developed in r esponse to concerns about inconsistencies in measuring and defining religiosity and spirituality, noting that such inconsistencies may be partially responsible for the conflicting findings
32 on the nature of the association(s) between religiosity and health. In a thorough review of existing measurements of religiosity and spirituality for health research, Hall and mited by the attempt 158). Defining Health Defining health is a complicated matter. The World Healt h Organization defines being and not merely the has received criticisms, mostly in relation to the impract icality of the term complete (Huber et al., 2011). A recent article by Huber and colleagues (2011) reviews the limitations of this definition and proposes a framework for reformulation. These authors discuss that on one hand, it may be sensible to consider health to be a general concept, and on the other hand, operational definitions are needed for practical purposes such as measurement, research, and evaluation of interventions. In addition assessing physical or mental health status (e.g., presence or abs ence of health conditions), from a preventative and practical perspective, it is useful to measure health behaviors. Although from a traditional medical perspective, people are considered healthy if they are uninjured and free of disease, an alternative pe rspective is that a person may be considered unhealthy due to having risk factors for a disease (Institute of Medicine Committee on Health and Behavior: Research, Practice, and Policy, 2001). Given that some of the prominent underlying risk factors for the leading causes of death in the U.S. are behavioral in nature (e.g., physical inactivity, poor
33 nutrition, tobacco use, and excessive alcohol use; USDHHS, 2011), it is sensible to include health behaviors when examining relationships between psychosocial va riables (e.g., religiosity) and health, particularly because of the potential for intervention. As care rather than expert care, 001, p. 166). Drawing conclusions from the body of literature examining potential associations between religiosity/spirituality (R/S) and health is complicated not only by the challenges related to measuring religiosity related variables (discussed in the prior section) but also by the great number of ways that health may be operationalized. Regarding this line of research, Seybold and Hill (2001) note that careful attention must be paid to the measurement of health in addition to the measurement of R/S. T hese authors explain that studies on associations between religiosity and health have used a wide variety of operationalizations of health, ranging from specific health indices (e.g., blood pressure), to mental health related behaviors (e.g., drug and alco hol abuse), to general and subjective measures (e.g., self S ome examples of the physical and mental health varia bles that have been examined in association with R/S include the following: overall mortality, cardiovascular disease incidence, cancer mortality, disability, cancer progression, and recovery from acute illness (e.g., see review by Powell, Shahabi, & Thore sen, 2003), outcomes after stroke (e.g., Colantonio, Kasl, Ostfeld, & Berkman, 1993), outcomes after heart surgery (e.g., Contrada et al., 2004), mobility (e.g., Fitchett, Rybarczyk, DeMarco, & Nicholas, 1999), depression (e.g., Fitchett et al., 1999), lif e satisfaction/quality of life (e.g., Fitchett
34 et al., 1999; Rippentrop, Altmaier, & Burns, 2006), cigarette smoking (e.g., Koenig et al., 1998), preventative health care utilization (e.g., Reindi Benjamins & Brown, 2004), and engagement in cancer screenin g (e.g., Abernethy, Houston, Bjorck, Gorsuch, & Arnold, 2009). Clearly, research on the associations between R/S and health (however defined) is vast and varied; thus, generalized statements about such associations may be risky, especially given the functi onal and conceptual differences in defining health in terms of physical health outcomes, versus mental health status, versus health behaviors. The remainder of this literature review will focus primarily on studies examining physical health outcomes (e.g., cardiovascular disease) and health behaviors (e.g., cancer screenings) rather than mental health status (e.g., depression) Determining the Population of Focus Drawing conclusions about the relationships between religiosity/spirituality (R/S) and healt h is additionally complicated by the variety of populations on which such research has been conducted. The majority of studies within this body of literature appear to focus on the health outcomes of clinical populations with serious illnesses or chronic d iseases. A few examples of such patient populations are heart surgery patients (e.g., Contrada et al., 2004; Oxman, Freeman, & Manheimer, 1995), stroke patients (e.g., Giaquinto, Spiridigliozzi, & Caracciolo, 2007), and chronic pain patients (e.g., Rippent rop, 2005). It is important to consider, however, that connections between R/S and health related factors may differ among individuals with acute or chronic illnesses versus persons from non clinical samples (i.e., community based samples). From a prevent ion perspective, it is necessary to understand how various psychosocial factors are associated with health or engagement in health behaviors prior to the development of chronic illnesses. Although studies on the associations between
35 R/S and health among ge neral (community based) populations seem to have increased within the past decade, more of such research is needed in order to understand the potential roles of religiosity in health promotion and early intervention efforts. Additionally, George and collea gues (2002) assert that more than half of the research on the religiosity health connection is based on samples of older adults (George, Ellison, & Larson, 2002). However, the older population may be a less prominent target with regard to preventative heal th efforts; thus, from a prevention orientation, it is important for future research in this area to include individuals from a spectrum of ages. Lastly, in order to ensure that findings from studies on religiosity and health are as useful as possible in d isease prevention efforts, it may be important for future research to examine the variables of discussion among specific (non clinical) population sub groups, especially those sub groups that experience significant disparities in relation to health (e.g., African American/Black men). Religiosity/Spirituality and Health: Positive Associations Generally speaking, there is thought to be a positive relationship between religiosity and health, such that self reported religiosity/spirituality (R/S) predicts bet ter health. A great deal of research has been conducted that supports this notion. Because the literature on positive associations between R/S and health outcomes/health status is vast, only meta analyses and literature reviews will be cited here with rega rd to this specific topic; on the other hand, individual research studies will be cited with regard to associations between R/S and health behaviors (i.e., as opposed to health outcomes/status variables) because such studies are less numerous. Possibly the most consistent finding among meta analyses that examine the relationship between R/S and health is that religious involvement (often measured as
36 attendance at religious services) is negatively associated with mortality. In other words, higher involvem ent is predictive of decreased mortality (e.g., McCullough, Hoyt, Larson, Koenig, & Thoresen, 2000; Williams & Sternthal, 2007). Interestingly, however, a recent meta analysis of 91 studies found that religious involvement was positively associated with re duced mortality among healthy populations, but not among diseased populations (Chida, Steptoe, & Powell, 2009). been found to have a protective effect against mortality (i.e., both mortality from cardiovascular disease and cancer as well as mortality from all causes), with this effect being strongest when religiosity is operationalized as attendance at religious services (i.e., as opposed to other measures of religiosity ). Findings from this review also suggest that religiosity is positively associated with (a) a reduced incidence of disease and disability, (b) recovery from physical illness, (c) increased tolerance of pain, and (d) prevention of and recovery from mental illness and substance abuse. The authors add that the association between religious involvement and health outcomes is stronger for mental health outcomes than for physical health outcomes or mortality. The authors also describe some limitations of the stu dies that they reviewed. For example, with regard to the common operationalization of religiosity as frequency of attendance at religious services, they indicate that it is important to consider the confounding factor that individuals must maintain a certa in baseline threshold of health in order to be capable of attending religious services (George, Larson, Koenig, & McCullough, 2000). In other words, operationalizing R/S as frequency of attendance at religious services
37 does not take into account those indi viduals who are, for example, highly religious but too ill to attend religious services. In a levels of evidence review of literature on the linkages between religiosity/spirituality and physical health, researchers concluded that (a) religiosity/spiritua lity has a protective effect against cardiovascular disease, and (b) weekly church/service attendance is associated with improved physical and mental health behaviors, with the association being stronger for women than for men (Powell et al., 2003). Additi onally, data from the well known National Health and Nutrition Examination Survey (NHANES III), which included a sample of over 14,000 American male and female adults, indicates that on average (after controlling for socio demographic and health variables) individuals who attend religious services weekly or more than weekly have lower systolic blood pressure as compared to individuals who never attend religious services; these frequent attenders also exhibited a somewhat lower hypertension prevalence than non attenders (Gillum & Ingram, 2006). It is notable, however, that these findings are based on a single item question about attendance at religious services, as opposed to a broader or multidimensional measure of religiosity and spirituality. Additionally positive associations between religiosity/spirituality and health have been found among African American/Black American adults specifically. For example, Brown and Gary (1994) examined the associations between religious involvement and health status/beha viors among 537 African American males living in an urban area. Results of this study indicated (a) lower depression levels among males who reported moderate to high frequency of church attendance and (b) better self reported mental
38 health among males who were affiliated with a religious denomination versus those who were unaffiliated. Similarly, in a study of 3,963 persons age 65 years or older, it was found that blood pressure was lower on average among individuals who attended religious services and pray ed or studied the Bible frequently, as compared with individuals who had a low frequency of engaging in these religious activities. This trend was found even after controlling for a number of potentially confounding variables such as age, education, and bo dy mass index. The researchers of the study noted that this difference in blood pressure was particularly notable among the Black participants in their sample (Koenig et al., 1998). Furthermore, Ellison and colleagues (2000) have reported that very frequen t church attendance among African Americans was positively associated with lower mortality and longer life, in that African Americans who reported that they never attend church had two times higher mortality rates (over a nine year period) than those who r eported attending church more than once per week Ellison and colleagues also noted that the degree of difference in mortality between very frequent church attenders versus non attenders in their (exclusively) African American sample was larger than that w hich had been reported in similar studies involving the general population (Ellison, Hummer, Cormier, & Rogers, 2000). Religiosity/Spirituality and Health: No Associations or Negative Associations As described above, there is a great deal of research to suggest that significant and positive associations exist between religiosity/spirituality (R/S) variables and health related variables. However, there are also findings from studies that suggest either negative associations between R/S and health or a lack of association between R/S and health. Some of these studies are described in the following paragraphs.
39 In a review of literature by Williams and Sternthal (2007), these authors report that despite significant support of positive associations between R/S and health, some research findings suggest a negative association between R/S and health. The potential negative effects summarized by these authors include the following: faith based social relations as a source of stress, feelings of guilt contributing to illness, reliance on faith healing rather than on medical care, extrinsic religious orientation and negative religious coping as predictors of depressive symptoms, and among medically ill patients, religious doubts and struggle as predictors of mortalit y. These authors also point out that due to conflicting research findings, the relationship between religious participation and obesity is ambiguous. Researchers of the large, population services and prevalence or levels of cardiovascular disease, dietary intake, or metabolic to this summary stat ement included the findings that (a) fish intake was higher among frequent attenders, (b) stroke prevalence was lower in frequently attending African American women, and (c) obesity and diabetes rates were higher among frequent attenders than among infrequ ent attenders. However, this latter association (i.e., with regard to obesity and diabetes) was not significant when the researchers controlled for demographic variables (Obisesan et al., 2006). In an analysis of literature (i.e., 266 articles) on claims about religion, spirituality, and health, Sloan and Bagiella (2002) suggest that there is not an abundance of literature demonstrating a protective effect of religiosity/spirituality on health. They
40 conclude that many of the studies cited by comprehensive reviews that support a beneficial effect of religiosity on health (i.e., on heart disease and hypertension in particular) (a) were not actually relevant to this claim, (b) had significant methodological flaws, and/or (c) were misrepresented in the secondar y source (e.g., a study was considered to be an examination of religion and health solely because it was conducted the area of heart disease and hypertension reveale d that there is little empirical support 20). There are some studies that have examined the association among R/S and health among African American/Black Americans and have found m ixed results or a lack of association between these variables. For example, Brown and Gary (1994) examined associations between religious involvement and health status/behaviors among 537 African American males living in an urban area. Results indicated th at although religious involvement was found to have a positive influence on mental health, it was not found to be associated with overall physical health status or with hypertension. Additionally, it is interesting to take a closer look at the Ellison and colleagues (2000) study which found lower mortality among African Americans who frequently attended church versus those who did not attend church [described in the previous section on positive associations between religiosity and health]. Some interesting gender differences were found when the effects of selectivity factors (e.g., sociodemographic factors and baseline health status), social ties, and health behaviors were controlled. Specifically, the difference in mortality between frequent church attender s and church non attenders increased among the African American women;
41 however, among the African American men, this differential decreased to the point of statistical nonsignificance (Ellison et al., 2000). Thus, for the African American males in this stu dy, it may be interpreted that differences in mortality rates in association with church attendance frequency may be attributable to third party variables rather than to a protective effect of involvement in organized religion. In relation to the idea that religiosity/spirituality (R/S) might have a negative association with health, George and colleagues (2002) contend that although religious involvement might have a negative association with health variables among small proportion of individuals, this is n ot a pattern that has been found among samples as a whole. George and colleagues (2000) summarize their examination of prior research on the relationship between religiosity and health by describing that religious involvement has been mostly found to have either a positive effect or no e ffect on health and that have found no evidence that religion can harm health in representative samples of community residents or in systematically sampled clinical populations 2000, p. 110). Specifica lly, George and colleagues (2002) explain that among research studies with large samples (i.e., not including case studies or studies with small clinical samples), the small percentage that has shown a negative association between religiousness and health have been related to either (a) religions which recommend against seeking professional medical care or (b) negative religious coping (e.g., feeling punished or abandoned by God, believing sickness results from sin; George Ellison, & Larson, 2002). Similar ly, Seybold and Hill (2001) point out that although positive effects of religiosity/spirituality on health are more commonly observed, negative effects of religious and spiritual experience on health are most likely to occur when a person
42 believes either t hat he or she has direct communication with God with little or no social accountability or uses a deferral to God problem solving strategy. Although Powell and colleagues (2003) also support the conclusion that there is a clear protective link between rel igious service attendance and mortality, they contend that there is little evidence that depth of religiousness showed a similar protective effect. Additionally, their review generally found a lack of support for the view that religiosity/spirituality (R/S ) protects against disability or assists in recovery from physical illness; in fact, they noted that religious beliefs may have harmful effects in patients who are ill if they (a) engage in negative religious coping, such as becoming upset that God has aba ndoned them, or (b) depend on their faith for recovery rather than on medical complex than h pose an interesting suggestion that re more powerful as a coping resource that buffers the impact of disease in patients who are ill than as a resistance res ource in healthy people confusing to interpret in light of the finding from a recent meta analysis that R/S has a protective effect against mortality among healthy population samples but not among disease d population samples (Chida et al., 2009). Mechanisms Involved in the Associations Between Religiosity/Spirituality and Health Given the plethora of research suggesting significant associations between religiosity/spirituality (R/S) and health, with the ma jority of studies supporting a positive relationship, more recent research has attempted to examine the mechanisms of this
43 association. The potential mechanisms of this association have been summarized by a number of authors (e.g., Ellison & Levin, 1998; G eorge, Ellison, & Larson, 2002; Park, 2007; Seybold & Hill, 2001). For example, in their review, Ellison and Levin (1998) summarize the potential explanatory mechanisms involved in the religion health connection as follows: (1) discouragement of certain he alth risk behaviors (e.g., substance use and abuse, risky sexual behaviors) and encouragement of low stress lifestyle choices (e.g., decreased marital disruption); (2) provision of social resources (e.g., opportunities for social integration and social sup port); (3) promotion of positive self feelings (e.g., self worth, self efficacy); (4) various forms of religious coping resources and coping behaviors (e.g., prayer, the perspective that difficult situations are opportunities for spiritual growth); (5) pos itive emotions (e.g., forgiveness, contentment, love); (6) healthy beliefs or learned optimism (e.g., expecting religious practice to benefit health or for piety to be rewarded with health and well being); and (7) additional possible mechanisms/pathways th at are not easily examined within the bounds of modern scientific research (e.g., direct effects of prayer or other types of spiritual healing). potential mechanisms by which r eligion may influence health: (a) social networking/fellowship that serves to reduce psychological and physical stressors; (b) lifestyle behaviors (e.g., abstinence from smoking, alcohol/drug abuse, and sexual risk behaviors); (c) psychological factors (e. g., coping strategies, purpose and meaning, acceptance, optimism); and (d) emotions (e.g., forgiveness) and practices (e.g., meditation) that may have physiological benefits (e.g., improved immune function) via
44 stress reduction. George and colleges (2002) present four main categories of mediators, which have received varying degrees of support from past research. These mediators include (1) health practices, (2) social support, (3) psychosocial resources (e.g., self esteem, self efficacy), and (4) sense of coherence or meaning. Additionally, with regard to health practices, these authors suggest two main mechanisms by which religious involvement may potentially promote health protecting and health promoting behaviors. One mechanism is the common religious te aching that one should respect p. 193). The second mechanism is certain health practices/habits that some particular religious groups (e.g., Mormons) engage in as par t of their religious teaching (e.g., abstaining from alcohol; George et al., 2002). However, questions about the effect of religiosity on health behaviors are raised by studies such as the one by Hixon and colleagues (1998), which found that although reli giosity was positively associated with decreased blood pressure among a sample of female adults of Judeo Christian faiths, path analyses demonstrated that the direct effects of religiosity were more powerful than the indirect effects of religiosity by way of engagement in health behaviors (e.g., physical activity, diet, abstaining from smoking; Hixon, Gruchow, & Morgan, 1998). In other words, the positive association between religiosity and low blood pressure did not appear to be due to increased engagement in health promotion behaviors among more religious persons; rather, this positive association appeared to be primarily due to some kind of direct effect of religiosity on blood pressure.
45 framework, she examined evidence from past research for the pathways of body sanctification, meaning in life, social support, positive and negative affect and stress moderation, treatment adherence, coping, health behaviors, and health locus of control. O ne of the religiosity/spirituality and health, as well as some data that suggests associations between these variables, few studies have actually tested the proposed pathways. One cumulatively influence their health. Religiosity/Spirituality and Health Behavior s: Mixed Results Findings on the examination of associations between religiosity/spirituality (R/S) and health behaviors (i.e., engagement in health promoting behaviors or avoidance of health risk behaviors) appear to be mixed. Some studies have found R/S to be positively associated with healthy lifestyle choices, such as not smoking cigarettes (e.g., Koenig et al., 1998), being less likely to abuse and/or become dependent on alcohol and other drugs, being more likely to recover from prior substance abuse/d ependence (George et al., 2000), and engaging in preventative health care behaviors such as cancer or cholesterol screenings (e.g., Reindl Benjamins, Ellison, Krause, & Marcum, 2011; Reindl Benjamins & Brown, 2004). However, studies that measure religiosit y and/or spirituality in a unidimensional manner are greatly limited in the depth of their findings, as compared to those that measure multiple dimensions of religiosity. For example, Reindl of Presbyterian adults examined if use of preventive services (e.g., cholesterol screening,
46 flu shot, and colonoscopy) was associated with religious attendance, four aspects of health related congregational support, and two health related religious beliefs (i.e., sanctity of the body and God locus of health control). Findings were significantly more complex than simply demonstrating a relationship between religiosity and preventive services. Specifically, two aspects of health related congregational support (e.g., discussing health issues with fellow church members) were associated with use of preventive services; however, neither sanctity of the body nor God locus of health control was related to use of preventive services. Furthermore, although religious a ttendance was significantly related to preventive services use, the authors report that the association appeared to be explained by age ( Reindl Benjamins et al., 2011), a finding which suggests the importance of controlling for demographic variables in suc h research. In a review and analysis of literature on the religion health connection, Ellison and Levin (1998) report that research has demonstrated an inverse relationship between various religious variables and health risk behaviors, such as risky sexual behaviors and substance abuse. For example, in a large epidemiological study of older adults, frequent attendance at religious services and frequent engagement in private religious activity were both significantly associated with being less likely to smok e cigarettes; ever start smoking, not more likely to quit smoking once they started (Koenig et al., 1998). Interestingly, a study of over 900 African American women l iving in South Carolina found no association between attending religious services and participating in physical
47 activities. The authors note that their findings differ from those of some other studies but also note that a limited range of variability of re ligiosity among participants in their study Thompson, Richter, & Henderson, 2003). In a large Utah based study involving over 6,000 Mormon and non Mormon adults, it was foun d that religious persons who attended church on a weekly basis exercised more frequently than religious persons who attended church less than weekly; however, when age, smoking status, education, marital status, and general health were controlled for, thes e differences were no longer found to be significant (Merrill & Thygerson, 2001). Data from the highly regarded National Health and Nutrition Examination Survey (NHANES III), which included nearly 12,000 adults from various regions of the country, found s ignificant differences in leisure time physical activity in association with attendance at religious services among women 60 years or older without m obility limitations, but not among men of any age group, nor women between the ages of 20 to 59; Gillum, 20 06). In a 28 year longitudinal study of data collected from 1965 to 1994, Strawbridge and colleagues (2001) found that weekly attendance at religious services was positively associated with quitting smoking and becoming physically active, and positively a ssociated with stopping heavy drinking (for women but not for men), but weekly service attendance was not associated with getting medical checkups. Overall, year period, those attending services weekly were more likely than those attending less or not at all to both establish good
48 (Strawbridge, Shema, Cohen, & Kaplan, 2001, p. 70). A measure of dietary behaviors was not include d in this study. The number of studies examining associations between religiosity/spirituality and health promoting behaviors that include an examination of healthy eating behaviors appears to be significantly limited; a sample of these studies follows. A recent community based study of adults 65 years or older (in which all of the participants were White) measured the associations among intrinsic and extrinsic religious orientation, body sanctification, relationship with God, and multiple health behavio rs (including healthy eating), and found that stronger religiosity was generally associated with healthier behaviors. However, having an extrinsic religious orientation predicted poorer nutrition and poorer health responsibility, while having an intrinsic religious orientation predicted better nutrition. (Intrinsic orientation was described as having internalized religious values and beliefs, while extrinsic religious orientation was described as using religion as an instrumental means to an end, such as to feel better). Body sanctification predicted engagement in exercise but was not related to engagement in healthy eating behaviors. These analyses were conducted while controlling for both sociodemographic variables and service attendance (Homan & Boyatzis, 2010). In a study of 546 adults living in upstate New York, religious factors were found to have few significant associations with fat intake and physical activity. Similarly to some of the previously mentioned studies, significant gender differences were found. Specifically, among women, religious commitment (operationalized as contributing money to religious causes) was positively associated with greater engagement in moderate and vigorous physical activity; however, it is difficult to interpret this fin ding
49 given that income was not included as a covariate. Among women, fat intake differed in association with religious denomination (i.e., Catholic women reported lower fat intake as compared with Conservative Protestant women, women with no religious pref erence, and women with Other religious denomination). Among men, divine social support was positively associated with engagement in moderate physical activity; however, for the males in this study, no relationship was found between religion and fat intake. The authors caution that due to the limited range of religiosity among the participants in their study, the results of this study may underestimate the effect of religion on health behaviors. Race/ethnicity was controlled for in the analyses for this stud y. However, race/ethnicity was measured only as White and Other and potential racial differences were not examined by the authors (Kim & Sobal, 2004). In the NHANES III study, no significant differences were found between frequent attenders of religious services and infrequent attenders with regard to total and saturated fat intake (Obisesan et al., 2006). Additionally, in a study of over 400 African American church members in the Midwest, religiosity and spirituality were not found to significantly infl uence exercise, weight management, or participation in cancer screenings; however, religion and spirituality did have a significant negative association with healthy eating behaviors, in that participants who were very or moderately religious were least li kely to report eating behaviors consistent with recommended nutrition guidelines, and participants who were slightly spiritual or slightly religious were more likely to report engagement in recommended dietary practices. The authors point out some limits o n the generalizability of their study, in that the majority of the sample had health insurance, described their health as good or excellent and was college
50 educated, middle or upper income, female, and recognized as leaders in their faith community (Under wood & Powell, 2006). God Locus of Health Control One religiosity variable that may influence the relationship between religiosity/spirituality and engagement in health promoting/prevention behaviors is God locus of health control God locus of health con trol is the degree to which one believes Although the general concept of locus of control has been a popular psychological research topic for decades, examination of the specific concept of God locus of health control has be en less frequent Historically, researchers (e.g., Welton Adkins, Ingle, & Dixon, 1996 ) have discussed the need for a measure of God locus of control that is distinct from as opposed to an extension of other forms of locus of control such as internal co ntrol external control and control by powerful others. Additionally, empirical support has been demonstrated for measuring God control as a separate dimension of locus of control (e.g., Chaplin et al., 2001; Wallston et al., 1 999; Welton et al. 1996). G od locus of health control is distinct from the general construct of God locus of control, which define God locus of health control as the extent to which an individual belie ves that God controls his/her health status McAuley and colleagues (2000) found qualitative differences in the beliefs about the roles of God in relation to health and illness between African American and White rural older adults. These researchers interviewed White rural older adults and African American rural older adults about their beliefs in God in relation to health and illness.
51 Despite many similarities in the beliefs of the Whi te participants and the African American participants, the authors also noted some important differences in their beliefs. For instance, the African American elders as compared to the White American elders more often mentioned sin as a cause of illness; in terestingly, however, participants who held this belief were more likely to attribute illness to sin in relation to others rather than to themselves The authors also reported that African American participants were more likely than White American particip ants to (a) believe that God verall, McAuley and colleagues concluded that there were substantial differences in beliefs about the roles of God in relation to health and illness between the African American participants and the White participants in their study, despite the fact that all participants were older adults and lived in the same rural area. Specifically, the African American elders articulated a greater number of and more specific roles that God plays with regard to health and illness (McAuley et al., 2000). Although this st udy is exploratory and utilizes a very small sample, the findings suggest the possibility of racial/ethnic differences in beliefs about how God Health Fatalism Beliefs as a Predictor of Health Behaviors Th e concept of God locus of health control shares some similarities with the more familiar concept of health fatalism Health fatalism beliefs t he idea that one cannot control the outcome of has been proposed as a potential barrier to engaging in preventative health practices such as cancer prevention behaviors. Fatalistic beliefs about cancer
52 prevention have been found to be relatively common among the American population and have been found to be neg atively correlated with engagement in health promotion behaviors such as exercising regularly, not smoking, and eating recommended amounts of fruits and vegetables (e.g., Niederdeppe & Levy, 2007). In fact, a national survey of over 6,000 Americans found t hat 27% of respondents agreed with this statement: (Niederdeppe & Levy, 2007, p. 1000). It is unclear whether or not racial/ethnic differences exist in the endorsement of fatalist ic health beliefs. For example, Franklin and colleagues (2007) reported that African American participants in their Nashville based study endorsed higher religious health fatalism tha n survey in dicated endorsement of lower health fatalism beliefs among African Americans as compared to White Americans. The differences in the findings of these studies may be due to differences in the measurement of the health fatalism construct. For instance, r elig ious health fatalism was measured in the first study, while in the second study fatalism was measured in relation to cancer specifically and the construct did not include a religious component In a study of over 1,000 Nashville residents religious fat alism was found to predict (a) increased engagement in emotional eating and (b) poorer self reported health status However religious fatalism but was not significantly associated with health care utilization behaviors, physical activity, or self reported overeating Religious fatalism beliefs were also found to be positively associated with both fat increasing behaviors (e.g., eating fast food often) and fat decreasing behaviors (e.g., choosing low fat
53 options) T his finding is difficult to interpret. Ove rall, findings from th e study suggest that the relationship between religious fatalism and engagement in health behaviors is unclear and potentially complex (Franklin et al., 2007). God Locus of Health Control: Development and Definition Although the gen eral concept of locus of control has been a popular psychological research topic for decades, examination of the specific concept of God locus of health control has been less frequent. Historically, researchers (e.g., Welton, Adkins, Ingle, & Dixon, 1996) have discussed the need for a measure of God locus of control that is distinct from as opposed to an extension of other forms of locus of control, such as internal control, external control, and control by powerful others. Additionally, empirical support h as been demonstrated for measuring God control as a separate dimension of locus of control (e.g., Chaplin et al., 2001; Wallston et al., 1999; Welton et al., 1996). The proposed study utilizes the God locus of health control measure by Wallston and colleag ues (1999). These authors define God locus of health control as the extent to which an individual believes that God controls his/her health status (Wallston et al., 1999). This construct is distinct from the general construct of God locus of control, which God Locus of Health Control as a Predictor of Health Behaviors A small amount of research has examined potential associations between God locu s of health control and engagement in health behaviors. It is interesting to note that although the amount of research on this topic is limited, a significant portion of this small body of research has examined these constructs specifically among the Afric an American population
54 health related research may be attributed to the traditional centrality of religion in African American culture and to increased efforts to develop community based interve ntions that target the health disparities impacting this population. Polzer and Miles (2005) suggest that because spirituality and religion are commonly and broadly infused in African American life, it is crucial that such concepts be taken into account wh en considering the health and health management of African Americans. In a qualitative study involving African Americans living with diabetes Polzer and Miles (2005) examined if certain religious and spiritual beliefs and practices interact ed with diabet es management. T hese authors distinguish ed between the perspectives of turning to God versus turning it over to God in the context of diabetes management, source of suppor God also [typically] believe that individuals have some responsibility in managing their 237). T his practice of turning to God is differentiated from the idea of turning it over to God : Some African Americans, although they turn to God to give them the support they need, may also, at times, turn it over to God. In other words, they may partially or wholly relinquish control of their health care to God, the ultimat e controller. For some individuals this can mean not following recommendations as set forth by health care providers. For persons with diabetes, this can increase the risk of complications and mortality from this illness. (p. 236) These authors believed th at among African Americans, those most likely to turn their health over to God are individuals who are low income, less educated, members of fundamentalist Christian faiths, elderly, and living in the southern United States (Polzer & Miles, 2005).
55 Similar ly, a small qualitative study by Holt and McClure (2006) examined the perceptions of the religiosity health connection among African American church members The se authors reported that the notion of the body as a temple of God was often mentioned in suppo rt for taking care of oneself. Interestingly, however, the authors point out that nearly everyone who referenced this idea of the body as a temple did so in relation to abstaining from health risk behaviors such as alcohol, tobacco, and drug use, or pre /e xtramarital sex, rather than in relation to supporting engagement in preventative health behaviors such as healthy eating. The authors also reported that participants spoke of giving ealth problems up to God not as a form of passivity but rather as all that they can do. Holt and McClure describe the importance of spiritual health as a form of coping with illness but also describe a specific perspective that is highly relevant to the concept of God locus of health control: Spiritual health could also account for the proportion of religious people who rely on God to heal them (e.g., if their faith is strong enough) and either refuse treatment or refuse to take what the medical commu nity sees as proactive steps to protect their health, such as getting routine screenings. Although none of the participants in this study espoused this belief, some acknowledged that they knew people who did. This type of behavior, and the rationale underl ying it, is poorly understood and is a potentially important area of further exploration. (p. 278) Some quantitative studies have also examined potential associations between God locus of health control (or a closely related variable) and health behaviors, such as diabetes management, preventative cancer screenings, healthy eating behaviors, and physical activity. A number of these studies suggest that having a high God locus of health control may be negatively associated with engagement in health behaviors An example of such a finding comes from a study involving patients from a hospital/health
56 clinic that serves mostly uninsured individuals with low income and low educational backgrounds. D iabetes management was examined among a sample of patients with ty pe 2 diabetes, the majority of whom were African American. One of the findings was that endorsing a high God locus of health control was associated with higher HbA1c (i.e., an indicator of inadequate blood sugar management), suggesting that individuals who strongly believe d that God controlled their health exhibited poorer diabetes management This trend was particularly significant for individuals who endorsed a low internal health locus of control in addition to a high God locus of health control; thus, t he authors concluded that the association between and poor treatment adherence may be particularly present for individuals who feel that they are not personally responsible for their own 2005). Multiple studies involving God locus of health control (or a similar construct) have focused on cancer screening behaviors. A mong a sample of over 300 men of African descent who were 50 to 70 years old, God control was found to be negatively associa ted with prostate screening be havior (Leaks, 200 9 ). Similarly, in a small study of 52 African American women in rural Louisiana who were at high risk for breast cancer, findings suggested that high endorsement of God locus of health control was associated with decreased seeking of clinical breast examinations and mammography (Kinney, Emery, Dudley, & Croyle, 2002). Another study examined the relationshi ps among spiritual health locus of control, breast cancer beliefs, and mammography utilization among Afric an American women living in an urban area. Results from this study suggested that active spirituality health locus of control (e.g., if I lead a good spiritual life, I will stay healthy) was positively associated with perceived barriers to
57 mammography and negatively associated with perceived benefits of mammography (Holt, Clark, Kreuter, & Rubio, 2003), suggesting that African American women who endorse a high spirituality health locus of control may be less likely to seek mammography. Other studies suggest a positive association or a lack of association between God locus of health control (or a closely related construct) and health behaviors. For example, in a study involving a national sample of Presbyterian adults, God locus of health control was not foun d to be significantly associated with cholesterol screenings, flu shots, nor colonoscopy (i.e., preventive health behaviors; Reindl Benjamins et al., 2011). On the other hand, a two sample study with college students that measured both God locus of contro l and God locus of health control found that God locus of control positively predicted health habits in both samples and that God locus of health control positively predicted health habits in only one of the two samples. It is important to note that the he alth habits construct used in the study was operationalized to include both physical and psychological health components (Welton et al., 1996). Finally, in a study of 100 African American adults living in a small city in the Midwest, God health control was positively associated with engagement in a health promoting lifestyle ( a construct which is based on a combination of measures of interpersonal relationships, stress management, spiritual growth, nutrition, health responsibility, and physical activity) H owever, when these constructs were analyzed separately, God locus of health control was not correlated with nutrition behaviors, physical activity, nor health responsibility (Gonnerman, Lutz, Yehieli, & Meisinger, 2008)
58 Thus, the findings are mixed with regard to the associations between God locus of health control and health behaviors. Additionally, it is interesting to note that the instances in which positive associations with God locus of health control were found occurred when health behaviors/health y lifestyle was operationalized to include psychological components (e.g., stress management). As aptly stated by Reindl sample, and outcomes examined, the findings to date [on the relationship between God 466). The Present Study The pr esent study utilize d a single time point survey de sign to collect information on religious beliefs healthy eat ing behaviors, and demographic characteristics from a community based sample of adult men who identify as Black. The general purpose of the present study was to examine the relationships among God locus of health control (GLHC), dimensions of religiosity a nd spirituality (i.e., organizational religiousness, private religious practices, religious coping, and daily spiritual experiences) and healthy eating within a sample of Black men. The primary specific purpose of the present study was to examine the cons truct of God locus of health control as a potential m oderator of the relationship s between religiosity/spirituality and engagement in healthy eating [i.e., to examine GLHC as a potential m oderator of the relationship between (a) organizational religiousnes s and healthy eating, (b) private religious practices and healthy eating, (c) religious coping and healthy eating, and (d) daily spiritual experiences and healthy eating].
59 Religiosity and spirituality were conceptualized multidimensionally, as has been rec ommended (e.g., Idler et al., 2003, Seybold & Hill, 2001) Multidimensional measurement of religiosity and spirituality is in contrast to the historically more common approach of measuring only one aspect of religiosity or spirituality using a single item (e.g., frequency of attendance at religious services). God locus of health control was conceptualized according to perception of GLHC as Res earch Questions The following research questions were explored: Research Question 1: Does engagement in healthy eating vary in association with age, education, and /or income? Research Question 2: Does degree of belief in God locus of health control vary in association with age, education, and/or income? Research Question 3: A re the four dimensions of religiosity/spirituality assessed in this study (i.e., organizational religiousness, private religious practices, religious coping, and daily spiritual exp eriences) significantly associated with healthy eating? Hypotheses Additionally, the following hypotheses were tested: Hypothesis 1: It is hypothesized that when controlling for age, education, and income God locus of health control will have a significa nt negative association with healthy eating Hypothesis 2: It is hypothesized that when controlling for age, education, and income God locus of health control (GLHC) will moderate the relationship s between dimensions of religiosity/spirituality (i.e., org anizational religiousness, private religious practices, religious coping, and daily spiritual practices) and healthy eating. Specifically, it is hypothesized that there will be (a) a negative relationship between healthy eating and each dimension of relig iosity/spirituality among men who ascribe to a high GLHC, and (b) a positive relationship between healthy eating and each dimension of religiosity/spirituality among men who ascribe to a low GLHC.
60 CHAPTER 3 RESEARCH METHODOLOGY Participants P articipant d ata used in the pr esent study w as collected as part of a larger study was to assess psychosocial variables (e.g., motivators, barriers, cultural values) that may affect engagemen t in healthy eating behaviors and prostate cancer screening behaviors among Black men. the University of Florida. The participants in the present study were Black males living within one or the other of two cities in Northeast and North Central Florida. Participant inclusion criteria were as follows: (a) self identif ies as a Black male, (b) is 21 years of ag e or older, and (c) understand s English. Participants were recruited at various community sites and venues, such as health fairs, conferences/meetings, and community centers where Black men were expected to be well represented. A total of 150 participants who met participant inclusion criteria completed the assessment battery (AB) for this study. Of these 150 participants, the data of nine participants were not included in the data analyses due to missing responses for more than 20% of the items of any giv en variable dropped due to being a significant outlier. Thus, data from 140 participants were analyzed in this study. Participants ranged in age from 21 to 82 years old ( M = 47.75, SD = 12.23, Mdn = 49.00). All participants identified as being Black (an inclusion criteria for the study). With regard to ethnicity,
61 and approximately 3% identified with another ethnic description. Education was operationalized as highest l evel of education achieved. This was a highly educated sample, in that approximately 31% of participants reported completing graduate school, 29% completed college, 25% completed some college or technical school, 13% completed high school or GED, and less than 2% did not report their education. With regard to employment status, approximately 61% of participants reported working full time, 16% do not work (e.g., retired, disability), 11% work part time, 10% were currently unemployed but looking for a job, a nd 2% did not report their employment status. Income range was operationalized as annual combined household income. The median annual household income range reported was $50,000 to $59,000. When collapsing some of the income ranges together, approximate ly 19% of respondents reported a household income below $30,000; 23% reported a household income between $30,000 and $50,000; 21% reported a household income between $50,000 to $70,000; 19% reported a household income between $80,000 to $100,000; and 15% r eported a household income over $100,000. Approximately 3% did not report their income range. In terms of relationship status, approximately 61% of respondents reported being married and living with their spouse; while 17% were single, living without a par tner; 10% were single, living with a partner; 8% we re divorced or separated; 2% were
62 married, not living with their spouse; and less than 2% were widowed. The majority of respondents (74%) reported living in an urban area (i.e., a city), while 24% reporte d living in a rural area, and less than 2% did not provide this information. With regard to religiosity and spirituality, 40% of respondents self identified as religiou entations of the levels of religiosity and spirituality endorsed by participants in this sample can be seen in Figure 3 1. Measures Participants were asked to complete an assessment battery that consisted of a series of psychosocial and health related que stionnaires as well as a demographic data questionnaire. The measures used include full scales or subscales of the following: (1) the God Locus of Health Control scale, (2) the Brief Multidimensional Measure of Religiousness/Spirituality, (3) the Health Pr omoting Lifestyle Profile II, and (4) the Eating Behavior Patterns Questionnaire. Below are descriptions of these measures. Demographic Data Questionnaire The demographic data questionnaire ( DDQ ) is a researcher constructed assessment designed to gather d emographic and basic health information from participants in the present study. The DDQ assesses self report information on demographic variables such as education level, annual household income, marital status, gender, age, and ethnicity.
6 3 God Locus of He alth Control Scale The God Locus of Health Control s cale (GLHC scale ; Wallston et al., 1999) is status. According to its authors the GLHC scale can be used to assess either (a) the (b) the belief that God is the locus of control for a specific disease/health condition, which can be achieved with a slight alteration of wording (Wallston et al., 1999 ). In the p roposed study, the GLHC scale will be used to assess the degree to which one believes that Multidimensional Health Locus of Control (MHLC) scales by Kennet h A. Wallston and colleagues, the GLHC s cale was developed to be used either in conjunction with the other MHLC scales or as a stand alone measure ; in the proposed study, it will be used as a stand alone measure. Support for the use o f GLHC as a separate d imension of external locus of control has been reported (Chaplin et al., 2001). A detailed discussion on the validity of the MHLC scales is available (see Wallston, 2005). Wallston and colleagues (1999) examined the validity of the GLHC measure in a multi sample study of persons with one of two specific chronic diseases (e.g., rheumatoid arthritis); in the study, authors reported finding (a) positive correlations ( r = .29 and .32) between the GLHC scores and ratings of importance of religion, (b) modest po sitive correlations ( r = .20 to .22) between GLHC scores and the Other People dimension of the MLHC but no significant correlations with the Internal or Doctors (locus of control) dimensions, and (c) positive correlations ( r = .42 and .47) between GLHC sco res and scores on a religious coping measure. Additionally, with
64 regard to discriminant validity, scores on the GLHC did not correlate with gender, income, or degree of physical disability. Responses to th e 6 item GLHC scale are provided using a 6 p oint Li kert scale ranging from strongly disagree =1 to strongly agree =6 ; higher score s represent higher belief that God is The internal consisten cy of the GLHC scale has been reported to be .93 among a sample of racially diverse college students (Masters & Wallston, 2005) and .88 among a community based sample of ethnically diverse women (Chaplin et al., 2001). y based sample of Black men was also .88. Brief Multidimensional Measure of Religiousness/Spirituality The Brief Multidimensional Measure of Religiousness/Spirituality (BMMRS; Fetzer Institute, 1999) assesses various domains of religious and spiritual expe rience. measure of all aspects of religion and spirituality; it has the limited goal of assessing r et al., 2003 p. 356 ). The following four subscales of the BMMRS were used in this study: Daily Spiritual Experiences, Private Religious Practices, Organizational Religiousness, and Religious Coping. Daily Spiritual Experiences measure s the perception of interaction with or involvement of God/the transcendent in daily life experiences. Private Religious Practices me asure s non organizational/non institutional religious/spiritual practices. Organizational Religiousness measure s involvement with a formal pub lic religious institution/place of worship ( Fetzer Institute, 1999). Religious Coping is operationalized as the extent to which one involves religion in understanding or dealing
65 with stressful situations. Although both the 3 item Positive Religious/Spiritu al Coping subscale and the 1 item Overall Religious Coping item were administered, the Overall Religious Coping item was chosen for use in analyses because it was less highly correlated with the other three BMMRS subscales (and therefore, exhibited less po tential for issues with multicollinearity) and also because it appears to capture a more general and distinct religious coping construct. Responses to items on the BMMRS are provided using a Likert scale, with the number of response choices varying by subs cale from four options to eight options. For example, the Private Religious Practices subscale uses an 8 point response scale ranging from never to m ore than once a day An example item from this subscale is On the other hand, the Daily Spiritual Experiences subscale uses a 6 point response scale ranging from never or almost never to many times a day Discriminant validity of the items on this measur e was assessed via a correlation matrix including all items and all factors/indices (see Idler et al., 2003). Responses from this measure collected as part of the 1998 General Social Survey (GSS) of the University of Chicago National Opinion Research Cente r (i.e., a nationally representative household survey) were utilized in a factor analysis, and convergent validity was established via correlations against a set of previously validated religion items from the GSS. Furthermore Neff (2006) examined the rel iability and validity of the complete BMMRS and determined that despite small differences in factor loadings in a ssociation with race/ethnicity, the measure appeared to be useful in research with African American participants as well as non Hispanic White American participants
66 Reported Cronbach 's alphas are .91 [Daily Spiritual Experiences], .72 [Private Religious Activities], and .82 [Organizational Religiousness] ( Idler et al., 2003). Cronbach 's alphas within the present sample similar to reported values were 87 [D aily Spiritual Experiences], .80 [Private Religious Activities], and .8 4 [Organizational Religiousness] Measures of Healthy Eating Due to the difficulty in measuring self reported healthy eating behaviors while avoiding time intensive measure ments, two measures of healthy eating were used. These measures were the Nutrition subscale of the Health Promoting Lifestyles Profile II (HPLP II; Walker & Hill Polerecky, 1996 ; Walker, Sechrist, & Pender, 1987) and the Low Fat Eating subscale of the Eati ng Behavior Patterns Questionnaire (EBPQ; Schlundt, Hargreaves, & Buchowski, 2003) A description of each measure follows. The Health Promoting Lifestyle Profile II The Health Promoting Lifestyle Profile II (HPLP II; Walker & Hill Polerecky, 1996 ; Walker, Sechrist, & Pender, 1987) assesses self reported engagement in a health promoting lifestyle. Although the 52 item HPLP II has six subscales, only the 9 item Nutrition subscale was used in the present study as a measure of healthy eating behaviors. An exam ple item on th is subscale is Responses to items on t h e HPLP II are made us ing a 4 point Likert scale ranging from never =1 to routinely =4 Reported the HPLP II are .94 3 for the total scale and .757 for the Nutrition subscale. Reported 3 week test retest stability fo r the total scale is .892. Authors of the HPLP II report that construct validity of this measure was supported by a factor analysis and that its criterion r elated validity was indicated by significant correlations (.269 to .491) between
67 the HPLP II and measures of perceived health status and quality of life (Walker & Hill Polerecky, 1996). In order to obtain information on three additional eating behaviors t hat are significant with regard to health promotion/disease prevention (i.e., sodium intake, eating whole grains, and drinking sugary drinks) but are no t directly assessed by this subscale three additional items were included for the purposes of this stud y, using wording and sentence structure similar to the existing items. Prior to running analyses with this slightly modified version of the HPLP II, bivariate correlations were out the three additional items. The 9 item and 12 item versions of the measure were extremely highly correlated ( r versions; the 12 items demonstrated a slightly higher internal relia bility (.81) than the 9 items (.79). Eating Behavior Patterns Questionnaire The Eating Behavior Patterns Questionnaire (EBPQ; Schlundt, Hargreaves, & Buchowski, 2003) assesses self reported prevalence of various eating behaviors. Although the scale is com prised of six subscales, only the Low Fat Eating subscale (14 items) was used in the present study ; this subscale was utilized as a second measure of healthy eating behaviors (i.e., in addition to the HPLP II). This subscale lists various low fat eating be respondents to what degree they agree with each behavior. Responses are provided on a 5 point Likert scale that ranges from strongly disagree = 1 to strongly agree = 5 Thus, higher scores indicate higher reported engagement in low fat eating behaviors.
68 The EBPQ was specifically developed for use with African American women and was validated with samples of African American women of diverse socioeconomic backgrounds who reside in the Southern United States. Unfortunately, there is a scarcity of published research that has used this measure with Black men. However, = .82) was quite close to the value reported in t he validation study involving a sample of African American women ( = .84) A uthors of the EBPQ report that the Low Fat Eating subscale and the other subscales of the EBPQ were determined based on the factor outc omes of a factor analysis. C onstruct valida tion was determined through the examination of correlations of on (a) a food frequency questionnaire and (b) the Eating Styles Questionnaire ( which measures eating behaviors that reduce fat inta ke; H argreaves et al., 1999). S cores from the Low Fat Eating subscale of the EBPQ have been reported to have a significant positive correlat ion with eating behaviors that reduce fat intake ( r = .65) and a significant negative correlat ion with total fat intake ( r = .37), saturated fat intake, ( r = .37), and percentage of energy from fat ( r = .50; Schlundt et al., 2003). To make the EBPQ less focused on ideas and behaviors that may potentially be considered more traditionally feminine (e.g., counting fat grams, using recipes), a few minor wording changes were and cutting portions male sample.
69 Procedure itutional Review Board approved the larger study of which this study is a part The procedures of this study consist ed of (a) recruitment and (b) a single data collection via completion of an assessment battery. Recruitment was conducted via tabling (i.e., setting up an informational booth and speaking with potential participants ) at events t argeted toward African American or Black men, such as health fairs in predominantly African American /Black communities and meetings of organizations comprised of Africa n American/Black men. Recruiters at each recruitment site were one of the principal investigators, a trained researcher (i.e., graduate student or post doctoral researcher), and/or a trained research assistant (i.e., undergraduate student). Prior to servi ng as recruiters for the proposed study, all persons were trained in the procedures of the study and in the implementation of ethical and culturally sensitive research. Recruiters informed potential participants that the purpose of the study is to find ef fective ways to promote health and prevent disease among African American and Black men. These potential participants were asked if they met inclusion criteria and if so, were invited to participate in the study. They were informed that participating in th e study involves completing a questionnaire packet on site that requires approximately 25 minutes to complete and that their responses w ould be anonymous (i.e., their name would not be written on their questionnaires). Potential participants w ere also info rmed that they w ould be given either a $10 gift card or an item of equal or greater value (e.g., a custom T shirt) f or their participation in the study and were specifically informed which compensation options were available prior to choosing whether or no t to participate.
70 Men who agreed to participate were provided an informed consent letter that further explain ed the study (i.e., the purpose of the study, inclusion criteria participation procedures, compensation) and were shown the assessment battery ( AB). A recruiter provided each participant with an AB to complete prior to leaving the event and i nvited the participant to ask questions and to inform the recruiter if he would prefer to have the questionnaire packet read to him. Participants were given u p to two hours to complete the AB on site, al though it was previously observed that male students could complete this AB with in approximately 25 minutes or less Upon returning a completed AB to the recruiter participants were provided the agreed upon co mpensation and thanked for their involvement in the study. Data was collected over approximately a 1.5 year period, depending on when certain community events occurred. Given the nature of the study, the Institutional Review Board approved for the consent not collected in any way in the case of individuals who chose a material compensation (e.g., a T shirt). Participants who chose to be compensated in the form of a gift card wer e asked to print and sign their name on a receipt and payment log. After completion, these documents and the AB were stored in a locked room within the Psychology Department of the host institution. Participant names could not be connected with their data in any way, given that no identifying information was recorded on their AB.
71 Figure 3 1. Levels of spirituality and religiosity endorsed by study participants.
72 CHAPTER 4 RESULTS Preliminary Analyses Prior to conducting analyses, data were examined for missing responses. All measures in the data set were examined, and cases with missing data within a particular scale or subscale were identified. Person mean substitution (i.e., person level mean substitution) was used to impute missing scores. In per son mean substitution, missing items on that scale (Downey & King, 1998). There is evidence to support using this method over listwise deletion or item level mean substitution for cross sectional data if half or more of the items of a scale are complete (Hawthorne & Elliott, 2005); nonetheless, given that as a result of this method, inter item correlations become inflated as the number of missing items increase (Downey & King, 1998), imputation wa s used only for participants whose data was missing no more than 20% of items on a given scale. This (Downey & King, 1998). Cases missing responses for more than 20% of item s within a scale (i.e., excluding demographic variables) were dropped from the data set ( n = 9). Of the remaining participants, less than 10% of participants ( n = 13) had one or more missing items on any of the subscales of the variables of study (i.e., ex cluding demographic variables). For the three demographic (i.e., single item) variables used as covariates in the analyses (i.e., age, income, education level), mean level substitution was utilized to impute missing data; less than 10% of data was imputed for each of these variables.
73 In addition to checking data for missing item responses, study variables were evaluated for univariate normality and multicollinearity. Bivariate correlations were conducted with all variables of interest to assess for multic ollinearity among variables (i.e., variables being so highly related that they are essentially redundant; Weston & Gore, 2006). None of the correlations appeared problematic with regard to multicollinearity (i.e., r s > .85; Kline, 2005). However, to reduce the possibility of problems arising from multicollinearity, scale scores were centered. In order to prepare the data for moderation analyses, interactions terms were created with God locus of health control (GLHC) and each of the four religiosity dimensio ns (e.g., GLHC x Religious Coping). Again, to reduce the threat of multicollinearity, these four newly created variables were centered as well. To assess univariate normality, histograms of each variable were visually inspected, and skewness and kurtosis values were evaluated. Additionally, Mahalanobis distances were calculated in AMOS 18.0 (Arbuckle, 2006) to identify multivariate outliers and assess their effect on departures from multivariate normality (Gao, Mokhtarian, & Johnston, 2008). One multivaria te outlier (i.e., one case) was deleted due to it having a particularly large effect on both the univariate kurtosis of one variable and outlier was deleted, distrib utions of all variables met guidelines for univariate normality above the critica l value), and five participants emerged as multivariate outliers with Mahalanobis distances significant at p < .001. Upon visual inspection of the data,
74 however, it appeared that none of these participant outliers demonstrated patterns of intentional or co model was tested with no additional cases deleted, given that (a) parameter estimates tend to remain valid under reasonable assumptions even when the data are nonnormal (McDonald & Ho, 2002), (b) testing assumptions for multivariate normality is typically impractical (Weston & Gore, 2006), (c) deletion of outliers until multivariate normality is reached may be undesirable due to loss of information and power (Gao et al., 2008), and (d) tests of univariate normality in the current sample provided acceptable results. Although the size of the final sample with which primary analyses were run ( n = 140) is smaller than the common ly reported sample size guideline for s tructural e quation m ode ling (SEM) of at least 200 cases (Weston & Gore, 2006), the size of the current sample does meet other documented guideline s of 5 to 10 participants per observed indicator ( Bryant & Yarnold, 1995 ). Additionally, it has been noted that sample size with rega rd to SEM is one of the multiple SEM related issues in which there is no consensus, and adequate sample size determination depends on a number of factors (Weston & Gore, 2006). For example MacCallum and colleagues (1996) indicate d that a smaller minimum s ample size is necessary if less power is required, if a researcher is testing for close fit versus exact fit, and if the model has less complexity. Given that this study is preliminary in nature and entails a l ow complex ity model with only one latent model ed variable, the current sample size was judged to be adequate. Additionally, the present sample fits within recommendations by Lei and Lomax (2009) that a sample size of more than 100 should be used in order to obtain more accurate and less biased paramet er estimates.
75 Bivariate Correlations variables of interest are presented in Table 4 1. As would be expected, the four examined dimensions of religiosity/spirituality (i.e., subscales of the Brief Multidimensional Measure of Religiosity and Spirituality [BMMRS]) were significantly correlated with each other; that is, Daily Spiritual Experiences (DSE), Private Reli gious Practices (PRP), Organizational Religiousness (OR), and Re ligious Coping (Cope ) were all significantly correlated ( p < .01) with r s ranging from .44 to .66, demonstrating medium to large effect sizes (Sink & Stroh, 2006) Despite these significant correlations, none of these variables were so highly correlated a s to be problematic with regard to multicollinearity (i.e., r > .85; Kline, 2005)]. God Locus of Health Control (GLHC) demonstrated a significant positive correlation with Organized Religiousness ( r = .18, p < .05), but did not correlate with any of the o ther three religiosity/spirituality dimensions. GLHC was negatively correlated with both measur es of healthy eating behaviors [ i.e., the Nutrition subscale of the Health Promoting Lifestyles Profile (HPLP) r = .18, p < .05, and the Low Fat Eating subscal e of the Eating Behavior Patterns Questionnaire (EBPQ) r = .22, p = .01], suggesting in healthy eating/low fat eating behaviors. These significant relationships with G LHC demonstrated small effect sizes (Sink & Stroh, 2006). Of the four religiosity/spirituality dimensions, only Daily Spiritual Experiences correlated with one of the healthy eating measures ( i.e., EBPQ; r = .19, p < .05), and this was a small positive cor relation (Sink & Stroh, 2006)
76 With regard to the three demographic variables (i.e., age, education level, income), neither age nor income level correlated with any of the religiosity variables; however, education level correlated negatively with Private Religious Practices and Religious Coping p < .05 with r s of .19 and .18 (small effect sizes; Sink & Stroh, 2006) Conversely, education level did not correlate with Daily Sp iritual Experiences or Organizational Religiousness Of the three demographic v ariables, age demonstrated a positive association with low fat eating (EBPQ), p < .01 (i.e., increased age was associated with increased engagement in low fat eating practices), and education level demonstrated a positive association with nutrition (HPLP), p < .05 (i.e., increased education was associated with increased engagement in healthy eating behaviors), with r s of .28 and .18 (small effect sizes; Sink & Stroh, 2006). GLHC Scores The mean total score of the God Locus of Health Control scale (GLHC) wa s 17.3 ( SD = 8.7), which indicates that on average, participants slightly disagreed with the statements about God controlling their health. However, the most frequently occurring th. In fact 17 there was considerable variation within the sample, with total scores covering the entire range of possible scores, from 6 to 36. When scores were divided into fou r categories to estimate frequencies of opinions, it appeared that approximately 58% of the sample expressed disagreement with GLHC, 11% expressed relatively neutral responses, 16% slightly to moderately agreed with GLHC, and 14% expressed strong agreement with GLHC. Thus, overall, approximately 30% of the sample expressed some degree of positive endorsement of GLHC.
77 Mean endorsement of GLHC by participants in this study is considerably lower as compared to some other studies examining GLHC among African A merican or Black adults (e.g., Kinney et al., 2002; Lutz, Yehieli, & Gonnerman, 2004) For example, a small study of African American females in rural Louisiana reported an average GLHC score of 26 (Kinney et al., 2002). A dditionally, a study by Lutz and c olleagues (2004) that involved a sample of Black and/or African American males and females living in Iowa who predominantly reported frequent church attendance found a mean GLHC score of 26.7 total and 26.0 for men only (as opposed to 17.3 in the present s tudy). Although the scores from the Lutz et al. study cannot be considered an exact comparison because a different measure of God locus of health control was used [i.e., the God Control Scale (Welton et al., 1996) vs. the God Health Locus of Control scale (Wallston et al., 1999) used in the present study], both measures use the same 6 point response scale and are 6 item scales with similar items. Reasons for the comparatively lower mean GLHC score in this sample as compared to these reference samples are u nknown However, in comparing these two studies with the present study, it is important to take into account substantial differences in sample demographics. For example, d espite the similar racial/ethnic make up of the sample from the Lutz et al. study and the sample from the present study t he majority (53%) of respondents in the Lutz et al. study reported an annual household income of less than $15,000, whereas the median annual household income range reported in the present study was $50,000 to $59,000. Additionally, 66% of respondents in the reference study completed at least some college or technical school (Lutz et al., 2004),
78 whereas 85% of the present sample completed a t least this level of education, and participants in these two studies lived in di fferent areas of the country (Iowa vs. Florida). Results for the Research Questions Research Question 1 Research Question 1: Does engagement in healthy eating vary in assoc iation with age, education and/or income ? To examine this research question, two simultaneous multiple linear regressions were performed in which the predictor variables were age, education, and income, and the criterion variable was one of the two measures of healthy eating. In the first multiple regression, the Low Fat Eating subsc ale of the Eating Behavior Patterns Questionnaire (EBPQ) was the criterion variable, and in the second multiple regression, the Nutrition subscale of the Health Promoting Lifestyles Profile II (HPLP) was the criterion variable. Results of the regression eq uations (Table 4 2) indicated that age accounted for unique positive variance in low fat eating ( = .27, p < .01), but income and education were not associated with low fat eating. The overall model of age, education, and income as predictors of low fat eating was significant, F (3, 139) = 3.69, p < .05, with 8% of total variance explained (a medium e ffect size; Sink & Stroh, 2006). Conversely, in the second multiple regression, none of the three demographic variables accounted for unique variance in nutritious eating as measured by HPLP. Research Question 2 Research Question 2 : Does degree of belie f in God locus of health control (GLHC) vary in association with age, education and/or income ? To examine this research question, a simultaneous multiple regression was performed in which the predictor variables were age, education, and income, and the cr iterion variable was
79 GLHC. Results of the regression equation (Table 4 3) indicated that none of these three demographic variables contributed unique variance in God locus of health control. In each of the previously mentioned analyses, collinearity stati stics (i.e., VIF and tolerance statistics) were within recommended limits (Field, 2005), suggesting that multicollinearity was not a concern. Research Question 3 Research Question 3: Are the four dimensions of religiosity/spirituality examined in this st udy (i.e., o rganizational religiousness, private religious practices, religious coping, and daily spiritual experiences) significantly associated with healthy eating? To examine this research question two simultaneous linear regressions were conducted wit h the four religiosity/spirituality dimensions as the predictor variable s and healthy eating as the criterion variable. In one of the regressions, healthy eating was operationalized with the EBPQ subscale and in the other regression, healthy eating was op erationalized with the HPLP subscale Collinearity statistics did not suggest cause for significant concern regarding multicollinearity (Field, 2005). For the regression in which healthy eating was operationalized using HPLP, daily spiritual experiences ( DSE) was the only one of the four religiosity/spirituality dimensions that accounted for a significant amount of unique variance in nutritious eating ( = .33, p < .01). DSE was positively related to nutritious eating, and in fact, was the only one of the dimensions with a positive B weight. T he overall model (i.e., including all four dimensions) predict ed a significant amount of the variance in nutritio us eating, F (4, 135) = 3.13, p < .05, and demonstrated a medium effect size ( R 2 = .09, Adj. R 2 = .06; Sink & Stroh, 2006).
80 For the regression equation u sing EBPQ to measure healthy eating (low fat eating) none of the four religiosity/spirituality dimensio ns predicted a significant amount of unique variance in low fat eating. However, the association between daily spiritual experiences and low fat eating approached significance ( = .20, p = .07 ) In sum, it appears that overall, of the four examined religiosity/spirituality dimensions, only daily spiritual experiences was significantly associated with healthy eating; in other words, healthy eating was not significantly associated with organizational religiousness, religious coping, nor private religious practices (Table 4 4). Results for the Hypotheses Hypothesis 1 Hypothesis 1 stated that when controlling for age, education, and income God locus of health control (GLHC) would have a significant negative association with healthy eating To test this hypothesis, two hierarchical linear regressions were conducted with GLHC as the predictor variable ; age, education, and income as covariates; and healthy eating as the criterion var iable. In one of the regressions, healthy eating was operationalized with the EBPQ subscale and in the other regression, healthy eating was operationalized with the HPLP subscale For both regression equations, collinearity statistics (i.e., VIF and tol erance statistics) were within recommended limits (Field, 2005), suggesting that multicollinearity was not a concern. Results of the regressions suggested partial support for Hypothesis 1 ( Table 4 5). For the regression in which healthy eating was operatio nalized using HPLP, GLHC accounted for unique negative variance in hea lthy eating ( = .17, p < .05); however, t he overall model with GLHC and the demographic
81 covariates (i.e., age, income, education) did not predict a significant amount of the variance i n healt hy eating. For the regression equation u sing EBPQ to measure healthy eating results indicated that GLHC accounted for unique negative variance in healthy eating, above and beyond the effects of age, education, and income ( = .23, p < .01) In fa ct, the addition of GLHC resulted in a significant increase in variance explained in healthy eating (i.e., an additional 5% variance) in step 2 of the model as compared with step 1 (demographic variables only) Of the three demographic variables, age expla ined significant unique variance in healthy eating ( = .29, p < .01) but education and income did not. T he overall model with GLHC and the demographic covariates predict ed a significant amount of the variance in healt hy eating, F (4, 139) = 5.01, p < .01, and demonstrated a medium effect size ( R 2 = .13, Adj R 2 = .10; Sink & Stroh, 2006). Overall, Hypothesis 1 was partially supported, in that GLHC predicted negative variance in healthy eating when controlling for age, education, and income when healthy eating was operationalized with EBPQ but not when health y eating was operationalized with HPLP. Hypothesis 2 Hypothesis 2 stated that after controlling for the effects of demographic variables ( age, education, and income ) God locus of health control (GLHC) would moderate the relationship between religiosity/ spirituality dimensions (i.e., organizational religiousness, private religious practices, religious coping, and daily spiritual practices) and healthy eating. Specifically, it was hypothesized that there would be (a) a significant negative relationship bet ween dimensions of religiosity/spirituality and healthy eating among men who ascribe to a high GLHC, and (b) a significant positive association between
82 dimensions of religiosity/spirituality and healthy eating among men who ascribe to a low GLHC. To test h ypothesis 2, SEM was conducted using AMOS 18.0 (Arbuckle, 2006) with maximum likelihood estimation. Given the difficulty in measuring engagement in healthy eating behaviors via brief self report surveys, two measures (i.e., instead of one) were used ( i.e ., HPLP and EBPQ) in this study. SEM provided a means of determining a single factor from the se two measures. Differences in findings from the previously described regression analyses which examined associations with these two measu res of healthy eating separately factor, particularly for the purpose of examining this slightly more complex moderation hypothesis. In addition to the healthy eating factor, the SEM model included the four examined dimensions of religiosity/spirituality, GLHC, and four interaction terms which were created with GLHC and each of the four religiosity/spirituality dimensions ( i.e ., R eligious C oping x GLHC, O rganizational R eligiousness x GLHC Private Re ligious Practices x GLHC, Daily Spiritual Experiences x GLHC ) Age, education, and income were included as covariates. Thus, th e model included a total of one latent variable (i.e., healthy eating, determined from two measures) and 1 2 additional observed v ariables (i.e., four religiosity dimensions, GLHC, four interaction terms, and three demographic variables). See F igure 4 1 for a graphical representation. All of the observed variables (indicators) were centered to reduce multicollinearity. T aking into account a range of fit indices has been recommended over making a determination of model fit based on nonsignificance of chi square alon e, particularly
83 g iven that the chi square statistic is sensitive to sample size (i.e., small samples may provide less po Bryant & Yarnold, 1995) and because there is no clear agreement on appropriate sample size nor acceptable fit criteria ( Hooper, Coughlan, & Mullen, 2008; Weston & Gore, 2006). Researchers have suggested using t he Comparative Fit Index (CFI) over other indices (e.g., chi square) when the sample size is smaller (Fan Thompson, & Wang, 1999; Hooper et al. 2008; Lei & Lomax, 2009) The Root Mean Square Error of Approximation (RMSEA ; Fan et al. 1999; Hooper et al. 2008) and the Standa rdized Root Mean Residual (SRMR; Hooper et al. 2008) are also recommended Thus, in addition to chi square, these three fit indices (i.e., CFI, RMSEA, and SRMR) were examined for determining model fit. In deciding which guidelines to use for determining acceptable fit of a model, Weston and Gore (2006) indicate d that sample size and model complexity should be considered applying more stringent criteria for large samples and less stringent crit eria for highly complex models. For CFI, criteria for fit (Weston & Gore, 2006) and for RMSEA and SRMR, recommended criteria range from Maxwell, 1999; Weston & Gore, 2006). Following recommendations to use a two step procedure for SEM (Kline, 2005; Muthn & Muthn, 2010; Weston & Gore, 2006), the adequacy of the observed indicators ( i.e., EBPQ and HPLP) in measuring their latent construct (healthy eating) was first assessed by testing the fit of the mea surement model. Next, the structural model was tested in order to evaluate the hypothesized relations.
84 Measurement m odel Fit index values for the measurement model demonstrated adequate fit, 2 (11, N = 140) = 16.57, p = .12, CFI = .99, RMSEA = .06 (90% CI: .00, .12), and SRMR = .03. Additionally, the two indicators (i.e., HPLP and EBPQ) loaded positively and significantly as would be expected, suggesti ng that the construct is adequately measured by the two observed variables. Unstandardized and standardized loadings for the measurement model are shown in Table 4 6. Structural m odel Next, the structural model ( Figure 4 1 ) was tested to evaluate the uni que relationships of the four religiosity/spirituality dimensions (i.e., organizational religiousness, private religious practices, religious coping, and daily spiritual experiences), GLHC, and the four interaction terms (e.g., R eligious C oping x GLHC), wi th healthy eating, including age, education, and income in order to covary out variance from each of these Results of structural equation modeling showed that the proposed model yielded adequate fit, 2 (67, N = 140) = 48.07, p = .13, CFI = .98, RMSEA = .0 4 (90% CI: .00, .08), and SRMR = .07. This model accounted for 28% of the variance in healthy eating. Main effects. Daily spiritual experiences was positively and uniquely related to healthy eating ( = .30, p < .01). Additionally, GLHC was negatively an d uniquely related to healthy eating ( = .23, p < .01), providing further support for Hypothesis 1. The other three religiosity/spirituality dimensions (i.e., organizational religiousness, religious coping, and private religious practices) did not demons trate significant associations with healthy eating, although the association between organizational religiousness and healthy eating approached significance ( = .19, p = .08) and was
85 negative. All unstandardized and standardized loadings for the hypothes ized structural model are shown in Table 4 7 Interaction effects. Two of the interactions Religious Coping x GLHC and Private Religious Practices x GLHC were positively and uniquely related to healthy eating ( = .24, = .27, respectively; p < .05). Add itionally, a moderately significant trend was observed in the effect of the Organizational Religiousness x GLHC interaction on healthy eating ( = .22, p = .059). The Daily Spiritual Experiences x GLHC interaction and healthy eating were not significantly related. Interpreting interaction effects can be particularly challenging, so to assist in decomposing interaction effects, general interpretations were made based on visual examination of graphs by using EBPQ as an approximation of the healthy eating con struct and graphing the relationship between healthy eating and the religious dimensions for the lower and upper quartiles of GLHC scores (i.e., the lowest 25% of GLHC scores and the highest 25% of GLHC scores). Alternative visual representations of the in teraction effects are also provided by the graphed relationships between healthy eating and GLHC for the lower and upper quartiles of each religiosity dimension. Religious C oping x GLHC Examining the graphical representation of the relationship between h ealthy eating (y axis) and religious coping (x axis) for individuals in the lower quartile versus the upper quartile of GLHC, it is apparent that healthy eating and religious coping were inversely related for individuals in the lower quartile of God locus of health control (i.e., healthy eating decrease d as engagement in religious coping increase d) ; however, for individuals in the upper quartile of GLHC, healthy eating and religious coping were positively associated (i.e., healthy eating increase d as
86 engage ment in religious coping increase d; Figure 4 2 ). Differences in healthy eating between participants in the lower and upper quartiles of GLHC were noticeable for those who engaged in low to moderate religious coping (i.e., those with low GLHC endorsed more healthy eating than those with high GLHC) but no differences were observed for those who engaged in the high est levels of religious coping (i.e., endorsement of healthy eating was about the same for individuals highly engaged in religious coping, regardle ). At most levels of religious coping, individuals who endorsed lower GLHC reported healthy eating more than individuals who endorsed higher GLHC; however, these differences disappeared at the highest level s of r eligious coping. General interpretations of the GLHC x Religious Coping interaction may alternatively be made by observing the graphed relationship between healthy eating and GLHC for participants at higher and lower levels of religious coping (Figure 4 3 ). Among individuals who endorsed a low to moderate degree of religious coping (i.e., religion is not involved in understanding or dealing with stressful situations to religion is somewhat involved in understanding or dealing with stressful situations), le vel of endorsement of GLHC was negatively associated with engagement in healthy eating behaviors (i.e., healthy eating decreased as belief in GLHC increased); however, among individuals who endorsed a high degree of religious coping (i.e., religion is very involved in understanding or dealing with stressful situations; 59% of the sample), no significant differences were observed in reported healthy eating behaviors in accordance with level of endorsement of GLHC.
87 Private R eligious P ractices x GLHC For in dividuals in the lowe r quartile of GLHC, levels of engagement in healthy eating did not appear to vary in association with levels of engagement in private religious practices ( PRP ) ; however, for individuals in the upper quartile of GLHC, engagement in heal thy eating appears to increase significantly as engagement in PRP increases (Figure 4 4 ). Differences in healthy eating between participants in the lower and upper quartiles of GLHC were notable for those who endorsed low to moderate engagement in PRP (i.e ., among individuals who endorsed low to moderate engagement in private religious practices, those with low GLHC tended to endorse higher engagement in healthy eating than those with high GLHC) but this difference disappeared at high levels of engagement i n PRP. In an alternative graphical representation, the relationship between healthy eating and GLHC for the lower and upper quartiles of private religious practices (Figure 4 5), it can be observed that among individuals in the lower quartile of reported engagement in private religious practices, level of endorsement of GLHC was negatively associated with engagement in healthy eating behaviors (i.e., healthy eating decreased as belief in GLHC increased); however, among individuals in the upper quartile of engagement in private religious practices, no significant relationship was observed between healthy eating and GLHC. Summary of results for hypothesis 2 In sum, GLHC was found to moderate the relationship between religiosity/spirituality and healthy eatin g for two of the four dimensions of religiosity/spirituality (i.e., religious coping and private religious practices). In this sense, Hypothesis 2 was partially supported; however, the specific hypothesized directional relationships were not supported. For these two religiosity/spirituality dimensions,
88 directions of the relationships among the variables were the opposite of what was hypothesized. In the case of religious coping, individuals with the highest endorsement of GLHC exhibited a positive relations hip between engagement in religious coping and engagement in healthy eating; conversely, individuals with the lowest endorsement of GLHC exhibited an inverse relationship between engagement in religious coping and engagement in healthy eating. Similarly, i n the case of private religious practices, individuals with the highest endorsement of GLHC exhibited a positive relationship between engagement in religious practices and engagement in healthy eating (i.e., similar to the association found for religious c oping); however, for individuals with the lowest endorsement in GLHC, no relationship was found between private religious practices and healthy eating. Individuals least likely to engage in healthy eating behaviors appeared to be those who reported low eng agement in religious practices/coping yet endorsed a high level of belief that God controls their health.
89 Table 4 1 Descriptive statistics and bivariate c orrelations among the variables of interest. Variable 1 2 3 4 5 6 7 8 9 10 1. God locu s of health control .08 .11 .18* .09 .18* .22** .08 .06 .06 2. Daily spiritual experiences .08 .61** .44** .45** .12 .19* .02 .14 .10 3. Private religious practices .11 .61** .66** .52** .11 .12 .12 .19* .02 4. Organized religiousness .18* .44** .66** .46** .14 .02 .14 .09 .03 5. Religious coping .09 .45** .52** .46** .10 .03 .02 .18* .02 6. Healthy eating (HPLP) .18* .12 .11 .14 .10 .53** .04 .18* .09 7. Hea lthy eating (EBPQ) .22** .19* .12 .02 .03 .53** .28** .08 .04 8. Age .08 .02 .12 .14 .02 .04 .28** .02 .26** 9. Education .06 .14 .19* .09 .18* .18* .08 .02 .37** 10. Income .06 .10 .02 .03 .02 .09 .04 .26** .37** M 17.26 29.86 29.61 8.44 3.41 2.48 45.31 SD 8.67 4.99 7.61 2.74 .82 .50 8.94 Possible Range 6 36 6 36 5 40 2 12 1 4 1 4 14 0 Note. Higher scores indicate higher levels of the variables assessed HPLP = Health Promoting Lifestyle Profile II; EBPQ = Eating Behavior Patterns Questionnaire. p < .05; ** p .01
90 Table 4 2. Relations of demographic variables with healthy eating m easures Note. p < .05; ** p < .01. Table 4 3. Relations of demographic variables with God locus of health c ontrol B SE B R 2 Adj. R 2 F df .02 .01 0.70 3, 139 Constant 17.14 4. 63 Education 0.74 0.77 .09 Age 0.05 0.07 .06 Income 0.23 0.27 .08 Note. No significance at p < .05. B SE B R 2 Adj. R 2 F df HPLP Regressed onto Demographic Variables .03 .01 1.45 3, 139 Constant 2.01 0.27 Education 0.0 8 0.04 .16 Age 0.00 0.00 .03 Income 0.00 0.02 .02 EBPQ Regressed onto Demographic Variables .08 .06 3.69* 3, 139 Constant 32.54 4.63 Education 0.83 0.77 .10 Age 0.21 0.07 .27** Income 0.19 0.27 .06
91 Table 4 4. Relations of religiosity/spirituality dimensions and healthy eating. B SE B R 2 Adj. R 2 F df Religiosity/Spiritu ality Dimensions Regressed onto HPLP Model .09 .06 3.13* 4, 135 Constant 0.01 0.04 DSE 0.03 0.01 .33** PRP 0.01 0.01 .17 OR 0.02 0.02 .13 Cope 0.06 0.06 .09 Religiosity/Spirituality Dimensions Regressed onto EBPQ Model .05 .03 1.91 4, 135 Constant 45.30 0.75 DSE 0.36 0.19 .20 PRP 0.15 0.15 .13 OR 0.54 0.37 .17 Cope 0.52 1.10 .05 Note. DSE = Daily spiritual experiences; PRP = Private religious practices; OR = Organizational religiousness; Cope = Religious coping. p < .05; ** p < .01.
92 Ta ble 4 5 Relations of Go d locus of health control and healthy eating. B SE B R 2 Adj. R 2 F df R 2 change GLHC Regressed onto HPLP Step 1 .03 .01 1.45 3, 139 Constant 2.01 0.27 Education 0.08 0.04 .16 Age 0.00 0.00 .03 Income 0.00 0.02 .02 Step 2 .06 .03 2.19 4, 139 .03* Constant 2.18 0.28 Education 0.07 0.04 .15 Age 0.00 0.00 .04 Income 0.01 0.02 .04 GLHC 0.01 0.01 .17* GLHC Regressed onto EBPQ Step 1 .08 .06 3.69* 3, 139 Constant 32.54 4.63 Education 0.83 0.77 .10 Age 0.21 0.07 .27** Income 0.19 0.27 .06 Step 2 .13 .10 5.01** 4, 139 .05** Constant 36.68 4.73 Education 0.65 0.76 .08 Age 0.22 0.06 .29** Income 0.14 0.27 .05 GLHC 0.24 0.08 .23** Note. GLHC = God locus of health control. p < .05; ** p < .01.
93 Table 4 6 Loadings for the measurement model. Parameter estimate Unstandardized loading SE Z Standardized loading HE HPLP .03 .01 4.14 .56** HE EBPQ 1.00 .94 Note. HE = healthy eating (latent construct). ** p < .01. Table 4 7 L oadings on to healthy eating in the hypothesized structural model. Indicator Unstandardized loading SE Z Standardized loading Religious coping .38 1.0 0 0.38 .04 Private religious practices .09 .14 0.66 .08 Daily spiritual experiences .50 .17 2.93 .30** Organizational religiousness .57 .17 1.75 .19 GLHC .22 .08 2.70 .23** GLHC x Religious Coping .29 .12 2.31 .24* GLHC x PRP .03 .02 2.08 .27* GLHC x DSE .02 .02 .95 .11 GLHC x OR .07 .04 1.89 .22 Age .20 .06 3.43 .28*** Education .78 .69 1.14 .10 Income .16 .24 .68 .06 Note. PRP = Private Religious Practices; DSE = Daily Spiritual Experiences; OR = Organizational Religiousness. p < .05; ** p < .01; *** p < .001.
94 Figure 4 1. Structural model of relations between var iables of interest and healthy eating. Standardized path coefficients are shown. Measurement errors are omitted for clarity. PRP = Private religious practices; DSE = Daily spiritual experiences; OR = Organizational religiousness; COPE = Religious coping; G LHC = God locus of health control; HPLP = Nutrition subscale of Health Promoting Lifestyles Profile II; EBPQ = Low Fat Eating subscale of Eating Behavior Patterns Questionnaire. Dashed lines indicate nonsignificant paths. Solid lines indicate p p p < .001
95 Figure 4 2. Linear relationships between religious coping and healthy eating at the upper and lower quartiles of endorsement of God locus of health control (GLHC).
96 Figure 4 3. Linear relationships between God locus of hea lth control and healthy eating behaviors at two levels of engagement in religious coping.
97 Figure 4 4 Linear relationships between private religious practices and healthy eating at the upper and lower quartiles of endorsement of God locus of health cont rol (GLHC).
98 Figure 4 5. Linear relationships between God locus of health control and healthy eating behaviors at two levels of engagement in private religious practices.
99 CHAPTER 5 DISCUSSION The pr esent study collected cross sectional data on religious /spiritual belie fs, healthy eating behaviors, and demographic characteristics from a community based sample of adult males who identif y as Black. The general purpose of the present study was to examine the relationships among dimensions of religiosity/spir ituality, God locus within a community based sample of Black men. A specific purpose was to examine GLHC as a potential m oderator of the relationship s between religiosi ty/spirituality dimensions and engagement in healthy eating. In this study, r eligiosity / spirituality was conceptualized multidimensionally, as has been recommended (e.g., Idler et al., 2003, Seybold & Hill, 2001) The common unidimensional measurement of r eligiosity ( Koenig et al., 2001), such as frequency of attendance at religious services, often provides an over emphasis on organizational religiousness and an under emphasis on more personal or non institutional practices or perspectives The creators of the Multidimensional Measure of Religiousness/Spirituality (Fetzer Institute, 1999) acknowledged that while some consider religiousness and behavioral, social, doctr inal, and denominational characteristics because it involves a with the assumption that 2). These authors note that it is possible to be spiritual without being religious and that, although spirituality is often a central aspect of religious involvement, it is possible to be
100 rel igious without being particularly spiritual (e.g., to practice religious doctrine without feeling a relationship to the transcendent/God/a higher power; Fetzer Institute, 1999). Four dimensions of religiosity/spirituality were assessed in this study: (1) o rganizational religiousness (i.e., frequency of attendance at religious services or other activities at a place of worship) (2) private religious practices (i.e. non institutional religious/spiritual practices such as praying or reading religious litera ture), (3) religious coping (i.e., involving religion in understanding or dealing with stressful situations), and (4) daily spiritual experiences (i.e., the perception of involvement with God and spirituality in daily life experiences ) This chapter presen ts a summary and interpretation of results from this study, limitations of the study, directions for future research and intervention, implications for counseling psychologists, and conclusions. Summary and Interpretation of Results The present study is n ovel both because of the topic of examination and the population of focus (i.e., Black men represent a demographic that both experiences significant health disparities and tends to be underrepresented in health research). This study attempts to expand the research literature for two areas in which there is a paucity of information: (a) the relationship between religiosity/spirituality and healthy eating behaviors and (b) the construct of God locus of health control (i.e., the degree to which one believes th the relationship between religiosity and health (broadly defined); however, little research has examined potential associations between religiosity and healthy eating in particular. Res earch examining God locus of health control, including potential associations between this construct and health behaviors, is also limited. The results from the
101 specific hypotheses and research questions examined in this study and the interpretations of th ese results are presented in the following sections. Research Question 1: Demographic Associations with Healthy Eating Behaviors Research Question 1 examined if engagement in healthy eating behaviors varied in association with age, education, and income i n this sample of Black men. It was found that age was positively associated with healthy eating when healthy eating was defined by one of the two measures of healthy eating used in this study. Conversely, neither income nor education was found to be signif icantly associated with engagement in healthy eating. The finding that healthy eating increased with age within this sample of Black men is similar to the trend found in the nationally representative Third National Health and Nutrition Examination Survey ( NHANES III) that diet quality increased with age. However, unlike the findings of this study, NHANES III found that income and education were also positively associated with diet quality (Guo, Warden, Paeratakul, & Bray, 2004). A number of explanations for these differences is possible. Potential explanations include different methods of measuring income and education, differing demographic characteristics of the samples, and the possibility of lower variation in eating behaviors within a single gender and racial group. Research Question 2: Demographic Associations with God Locus of Health Control Given the limited research that has been conducted on the construct of God locus of health control (GLHC), it seemed relevant to examine if and how the demographi c variables of study (i.e., age, education, income) were related to this construct. Thus, Research Question 2 asked if degree of belief in GLHC var ies in association with age, education and/or income ? Findings from the present study did not
102 indicate signi ficant relationships between any of these demographic variables and GLHC. Findings from prior studies that examined associations between GLHC and these ( 2005) qualitati ve study involving African Americans with diabetes found that those who were most likely to turn their health over to God were individuals who were low income, less educated, and elderly. On the other hand, in a breast cancer screening study of African Ame rican women in rural Louisiana, GLHC was negatively correlated with age and not correlated with income nor education level (Kinney et al., 2002). In a study of African American Christians the majority of whom were low income health God control (i.e., a sim associated with age (and associations with education and income were not examined; Gonnerman et al., 2008). Conversely, a study by the authors of the GLHC scale that included three samples o f individuals with a chronic health condition the majority of whom we re female and European American found that (a) age was negatively correlated with GLHC in one sample but not in the other two samples, (b) level of education was negatively associated wit h GLHC in two of the three samples, and (c) income was not significantly related to GLHC The researchers also found no gender differences in GLHC (Wallston et al., 1999). Thus, findings from prior studies examining associations between GLHC and these demo graphic variables are diverse and inconsistent.
103 Research Question 3: The Relationship Between Religiosity/Spirituality and Healthy Eating Behaviors Although a great deal of research has found significant associations between religiosity/spirituality and h ealth, including many studies that report a positive relationship between these variables, little of this research has focused on health behaviors (i.e., as opposed to health status, health outcome s, or mortality). Among the few studies found that have exa mined potential associations between religiosity and healthy eating, results have been mixed, including whether or not an association exists at all (e.g., Homan & Boyatzis, 2010; Obiesesan et al., 2006; Underwood & Powell, 2006). Research Question 3 asked if the four dimensions of religiosity/spirituality examined in this study (i.e., organizational religiousness, private religious practices, religious coping, and daily spiritual experiences) are significantly associated with healthy eating. Results found that only one of the four religiosity/spirituality dimensions explained unique variance in healthy eating; daily spiritual experiences (i.e., the perception of spirituality or interaction with/ involvement of God in daily life experiences ) was positively an d uni quely related to healthy eating. On the other hand, no associations with healthy eating were found for private religious practices, organizational religiousness, nor religious coping. Overall, findings suggest that spirituality not religiosity is posi tively related to engagement in healthy eating among this sample of Black men. Hypothesis 1 : The Relationship Between GLHC and Healthy Eating God locus of health control the degree to which one believes that God controls has been conceptualiz ed as a distinct dimension of locus of control
104 (Wallston et al., 1999; Welton et al. 1996 ). Although the amount of research on this construct has been limited, some studies have suggested that having a high God locus of health control (GLHC) is negatively associated with engagement in health behaviors (Kinney et al., 2002; Leaks, 200 9 ; or is not associated with health behaviors ( Reindl Benjamins et al., 2011; Gonnerman at al., 2008). It is notable that the couple of studies found that s uggest a positive relationship between GLHC and healthy behaviors/lifestyle used an operationalization of healthy behaviors/lifestyle that included mental health behaviors (e.g., stress management) in addition to physical health behaviors ( Gonnerman at al. 2008 ; Welton et al., 1996). The only study this author is aware of that has examined GLHC specifically in relation to healthy eating found no association between GLHC and healthy eating ( Gonnerman at al., 2008). Given prior research suggesting a negativ e association of GLHC with other health promoting behaviors, hypothesis 1 in the present study stated that God locus of health control would have a significant negative association with healthy eating above and beyond the effects of age, education, and in come. Hypothesis 1 was tested separately with two measures of healthy eating (i.e., HPLP and EBPQ). Findings from th e present study largely supported hypothesis 1. When healthy eating was operationalized with the HPLP measure, GLHC accounted for unique ne gative variance in healthy eating, and the model predicted 6% of the variance in healthy eating. Results were more significant when healthy eating was operationalized with EBPQ, in that the model predicted 13% of the variance in healthy eating, and GLHC sp ecifically explained an additional 5% of variance above and beyond that which
105 was explained by the demographic variables. Thus, findings suggest that as belief that This f inding is similar to findings from some other studies with samples that consist fully or predominantly of Black, African, or African American individuals, which found an inverse relationship between GLHC (or a closely related construct) and health behavior te cancer screening (Leaks, 200 9 ), and breast cancer screening ( Holt et al. 2003 ; Kinney at al. 2002). Hypothesis 2: GLHC as a Moderator of Religiosity/Spirituality and Healthy Eating Hypothesis 2 proposed that when controlling for age, education, and income, GLHC would moderate the rel ationship between dimensions of religiosity/spirituality and healthy eating. Structural Equation Modeling was used to test this hypothesis, as this analysis allow ed for the creation of a latent healthy eating variable that was determined from responses on the two healthy eating measures. The tested model accounted for 28% of the variance in healthy eating. Hypothesis 2 was partially supported, in that GLHC was foun d to moderate the relationship between religiosity/spirituality and healthy eating for two of the four dimensions of religiosity/spirituality: religious coping and private religious practices. For individuals with the highest endorsement of GLHC, healthy e ating increased as engagement in religious coping and private religious practices increased. Conversely, for individuals with the lowest endorsement of GLHC, healthy eating decreased as religious coping increased, but no relationship was found between heal thy eating and private religious practices. Another way these interaction effects can be understood is as follows: Healthy eating decreased as belief in GLHC increased among (a) individuals with the lowest
106 engagement in private religious practices and (b) individuals who endorsed a low to moderate degree of religious coping (i.e., religion is not involved to somewhat involved in understanding or dealing with stressful situations). Conversely, no significant relationship was observed between healthy eating a nd GLHC among (a) individuals who endorsed a high degree of religious coping (i.e., religion is very involved in understanding or dealing with stressful situations; 59% of the sample) and (b) individuals with the highest engagement in private religious pra ctices. Overall, within this sample of Black men from north/north central Florida, it appears that GLHC had little to no association with healthy eating behaviors among the men who endorsed the highest engagement in private religious practices and religio us coping; however, among the men with comparatively lower levels of private religious practices and religious coping, healthy eating decreased as GLHC increased. The individuals least likely to engage in healthy eating appeared to be those who reported lo w engagement in religious practices/coping yet endorsed a high level of belief that God controls their health. Limitations of the Present Study Findings from this study must be interpreted in light of some limitations. First, it is important to note that the findings of this study are certainly not generalizable to all Black men in the U.S., as this sample is relatively small, was geographically limited to north/north central Florida, and was highly educated. Sixty percent (60%) reported that they had com pleted college, and half of that 60% reported that they had completed graduate school as well. Also, these findings are of course not generalizable to Black or African American women, nor to men of other racial/ethnic groups.
107 Additionally, religious affil iation was not taken into account in this study, so it is not possible to know the exact make up of the sample with regard to religious affiliation, or if the findings may differ in accordance with religious affiliation. Although affiliation was included i n the demographic questionnaire, and nearly all participants endorsed some form of Christianity, the variability of responses from this write in item made this data difficult to quantify. Limitations of the present study include the cross sectional natur e of this data and shortcomings in relation to assessment. While data from a single time point can provide reports of their r eligiosity or eating habits might change based on time related factors, such as time of year or proximity to religious holidays. Also, of course no causal directions can be inferred from this cross sectional data. However, research testing casual relations between health and locus of control current health status (Luszczynska & Schwarzer, 2005). Similarly, the present study does not take into account health status nor ha ve a manner of assessing the duration of time an individual has engaged in his current level of healthy eating habits. Thus, it is impossible to consider information such as how health status might be related to God locus of health control, or for what per iod of time an individual has been engaging in his current eating behaviors. Such information on duration of current habits is difficult to obtain, but could help to answer questions such as: Is healthy eating his norm? Has he recently started following sp ecific recommendations in order to combat a diagnosed health condition?
108 Given the intent to collect information on eating habits without requiring time questionnaires wit h demonstrated reliability and validity were utilized in the present study. However, the potential limitations to self report data in the assessment of daily behaviors (e.g., recall reliability, social desirability bias) are important to note, particularly with regard to eating habits. ultidimensional assessment of religiosity and spirituality can be perceived as a significant improvement over the historically common practices of operationalizing religiosity as religious affiliation only (Koe nig et al., 2001) or using brief or single item global measures such as frequency of church attendance or self rated religiousness and spirituality (Hill & Pargament, 2003) Even so, the subscales used in the present study may be limited by their brevity, and one of the four subscales (i.e., religious coping) was comprised of only a single item. Additional information relevant to informing health behavior intervention programs might have been collected through inclusion of a measure of internal locus of h ealth control. Although the authors of the God Locus of Health Control scale approve the use of this measure either alone or in conjunction with their other locus of health control measures (Wallston et al., 1999), it is possible that inclusion of a measur e of internal locus of health control in the tested model may have further elucidated the relationships among the variables of study, possibly by delineating the effects of God locus of health control (GLHC) from other forms of locus of health control. How ever, findings from past research suggest that the relationship between internal locus of health control (or internal locus of control) and GLHC (or God locus of control) is currently unclear and
109 may vary with demographic characteristics. For example, GLHC beliefs were positively associated with internal locus of health control in a sample of low income African Americans (Gonnerman et al., 2008; Holt et al., 2003); similarly, God locus of control and internal locus of control were positively related in a sa mple of Black congregants who had an average of three years of college education (Jackson & Coursey, 1988). On the other hand, negative correlations were found between God health control and internal health control in a study of university students (race/e thnicity not provided; Welton et al., 1996). It is important to note that varying measures of GLHC were used in these studies. In sum, future research in this area may benefit from larger and more diverse samples (e.g., geographically diverse, educationall y diverse), multiple measures to assess each construct and more research to inform the selection of such measures, and the inclusion of other forms of health locus of control. Future Directions for Research and Interventions The current study lays some gr oundwork for future research examining potential associations between religiosity/spirituality variables and health behaviors, particularly among Black and/or African American adults. Although this study specifically examines healthy eating, similar resear ch could be conducted focusing on other health behaviors such as physical activity. It appears that the vast majority of research conducted thus far on the relationship between religiosity and health has examined health status variables (e.g., presence of disease, self rated health) or mortality rates, and that very little of such research has focused on health behaviors. Additionally, past religiosity health research seems to have focused more often on patient populations rather than community based sample s, the latter of which may be more relevant to the development of prevention and early intervention programs.
110 Given the high prevalence of chronic disease and the significant health disparities that disproportionately impact African American and Black ind ividuals ( NCHS, 2010 ; Roger et al., 2011; USDHHS, OMH, 2009 ), it is important that future health research includes a particular focus on this racial/ethnic minority group. Furthermore, given the trend of religious organizations housing health programs (Cam pbell et al., 2007; Trinitapoli et al., 2009), particularly within African American/Black communities, it is crucial that health researchers and individuals responsible for developing health promotion/intervention programs, develop an understanding of how religious and spiritual variables may (or may not) be associated with health promoting behaviors. Findings of the present study provide an example of the complexity of examining religiousness and spirituality, particularly in relation to health behaviors. In this study, some religiosity/spirituality variables were positively associated with healthy eating, while other religiosity/spirituality variables were negatively associated with healthy eating, and yet other religiosity/spirituality variables demonstr ated no associations with healthy eating at all. Further research examining which religious and/or spiritual factors may be motivators of or barriers to health behavior change could assist in determining the potential for the inclusion of religious compon ents in faith based intervention programs. If religious settings are to continue to be utilized as sites for health promotion and intervention programs, it is important that researchers explore ways in which religiosity or spirituality might impact the ou tcomes of such programs or be incorporated into such programs to increase their efficacy. Future research could elucidate how health promotion messages and religiosity messages could be combined to effectively promote
111 health, and if incorporating such mess ages would provide an added benefit to such commonly applied to efforts to promote abstinence from alcohol and drugs, yet such a message clearly holds potential for increasing h ealth promoting behaviors such as healthy eating as well. In order for such an approach to be conducted in a culturally sensitive manner, of course it will be crucial to include religious leaders in the planning and implementation of such programs. [For a discussion of incorporating cultural and spiritual sensitivity into community based health promotion research, see writings such as Campbell et al., 2007]. Particularly in relation to African American/Black communities, it is also important to consider tha t the influence of pastors in these communities is significant. The influence of such pastors often extends beyond serving as spiritual leaders to serving as potential agents of health behavior changes (Levin, 1986). There may be some religious or spirit ual beliefs that impact health behaviors (positively or negatively) beliefs that could be influenced by religious leaders. For engagement in healthy eating (except i n the case of the most highly religious individuals). If further research supports this finding, then addressing this issue in a culturally and spiritually sensitive manner through collaborations with religious leaders in African American/Black communities could potentially lead to increased involvement of members of these communities in healthy eating. spirituality had differing associations with health behavior, future resear ch in this area
112 should include a multidimensional assessment of religiosity/spirituality, as has been previously recommended by other researchers (Idler et al., 2003; Seybold & Hill, 2001), with health variables. Also, for the purposes of health related research, it is important that spirituality is specifically assessed in addition to religiosity (Weaver et al. 2006). Finally, future research should further examine the construct of God loc us of health control (GLHC) and determine how this construct may relate to engagement in health behaviors. Polzer and Miles (2005) wrote about the distinction between turning to God versus turning it over to God ; future research may benefit from a measure intended for use in health research that specifically provides a way of distinguishing between these two related but meaningfully distinct constructs. Given that studies examining GLHC and related constructs (e.g., religious health fatalism) are currently limited in number, research comparing measures that assess GLHC and similar constructs is especially needed to elucidate which measure(s) are most relevant for health related research. It is additionally pertinent that future research assesses which multi dimensional religiosity/spirituality measures are most relevant for health research, while taking into account the benefits of brief measurement in community based research. Finally, it is interesting to note that the limited associations found in this st udy between the four examined religiosity/spirituality dimensions and healthy eating are in contrast to the plethora of research that suggests a protective effect of religiosity on mortality and/or chronic disease (George et al., 2000; Gillum & Ingram, 200 6 ; Powell et al., 2003; Williams & Sternthal, 2007). Thus, future research on the religion health
113 connection (Ellison & Levin, 1998) should compare the relationships between religiosity and health status/mortality variables with the relationships between r eligiosity and health behaviors. Such research could determine if and how religiosity may be differentially associated with varying operationalizations of health and could examine the underlying mechanisms that explain the nature of such relationships. Imp lications for Counseling Psychologists The health focus of the present study is highly relevant for the field of counseling psychology, as this field is making increasing contributions to health psychology. As stated in a recent publication by Raque Bogdan Torrey, Lewis, and Borges (2012): Counseling psychologists are poised to make unique contributions to standing focus on wellness and prevention; a holistic, developmental, and strengths based approach to individual health (Chwalisz & Obasi, 2008); and multiculturalism (Eppserson, Fouad, Stoltenberg, & Murdock, 2005). (p. 2) Indeed, this position is reinforced by (a) the existence of specialized sections such as the Counseling Health Psychology section and the Prevention section in the publications such as the 2007 Major Contribution in The Counseling Psychologist that focused on the roles of counseling psychologists i n addressing health disparities through culturally sensitive health care research and interventions (e.g., Tucker et al., 2007). This study is a prime example of counseling health psychology research, in that it examines associations between psychosociocu ltural factors (i.e., religiosity/spirituality variables) and a behavioral factor (i.e., healthy eating) that relates to multiple chronic health conditions. Given that eating behaviors are associated with many of the most common chronic health conditions i mpacting our country, it is pertinent that counseling
114 psychologists (a) elucidate the psychosociocultural factors that are associated with engagement (or lack of engagement) in healthy eating behaviors and (b) contribute to the development of research base d health promotion programs that take into account psychosociocultural factors. Given that approximately 89% of Americans identify with religiosity (Pew Forum on Religion & Public Life, 2008), it is sensible that health professionals develop a specific un derstanding of if and how religious and spiritual beliefs relate to health behaviors. Furthermore, given that African Americans/Black Americans are the most likely of all major racial and ethnic groups in the U.S. to report a formal religious affiliation ( Pew Forum on Religion & Public Life, 2008), it may be particularly important to consider religious factors when working with African Americans/Black Americans. If further research supports the existence of associations between religious/spiritual beliefs and health behaviors such as healthy eating, counseling psychologists could be particularly useful in determining how to increase the efficacy of health education, prevention, and intervention programs by incorporating religiosity/spirituality in a cultura lly and spiritually sensitive manner. Counseling psychologists also often work with clients/patients on an individual level to change health related behaviors, such as those who have been advised to alter their eating after diagnosis of a chronic health co ndition. The present research suggests that in such scenarios, it may be useful to ask clients/patients about their religious/spiritual beliefs and to assess how these beliefs relate for example, to their motivation to engage in healthy eating behaviors.
115 Conclusions Within the sample of African American/Black men from north/north central Florida in the present study, only one of the four examined dimensions of religiosity/spiritualit y demonstrated a significant association with healthy eating. Specifically, endorsement of daily spiritual experiences was positively related to engagement in healthy eating, but none of the other three dimensions (i.e., organizational religiousness, priva te religious practices, and religious coping) demonstrated a significant relationship with healthy eating. On the other hand, God locus of health control (GLHC) was found to be negatively associated with healthy eating. Furthermore, the relationships amo ng these variables demonstrated greater complexity when the interactions between GLHC and religiosity/spirituality dimensions were examined in association with healthy eating. Specifically, for individuals with low endorsement of private religious practice s/religious coping, engagement in healthy eating differed in accordance with level of endorsement of GLHC (i.e., those with low GLHC exhibited greater engagement in healthy eating than those individuals with high GLHC). On the other hand, among individuals with high endorsement of religious practices/coping, no differences were observed in engagement in healthy eating in accordance with GLHC. Thus, GLHC may be a relevant variable to consider in the context of healthy eating interventions and health researc h among Black men. One interpretation of these negatively associated with engagement in healthy eating behaviors, high religiosity may be protective against this asso ciation. Given the need for effective interventions to promote healthy eating, and given the trend of using religious institutions as a setting for
116 such interventions, it is important that future research and public health initiatives take into considerati on both the potential relationships between religiosity/spirituality variables and engagement in health behaviors.
117 APPENDIX A DEMOGRAPHIC DATA QUESTIONNAIRE Directions : Please fill in the blanks and answer the questions on this questionnaire. Please ans wer the questions honestly and to the best of your knowledge. The information you provide will be kept confidential. Do not write your name on this questionnaire. For questions that have bubbles ( O ), completely fill in the bubble beside the information t hat best answers the question. Filled in bubbles should look like this: What is your employment status? O Work full time O Work part time O Currently unemployed but looking for a job O Do not work (stay at home parent, retired, on disability, etc.). What is the highest level of education that you have completed ? O Elementary school O Junior high/middle school O High school or GED O Some college/Technical school O College O Graduate school What is your current relationship status? O Sin gle, living wi th a partner O Single, living without a partner O Married, living with your spouse O Married, not living with your spouse O Divorced or separated O Widow/Widower
118 What is your gender? O Female O Male What is your age ? _____________ Ho w tall are you? ________ feet and _________ inches What was your weight the last time you were weighed? _________ pounds In what country were you born ? ____________________________________________ In what country were your parents born ? ____ _________________________________ In what country have you spent most of your life? ________________________________ What is the combined annual income of all of the people living in your household? O Less than $10,000 O $60,000 to $69,999 O $10,000 to $19,999 O $70,000 to $79,999 O $20,000 to $29,999 O $80,000 to $89,999 O $30,000 to $39,999 O $90,000 to $99,999 O $40,000 to $49,999 O More than $100,000 O $50,000 to $59,999 How many people currently live in your household (including yourself)? O 1 O 4 O 7 O 2 O 5 O 8 O 3 O 6 O 9 or more
119 Have you been told by a doctor or other health professional that you have any of the following health problems? (Please bubble in all of the health problems that you have been told you have.) O H igh cholesterol O High blood pressure (Hypertension) O Overweight/obesity O Heart/Cardiovascular disease O Diabetes O Prostate Cancer O Asthma O Other Cancer (type: ___________) O Arthritis O Depression O Stress O Mental Illness O Other Health Problem(s) Not Listed (please list): ________________________ If you have been diagnosed with diabetes, what type of diabetes do you have? O I have not been diagnosed with diabetes. O Type 1 diabetes O Type 2 diab etes O Not sure what type of diabetes I have From your perspective, h ow important is it for you to lose weight? O N ot at all important O O nly a little important O M oderately important O V ery important Are you currently trying to lose weight ? O No. O Yes, I am trying a little. O Yes, I am actively trying, but it is not a major focus in my life right now. O Yes, I am actively trying. It is a major focus in my life right now.
120 Your ethnicity is: O African American of American origin (born and grew up in America) O African American of African origin (born in Africa but now an American citizen) O African American of Caribbean origin (born in one of the Caribbean Islands but now an American citizen) O African O Caribbean O Other (please state) ________________________________ Your current residence is classified as: O Rural (Country) O Urban (City) To what extent do you consider yourself a religious person ? O Very religious O Moderately religious O Slightly religious O Not religious at a ll To what extent do you consider yourself a spiritual person ? O Very spiritual O Moderately spiritual O Slightly spiritual O Not spiritual at all
121 Healthcare Services: Where do you usually receive your health care services ? O health care center/cl inic O private practice O hospital O health department O other (please specify): _______________________________________ medication), what portion do you usually pay? O All the bills O More than half of the bills O About half of the bills O Less than half of the bills O None of the bills How often do you have physical health examinations? O Every year O Every other year O Occasionally O Rarely O Never Do you h ave a regular doctor you see for your medical care? O Yes O No
122 Prostate Cancer Screening : I was tested for prostate cancer with the Digital Rectal Examination (DRE) within the last two years ( not including today). O Yes O No I was tested for prostate cancer with the Digital Rectal Examination (DRE) within the last year ( not including today). O Yes O No I was tested for prostate cancer with the PSA test within the last two years ( not including today). O Yes O No I was tested for prostate cancer with the PSA test within the last year ( not including today). O Yes O No THANK YOU FOR YOUR PARTICIPATION!
123 APPENDIX B GOD LOCUS OF HEALTH CONTROL SCALE Directions : Each item below is a belief statement about your health with which you may agree or disagr ee. For each item, please circle the bubble that represents how much you agree or disagree with that statement. Please try to answer EVERY ITEM and fill in ONLY ONE bubble per item. This is a measure of your personal beliefs; there are no right or wrong an swers. Your answers should look like this: Strongly Disagree Moderately Disagree Slightly Disagree Slightly Agree Moderately Agree Strongly Agree 1. If my health worsens, it is up to God to determine whether I will feel better again. 2. Most things that affect my health happen because of God. 3. God is directly responsible for my health getting better or worse. 4. Whatever happens to my health is 5. Whether or not my health improves is up to God. 6. God is in control of my health.
124 APPENDIX C BRIEF MULTIDIMENSIONAL MEASURE OF RELIGIOUSNESS/SPIRITUALITY (SELECTED SUBSCALES) Directions : The following questions deal with possible spiritual experiences. To what extent do yo u experience the following? Please shade in the circle beneath the answer you choose like this: Many times a day Every day Most days Some days Once in a while Never or almost never 1. presence. 2. I find strength and comfort in my religion. 3. I feel deep inner peace or harmony. 4. I desire to be closer to or in union with God. 5. me, directly or through others. 6. I am spiritually touche d by the beauty of creation.
125 More than once a week Every week or more often Once or twice a month Every month or so Once or twice a year Never 7. How often do you go to religious services? 8. Besides religious services, ho w often do you take part in other activities at a place of worship? 9. What is your current religious preference? More than once a day Once a day A few times a week Once a week A few times a month Once a month Less than once a mon th Never 10 How often do you pray privately in places other than at church? 11 Within your religious or spiritual tradition, how often do you meditate? 12 How often do you watch or listen to religious programs on TV or radio? 13 How often do you read the Bible or other religious literature? 14 How often are prayers or grace said before or after meals in your home?
126 Directions: Think about how you try to understand and deal with major problems in your life. To what extent is each of the following involved in the way you cope. Please shade in the circle beneath the answer you choose like this: A great deal Quite a bit Somewhat Not at all 15 I think about how my life is part of a larger spiritual force. 16 I work together with God as partners. 17 I look to God for strength, support, and guidance. Very involved Somewhat involved Not very involved Not involved at all 18 To what extent is your religion involved in understand ing or dealing with stressful situations in any way? Copy right 1999, Fetzer Institute
127 APPENDIX D HEALTH PROMOTING LIFESTYLE PROFILE II: NUTRITIO N SUBSCALE (REVISED) Directions : We want to know about your way of life and personal ha bits. Please answer each question as honestly as you can, and try not to skip any question. Show how often you do each behavior by filling in one circle for each question. Your answers should look like this: How often do you: Never Sometimes Often Routinely (Very Often) 1. Choose a diet low in fat, saturated fat, and cholesterol? 2. Limit use of sugar and food containing sugar? 3. Eat 6 11 servings of bread, cereal, rice, and pasta each day? 4. Eat 2 4 servings of fruit each day ? 5. Eat 3 5 servings of vegetables each day? 6. Eat 2 3 servings of milk, yogurt, or cheese each day? 7. Eat only 2 3 servings from the meat, poultry, fish, eggs, and nuts group each day? 8. Read labels to identify nutrients, f ats, and sodium content in packaged food? 9. Eat breakfast? 10. Add salt to your food? 11. Eat whole grains (such as whole wheat bread, whole wheat pasta, or brown/wild rice) each day? 12. Limit sugary drinks to no more than one pe r day (such as non diet soda, energy drinks, sports drinks, fruit drinks) ?
128 APPENDIX E EATING BEHAVIOR PATTERNS QUESTIONNAIRE: LOW FAT EATING SUBSCALE (REVISED) Directions : We want to know about your eating habits. Please answer each question a s honestly as you can, and try not to skip any questions. Show how much you agree with each behavior by filling in one circle for each question. Your answers should look like this: 1 Strongly Disagree 2 Disagree 3 Neutral 4 Agree 5 Strongly Agree 1) I reduce fat in recipes by substituting ingredients and cutting portions. 2) I am very conscious of how much fat is in the food I eat. 3) I use low fat food products. 4) I choose healthy foods to prevent heart disease. 5) I count fat grams. 6) I carefully watch the portion sizes of my foods. 7) When choosing fast food, I pick a place that offers healthy foods. 8) Fish and poultry are the only meats I eat. 9) I like to eat vegetables seasoned with fatty meat. 10) I eat meatless meals from time to time because I think that is healthier for me. 11) I try to limit my intake of red meat (beef and pork). 12) I buy snacks from vending machin es. 13) I take a shopping list to the store. 14) Instead of planning meals, I choose what is available and what I feel like eating.
129 LIST OF REFERENCES Aberne thy, A. D., Houston, T. R., Bjorck, J. P., Gorsuch, R. L., & Arno ld, H. L. Jr. (2009). Religiousness and prostate cancer screening in African American men. Journal of Psychosocial Oncology, 27 (3), 316 331. doi: 10.1080/07347330902979036 American Cancer Society. (2011). Cancer facts and figures for African Americans 2011 2012. Atlanta, GA: American Cancer Society. Retrieved from http://www.cancer. org/acs/groups/content/@epidemiologysurveilance/documents/d ocument/acspc 027765.pdf American Heart Association. (2011). Coronary Artery Disease: The ABCs of CAD Retrieved from http://www.heart.org/HEARTORG/Conditions/More/MyHeartand StrokeNews/Coronary Artery Disease --The ABCs of CAD_UCM_436416_ Article.jsp Basen Engquist, K., & Chang, M. (2011). Obesity and cancer risk: Recent review and evidence. Current Oncol ogy Reports, 13 (1), 71 76. doi: 10.1007/s11912 010 0139 7 Brown, D. R., & Gary, L. E. (1994). Religious involvement and health status among African American males. Journal of the National Medical Association 86 825 831. Bryant, F. B., & Yarnold, P. R. ( 1995). Principal components analysis and exploratory and confirmatory factor analysis. In L. G. Grimm & P. R. Yarnold (Eds.), Reading and understanding multivariate statistics (pp. 99 136). Washington, DC: American Psychological Association. Calle, E. E., Rodriguez, C., Walker Thurmond, K., & Thun, M. J. (2003). Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. New England Journal of Medicine, 348 (17), 1625 1638. Campbell, M. K., Hudson, M. A., Resnicow, K., Blakeney, N., Paxton, A., & Baskin, M. (2007). Church based health promotion interventions: Evidence and lessons learned. Annual Review of Public Health, 28, 213 234. doi: 10.1146/annurev. publhealth.28.021406.144016 Carter Pokras, O., & Baquet, C. (2002). Public Health Reports, 117, 426 434. Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Division of Adolescent and School Health. (2011). Health disparities. Retr ieved from http://www.cdc.gov/healthyyouth/ disparities/index.htm
130 Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Divis ion of Adult and Community Health. (2011). Health disparities among racial/ethnic populations. Retrieved from http://www.cdc.gov/nccdphp/dach/ chhep/disparities.htm Centers for Disease C ontrol and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Division of Diabetes Translation (2011). National diabetes fact sheet: National estimates and general information on diabetes and prediabetes in the United States. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention. Retrieved from http://www.cdc.gov/diabetes/pubs/factsheet11.htm Centers for Disease C ontrol and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Division of Diabetes Translation. (2010). Diabetes data and trends, national survelliance data. Retrieved from http://apps.nccd.cdc.gov/DDTSTRS/ NationalSurvData.aspx Chaplin, W. F., Davidson, K., Sparrow, V., Stuhr, J., Van Roosmalen, E., & Wallston, K. A. (2001). A structural evaluation of the expanded Multidimensional Health Locus of Contr ol Scale with a diverse sample of Caucasian/European, Native, and Black Canadian women. Journal of Health Psychology, 6 (4), 447 455. doi: 10.1177/13591053010006000407 Chida, Y., Steptoe, A., & Powell, L. H. (2009). Religiosity/spirituality and mortality: A systematic quantitative review. Psychotherapy and Psychosomatics, 78 (2), 81 90. doi: 10.1159/000190791 Chiu, L., Emblen, J. D., Hofwegen, L. V., Sawatzky, R., & Meyerhoff, H. (2004). An integrative review of the concept of spirituality in the health scien ces. Western Journal of Nursing Research, 26 (4), 405 428. doi: 10.1177/0193945904263411 Chou, C. P. & Bentler, P. M. (1995). Estimates and tests in structural equation modeling. In R. H. Hoyle (Ed.), Structural equation modeling: Concepts, issues and appli cations (pp. 37 55). Thousand Oaks, CA: Sage. Colantonio, A., Kasl, S. V., Ostfeld, A. M., & Berkman, L. F. (1993). Psychosocial predictors of stroke outcomes in an elderly population. Journal of Gerontology, 48 (5), S261 S268. Contrada, R. J., Goyal, T. M., Cather, C., Rafalson, L., Idler, E. L., & Krause, T. J. (2004). Psychosocial factors in outcomes of heart surgery: The impact of religious involvement and depressive symptoms. Health Psychology, 23, 227 238. Downey, R. G., & King, C. V. (1998). Missin g data in Likert ratings: A comparison of replacement methods. The Journal of General Psychology, 125 (2), 175 191. doi: 10.1080/00221309809595542
131 Ellison, C. G., Hummer, R. A., Burdette, A. M., & Benjamins, M. R. (2010). Race, religious involvement, and h ealth: The case of African Americans. In C. G. Ellison, & R. A. Hummer (Eds.), Religion, families, and health: Population based research in the United States (pp. 330 331). Piscataway, NJ: Rutgers University Press. Ellison, C. G., Hummer, R. A., Cormier, S ., & Rogers, R. G. (2000). Religious involvement and mortality risk among African American adults. Research on Aging 226 630 667. doi: 10.1177/0164027500226003 Ellison, C. G., & Levin, J. S. (1998). The religion health connection: Evidence, theory, and f uture directions. Health Education & Behavior, 25 (6), 700 720. doi: 10.1177/109019819802500603 Fetzer Institute, National Institute on Aging Workgroup. (1999). Multidimensional Measurement of Religiousness and Spirituality for Use in Health Research. Kala mazoo, MI: Fetzer Institute. Retrieved from http://www.acperesearch.net/ Fetzer_Book_extracts.pdf Fitchett, G., Rybarczyk, B. D., DeMarco, G. A., & Nicholas, J. J. (1999). The role of re ligion in medical rehabilitation outcomes: A longitudinal study. Rehabilitation Psychology, 44 (4), 333 353. doi 10.1037/0090 55184.108.40.2063 Franklin, M. D., Schlundt, D. G., McClellan, L. H., Kinebrew, T., Sheats, J., Belue, us fatalism and its association with health behaviors and outcomes. American Journal of Health Behavior, 31 (6), 563 572. Gao, S., Moktharian, P. L., & Johnston, R. A. (2008). Nonnormality of data in structural equation models. Transportation Research Reco rd: Journal of the Transportation Research Board, No. 2082, 116 124. Washington, DC: Transportation Research Board of the National Academies. doi: 10.3141/2082 14 George, L. K., Larson, D. B., Koenig, H. G., & McCullough, M. E. (2000). Spirituality and hea lth: What we know, what we need to know. Journal of Social and Clinical Psychology, 19 (1), 102 116. Giaquinto, S., Spiridigliozzi, C., & Caracciolo, B. (2007). Can faith protect from emotional distress after stroke? Stroke, 38 993 997. doi: 10.1161/01.STR 0000257996.26950.59 Gillum, R. F. (2006). Frequency of attendance at religious services and leisure time physical activity in American women and men: The Third National Health and Nutrition Examination Survey. Annals of Behavioral Medicine, 31 (1), 30 35. doi: 10.1207/s15324796abm3101_6 Gillum, R. F., & Ingram, D. D. (2006). Frequency of attendance at religious services, hypertension, and blood pressure: The Third National Health and Nutrition Examination Survey. Psychosomatic Medicine, 68 (3), 382 385. do i: 10.1097/01. psy.0000221253.90559.dd
132 Gonnerman, M. E., Lutz, G. M., Yehieli, M., & Meisinger, B. K. (2008). Religion and health connection: A study of African American, Protestant Christians. Journal of Health Care for the Poor and Underserved, 19 (1), 19 3 199. doi: 10.1353/hpu. 2008.0020 Guo, X., Warden, B. A., Paeratakul, S., & Bray, G. A. (2004). Healthy Eating Index and obesity. European Journal of Clinical Nutrition, 58, 1580 1586. doi: 10.1038/sj.ejcn.1601989 Hall, D. E., Meador, K. G., & Koenig, H. G. (2008). Measuring religiousness in health research: Review and critique. Journal of Religion and Health, 47, 134 163. doi: 10.1007/s10943 008 9165 2 Hargreaves, M. K., Schlundt, D. G., Buchowski, M. S., Hardy, R. E., Rossi, S. R., & Rossi, J. S. (1999 ). Stages of change and the intake of dietary fat in African American women: Improving stage assignment using the Eating Styles Questionnaire. Journal of the American Dietetic Association, 99 (1), 1392 1399. doi: 10.1016/S0002 8223(99)00338 7 Hawthorne, G., & Elliott, P. (2005). Imputing cross sectional missing data: Comparison of common techniques. Australian and New Zealand Journal of Psychiatry, 39, 583 590. doi: 10.1080/j.1440 1614.2005.01630.x Hill P. C., & Pargament, K. I. (2003). Advances in the conc eptualization and measurement of religion and spirituality: Implications for physical and mental health research. American Psychologist, 58 (1), 64 74. doi: 10.1037/0003 066X.58.1.64 Hixson, K.A., Gruchow, H.W., & Morgan, D. (1998). The relation between rel igiosity, selected health behaviors, and blood pressure among adult females. Preventive Medicine, 27 545 552. Holt, C. L., Clark, E. M., Kreuter, M. W., & Rubio, D. M. (2003). Spiritual health locus of control and breast cancer beliefs among urban Africa n American women. Health Psychology, 22 (3), 294 299. doi: 10.1037/0278 6220.127.116.114 Holt, C. L., & McClure, S. M. (2006). Perceptions of the religion health connection among African American church members. Qualitative Health Research, 16, 268 281. doi: 10.1177/1049732305275634. Homan K. J., & Boyatzis, C. J. ( 2010 ). Religiosity, sense of meaning, and health behavior in older adults. International Journal for the Psychology of Religion, 20 (3), 173 186. doi: 10.1080/10508619.2010.481225 Hooper, D., Cough lan, J., & Mullen, M. R. (2008). Structural equation modeling: Guidelines for determining model fit. The Electronic Journal of Business Research Methods, 6 (1), 53 60.
133 Huber, M., Knottnerus, J. A., Green, L., van der Horst, H., Jadad, A. R., Kromhour, id, H. (2011). How should we define health? Biomedical Journal, 343 (d4163), 1 3. doi: 10.1136/bmj.d4163 Hwu, Y J. (2001). The impact of concept of health on health behavior. Journal of Nursing Research, 9 (5), 166. doi: 10.1097/01.JNR.0000347574.45553.2e Id ler, E. I., Musick, M. A., Ellison, C. G., George, L. K., Krause, N., Ory, M. spirituality for health research: Conceptual background and findings from the 1998 General Social Survey. Research on Aging, 25, 327 365. doi: 10.1177/0164027503025004001 Institute of Medicine Committee on Health and Behavior: Research, Practice and Policy. (2001). Health and Behavior: The Interplay of Biological, Behavioral, and Societal Influences. Nationa l Academy Press: Washington, D.C. Retrieved from http://www.nap.edu/openbook.php?record_id=9838&page=1 Jackson, L. E., & Coursey, R. D. (1988). The relationship of God control and inter nal locus of control to intrinsic religious motivation, coping and purpose in life. Journal for the Scientific Study of Religion, 27 (3), 399 410. Kannan, S., Sparks, A. V., DeWitt Webster, J. Krishnakumar, A., & Lumen g J. ( 2010 ). Healthy eating and hara mbee: Curriculum Development for a culturally centered bio medically oriented nutrition education program to reach African American women of childbearing age. Maternal and Child Health Journal, 14 (4), 535 547. doi: 10.1007/s10995 009 0507 9 Kim, K. H., & S obal, J. (2004). Religion, social support, fat intake and physical activity. Public Health Nutrition, 7, 773 781. doi:10.1079/PHN2004601 Kinney, A. Y., Emery, G., Dudley, W. N., & Croyle, R. T. (2002). Screening behaviors among African American women at hi gh risk for breast cancer: Do beliefs about God matter? Oncology Nursing Forum, 29 (5), 835 843. Kline, R. B. (2005). Principles and practice of structural equation modeling (2 nd ed.). New York: Guilford. Koenig, H., George, L. K., Larson, D. B., Cohen, H J., & Blazer, D. G. (1998). The relationship between religious activities and blood pressure in older adults. International Journal of Psychiatry in Medicine 28 189 213. Larson, D. B., Larson, S. S., & Koenig, H.G. (2002). Mortality and religion/spiri tuality: A brief review of the research. The Annals of Pharmacotherapy, 36 (6), 1090 1098. doi: 10.1345/aph.1A438
134 Leaks, B. W. (2009). The influence of fatalistic control, God locus of health control, and religious coping on prostate cancer screening (Docto ral dissertation). Fuller Theological Seminary, School of Psychology, Pasadena, California. Retrieved from Dissertations & Theses: Full Text (Publication No. AAT 3385582). Levin, J. S. (1986). Roles for the Black pastor in preventive medicine. Pastoral Psy chology, 35 (2), 94 104. doi: 10.1007/BF01768709 Lutz, G. M., Yehieli, M., Gonnerman, M. E., Jr., with Philipp, M., Park, K H., Mayfield, J., & Jensen, N. (2004). Health care attitudes and service utilization among African Americans in Waterloo, Iowa. Cedar Falls, IA: University of Northern Iowa, Center for Social and Behavioral Research. Retrieved from www.csbs.uni.edu/dept/csbr/ pdf/BlackHawk_AA Health 2004.pdf Luszczynska, A., & Schwarzer, R. (2005). Multidimensional health locus of control: Comments on the construct and its measurement. Journal of Health Psychology, 10 (5), 633 642. doi: 10.1177.1359105305055307 Mardia, K. V. (1970). Measures of multivariate skewness and ku rtosis with applications. Biometrika, 57 (3), 519 530. doi: 10.1093/biomet/57.3.519 Masters, K. S. (2007). Religiosity/spirituality and behavioral medicine: Investigations concerning the integration of spirit with body. Journal of Behavioral Medicine, 30 (4) 287 289. doi: 10.10007/s10865 007 9116 5 Mayo Clinic. (2009). Prostate cancer prevention: What you can do. Retrieved from http://www.mayoclinic.com/health/prostate cancer prevention/MC00027 McAuley,W., Pecchoni, J., & Grant, L. (2000). Personal account of the role of God in health and illness among older African American and White residents. Journal of Cross Cultural Gerontology 15 13 35. doi: 10.1023/A:1006745709687 M cCullough, M. E., Hoyt, W. T., Larson, D. B., Koenig, H. G., & Thoresen, C. ( 2000 ). Religious involvement and mortality: A meta analytic review. Health Psychology, 19 (3), 211 222. doi: 10.1037 //0278 618.104.22.168 McDonald, R. P. & Ho, M. H. R. (2002). Prin ciples and practice in reporting structural equation analyses. Psychological Methods, 7, 64 82. doi: 10.1037//1082 989X.7.1.64. Mensah, G. A., Mokdad, A. H., Ford, E. S., Greenlund, K. J., & Croft, J. B. (2007). State of disparities in cardiovascular healt h in the United States. Circulation, 111 (10) 1233 1241. doi: 10.1161/01.CIR.0000158136.76824.04 Merrill, R. M., & Thygerson, A., L. (2001). Religious preference, church activity, and physical exercise. Preventive Medicine, 33, 38 45. doi: 10.1006/pmed.2001.0851
135 Moberg, D. (2002). Assessing and measuring spirituality: Confronting dilemmas of universal and particular evaluative criteria. Journal of Adult Development, 9, 47 60. doi: 10.1023/A:101387720 1375 Mokdad, A. H., Bowman, B. A., Ford, E. S., Vinicor, F., Marks, J. S., & Koplan, J. P. (2001). The continuing epidemics of obesity and diabetes in the United States. Journal of the American Medical Association, 286 (10), 1195 1200. doi:10.1001/jama.286. 10.1195 National Center for Health Statistics. (2010). Health, United States, 2009: With special feature on medical technology. (Library of Congress No. 76 641496). Retrieved from http://www.cdc.go v/nchs/data/hus/hus09.pdf National Center for Health Statistics. (2011). Health, United States, 2010: With special feature on death and dying. (Library of Congress No. 76 641496). Retrieved from h ttp://www.cdc.gov/nchs/data/hus/hus10.pdf National Stroke Foundation. (2012). African Americans and stroke. Retrieved from http://www.stroke.org/site/PageServer?pagename=aamer Neff, J. Fetzer multidimensional measure. Journal for the Scientific Study of Religion, 45 (3), 449 459. doi: 10.1111/j.1468 5906.2006.00318.x. Newlin, K., Dyess, S. M., Allard, E., C hase, S., & Melkus, G. D. (2011). A methodological review of faith based health promotion literature: Advancing the science to expand delivery of diabetes education to African Americans. Journal of Religion and Health Advance online publication. doi: 10.1 007/s10943 011 9481 9 Niederdeppe, J., & Levy, A. G. (2007). Fatalistic beliefs about cancer prevention and three prevention behaviors. Cancer Epidemiology, Biomarkers & Prevention, 16 998 1003. doi: 10.1158/1055 9965.EPI 06 0608 Obisesan, T., Livingston, I., Trulear, H. D., & Gillum, F. ( 2006 ). Frequency of attendance at religious services, cardiovascular disease, metabolic risk factors and dietary intake in Americans: An age stratified exploratory analysis. International Journal of Psychiatry Medicine, 36 (4), 435 448. Ogden, C. L., & Carroll, M. D. (2010, June). Prevalence of overweight, obesity, and extreme obesity among adults: United States, trends 1960 1962 through 2007 2008. Hyattsville, MD: National Center for Health Statistics. Retrieve d from http://www.cdc.gov/NCHS/data/hestat/obesity_adult_07_08/obesity_adult_07_ 08.pdf Ogden, C. L., Carroll, M. D., Curtin, L. R., McDowell, M. A., Tabak, C. J., & Fl egal, K. M. (2006). Prevalence of overweight and obesity in the United States, 1999 2004. Journal of the American Medical Association, 295 (13) 1549 1555. doi: 10.1001/ jama.295.13.1549
136 M. A., & Brantley, P. J. (2005). Predicting medical regimen adherence: The interactions of health locus of control beliefs. Journal of Health Psychology, 10, 705 71 7. doi:10.1177/ 1359105305055330 Oxman, T. E., Freeman, D. H. Jr., & Manheimer, E. D. (1995) Lack of social participation or religious strength and comfort as risk factors for death after cardiac surgery in the elderly. Psychosomatic Medicine, 57 (1), 5 15. Park, C. L. (2007). Religiousness/spirituality and health: A meaning systems perspective. Journal of Behavioral Medicine, 30, 319 328. doi: 10.1007/s10865 007 9111 x Pew Forum on Religion & Public Life. (2008). U.S. Religious Landscape Survey. Pew Research Center: Washington, D.C. Retrieved from the Pew Forum on Religion & Public Life website: http://religions.pewforum.org/reports# Pleis, J. R., & Lucas, J. W. (2009). Summary health statistics for U.S. adults: National Health Interview Survey, 2007. Vital Health Statistics, 10 (240). Retri eved from Centers for Disease Control and Prevention, National Center for Health Statistics website: http://www.cdc.gov/nchs/data/series/sr_10/sr10_240.pdf Polzer, R., & Miles, M. S. ( 2005). Spirituality and self management of diabetes in African Americans. Journal of Holistic Nursing, 23 (2), 230 250. doi:10.1177/ 0898010105276179 Powell, L. H., Shahabi, L., & Thoresen, C. E. (2003). Religion and spirituality: Linkages to physical healt h. American Psychologist, 58 (1), 36 52. doi:10.1037/0003 066X.58.1.36 Prostate Cancer Foundation. (2011). Understanding prostate cancer prevention. Retrieved from http:// www.pcf.org/site/c.leJRIROrEpH/b.5802029/k.31EA/ Prevention.htm Raque Bogdan, T. L., Torrey, C. L., Lewis, B. L., & Borges, N. J. (2012). Counseling health psychology: Assessing health psychology training within counseling psychology doctoral programs. Th e Counseling Psychologist. Advance online publication. doi: 10.1177/0011000012439611 Reindl Benjamins, M. & Brown, C. (2004). Religion and preventative health care utilization among the elderly. Social Science and Medicine, 58, 109 118.s. doi: 10.1016/S0277 9536(03)00152 7 Reindl Benjamins, M. Ellison, C. G., Krause, N. M., & Marcum, J. P. (2011). Religion and preventive service use: Do congregational support and religious beliefs explain th e relationship between attendance and utilization? Journal of Behavioral Medicine, 34, 462 476. doi: 10.1007/s10865 011 9318 8
137 Resnicow, K., Kramish Campbell, M., Carr, C., McCarty, F., Wang, T., Periasamy, ntervention conducted through African American churches. American Journal of Preventive Medicine, 27 (2), 97 105. doi: 10.1016/j.amepre.2004.04.009 Rippentrop, A. E. (2005). A review of the role of religion and spirituality in chronic pain populations. Reha bilitation Psychology, 50, 278 284. doi: 10.1037/0090 5522.214.171.1248 Rippentrop, A. E., Altmaier, E. M., & Burns, C. P. (2006). The relationship of religiosity and spirituality to quality of life among cancer patients. Journal of Clinical Psychology in Medical Settings, 13 (1), 29 35. doi: 10.1007/s10880 005 9000 9 Roger, V. L., Go, A. S., Lloyd Jones, D. M., Adams, R. J., Berry, J. D., Brown, T. Rosett, J. (2011). Heart disease and stroke s tatistics 2011 update: A report from the American Heart Association. Circulation, 123 (e18 e209). doi: 10.1161/CIR.0b013e3182009701 Schlundt, D. G., Hargreaves, M. K., & Buchowski, M. S. (2003). The Eating Behavior Patterns Questionnaire predicts dietary f at intake in African American women. Journal of the American Dietetic Association, 103 (3), 338 345. doi: 10.1053/jada.2003.50049 Seybold, K. S. (2007). Physiological mechanisms involved in religiosity/spirituality and health. Journal of Behavioral Medicin e, 30, 303 309. doi: 10.1007/s10865 007 9115 6 Sink, C. A., & Stroh, H. R. (2006). The use of effect sizes in school counseling research. Professional School Counseling, 9 (5), 401 411. Sloan R. P., & Bagiella E. (2002) Claims about religious involvemen t and health outcomes. Annals of Behavioral Medicine, 24 (1), 14 21. doi: 10.1207/S15324796A BM2401_03 Strawbridge, W. J., Shema, S. J., Cohen, R. D., & Kaplan, G. A. (2001). Religious attendance increases survival by improving and maintaining good health b ehaviors, mental health, and social relationships. Annals of Behavioral Medicine, 23 (1), 68 74. doi: 10.1207/S15324796ABM2301_10 Trinitapoli, J., Ellison, C. G., & Boardman, J. D. (2009). U.S. religious congregations and the sponsorship of health related p rograms. Social Science and Medicine, 68, 2231 2239. doi: 10.1016/j.socscimed.2009.03.036 www.healthyamericans.org
138 Tucker, C. M., Ferdinand, L. A., Mirsu Paun, A., Herman, K. C., Delgado Romero, E., van den Berg, J. J., & Jones, J. D. (2007). The roles of counseling psychologists in reducing health dispa rities. The Counseling Psychologist, 35 650 676. doi: 10.1177/0011000007301687 Underwood S. M., & Powell, R. L. ( 2006 ). Religion and spirituality: Influence on health/risk behavior and cancer screening behavior of African Americans. The Official Journal of the Association of Black Nursing Faculty, 17 (1), 20 31. U.S. Department of Health and Human Services. (2011, May 2). 2020 topics and objectives: Nutrition and weight status. Retrieved from http://www.healthypeople. gov/2020/topicsobjectives2020/overview.aspx?topicid=29 U.S. Department of Health and Human Services, National Institutes of Health, National Canc er Institute. (2011a). Cancer prevention overview: Risk factors. Retrieved from http://www.cancer.gov/cancertopics/pdq/prevention/overview/patient/page3 U.S. Departm ent of Health and Human Services, National Institutes of Health, National Cancer Institute. (2011b). Prostate cancer prevention. Retrieved from http://www .cancer.gov/cancertopics/pdq/prevention/prostate/Patient/page3#Keypoi nt16 U.S. Department of Health and Human Services, National Institutes of Health, National Heart, Lung, and Blood Institute. (n.d.) What is heart disease? Retrieved from http://www.nhlbi.nih.gov/educational/hearttruth/lower risk/what is heart disease.htm U.S. Department of Health and Human Services, National Institutes of Health, Nation al Heart, Lung, and Blood Institute. (2011, Feb 1). What is a stroke? Retrieved from http://www.nhlbi.nih.gov/health/health topics/topics/stroke/ U.S. Department of Health and Hu man Services, National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, National Diabetes Information Clearinghouse. (2008). Diabetes overview (NIH Publication No. 09 3873). Retrieved from http://diabetes.niddk.nih.gov/dm/pubs/overview/#scope U.S. Department of Health and Human Services, National Prevention, Health Promotion, and Public Health Council. (2010, July 1). 2010 annual status report. Re trieved from http://www.hhs.gov/news/reports/nationalprevention2010report.pdf U.S. Department of Health and Human Services, Office of Minority Health. (2011). African A merica n profile. Retrieved from http://minorityhealth.hhs.gov/templates/browse. aspx?lvl=2&lvlid=51 Walker, S. N., & Hill Polerecky, D. M. (1996). Psychometric evaluation of t he Health Promoting Lifestyle Profile II. Unpublished manuscript, University of Nebraska Medical Center. Retrieved from http://www.unmc.edu/nursing/docs/HPLPII_ Abstract_ Dimensions.pdf
139 Walker, S. N, Sechrist, K. R., & Pender, N. J. (1987). The Health Promoting Lifestyle Profile: Development and psychometric characteristics. Nursing Research, 36 (2) 76 81. doi: 10.1097/00006199 198703000 00002 Wallston, K. A. (2005). The validity of the Multidimensional Health Locus of Control scales. Journal of Health Psychology, 10 (4), 623 631. doi: 10.1177/1359105305055304 Wallston, K. A., Malcarne, V. L., Flores, L., Hansdottir, I., Smith, C. A., Stein, M. Does God determine your health? The God Locus of Health Control scale. Cognitive Therapy and Research 23 131 142. Retrieved from http://www.vanderbilt.edu/nursing/kwallston/A71.pdf Weav er, A., Pargament, K., Flannelly, K., & Oppenheimer, J. (2006). Trends in the scientific study of religion, spirituality, and health: 1965 2000. Journal of Religion and Health, 45 (2), 208 214. doi: 10.1007/s10943 006 9011 3 Welton, G. L., Adkins, A. G., In gle, S. L., & Dixon, W. A. (1996). God control: The fourth dimension. Journal of Psychology and Theology, 24 (1), 13 25. Retrieved from http://apps.biola.edu/jpt/volumes/24/issues/1/a rticles/13 Williams, D. R., & Sternthal, M. J. (2007). Spirituality, religion and health: Evidence and research directions. Spirituality and Health, 186 (10), S47 S50. Retrieved from https://www.mja.com.au/public/issues/186_10_210507/wil11060_fm.pdf Wilcox, S., Laken, M., Parrott, A. W., Condrasky, M., Saunders, R., Addy, C. L....Samuel, M. (2010). The Faith, Activity, and Nutrition (FAN) Program: Design of a participatory research intervention to increase physical activity and improve dietary habits in African American churches. Contemporary Clinical Trials, 31 (4), 323 335. doi: 10.1016/j.cct.2010.03.011 Wolin, K. Y., Carson, K., & Colditz, G. A. (2010). Obesity and cancer. The Oncologist, 15 (6), 556 565. doi: 10.1634/theoncologist.2009 0285 World Health Organization. (2008). Closing the Gap in a Generation: Health Equity Through Action on the Social Determinants of Health. Final Report of the Commission of Social Determinan ts of Health. Geneva, Switzerland: World Health Organization. Retrieved from http://www.who.int/social_determinants/ thecommission/finalreport/en/index.html Xu J. Q., Kochanek, K. D., Murphy, S. L., & Tejada Vera, B. (2010). Deaths: Final data for 2007. National Vital Statistics Reports, 58 (19). Hyattsville, MD: National Center for Health Statistics. Retrieved from http://www.cdc.gov/nchs/data/nvsr/nvsr58/ nvsr58_19.pdf
140 BIOGRAPHICAL SKETCH Lillian (Lily) Boynton Kaye was born and raised in Tallahassee, Florida, where she co owns a business Railroad Square, LLC She graduated summa cum laude from the University of Florida with a Bachelor of Science in psychology and a Bachelor of Arts in linguistics and alcohol/drug use among college students. A year later, s he entered t he Counseling Psychology Ph.D. P rogram in the University of Florida Department of Psychology under the mentorship of Dr. Carolyn Tuc ker. In 2008, Lily completed a m on the motivators of and barriers to healthy eating behaviors among low income children and r eceived a Master of Science in p sychology. Throughout graduate school, Lily held leadership positions focused on the funded community based health promotion and intervention programs targeting low income and racial/ethnic minority families and individuals. Lily co led the production of The Family Health Self Empowerment DVD and the associated participant resource guide. She also trained community leaders in Gaine sville, FL, Lexington, KY, and the Bronx, NY to implement these interventions. Clinically, Lily completed practicums in a number of settings throughout her doctoral training, including the North Florida/ South Georgia Veterans Hospital and the Behavioral Health Unit and Anxiety Disorders/OCD Clinic at Shands Hospital. She completed her predoctoral internship at the University of Florida Counseling and Wellness Center. In 2011, she received the Ted Landsman Award in Counseling Psychology from her home dep artment. In summer 2012, Lily defended her dissertation and graduated with a Ph.D. in counseling psychology.