BREAKFAST HABITS AND THEIR RELATION TO DIET QUALITY AND HEALTH-RELATED QUALITY OF LIFE IN AN URBAN, SOCIOECONOMICALLY DIVERSE SAMPLE

BREAKFAST HABITS AND THEIR RELATION TO DIET QUALITY AND HEALTH-RELATED QUALITY OF LIFE IN AN URBAN, SOCIOECONOMICALLY DIVERSE SAMPLE OF AFRICAN AMERICAN AND WHITE ADULTS by Megan Grimes A thesis submitted to the Faculty of the University of Delaware in partial fulfillment of the requirements for the degree of Master of Science in Human Nutrition Spring, 2017 2017 Megan Grimes All Rights Reserved

BREAKFAST HABITS AND THEIR RELATION TO DIET QUALITY AND HEALTH-RELATED QUALITY OF LIFE IN AN URBAN, SOCIOECONOMICALLY DIVERSE SAMPLE OF AFRICAN AMERICAN AND WHITE ADULTS by Megan Grimes Approved: Marie F. Kuczmarski, Ph.D. Professor in charge of thesis on behalf of the Advisory Committee Approved: Michael Peterson, Ed.D. Chair of the Department of Behavioral Health and Nutrition Approved: Kathleen S. Matt, Ph.D. Dean of the College of Health Sciences Approved: Ann L. Ardis, Ph.D. Senior Vice Provost for Graduate and Professional Education

ACKNOWLEDGMENTS I would like to take this opportunity to thank the HANDLS dietary research team. Without the help of Emily Shupe, Samantha Reilly, and especially Dr. Marie Kuczmarski, writing this thesis would not have been possible. Thank you to Dr. Kuczmarski for offering me so many opportunities throughout my undergraduate and graduate education. I would also like to acknowledge Dr. Barry Bodt for his countless hours teaching me statistics. Thank you for your patience! To Dr. Sandra Baker and Dr. Richard Fang, thank you for your guidance and support throughout the years. Finally, I would like to thank my family and Austin for their encouragement throughout this process. iii

TABLE OF CONTENTS LIST OF TABLES... v LIST OF FIGURES... vi ABSTRACT... vii Chapter 1 DEFINITIONS... 1 2 REVIEW OF THE LITERATURE... 5 3 PURPOSE OF THE STUDY... 9 4 METHODS... 10 5 RESULTS... 15 6 DISCUSSION... 19 7 CONCLUSION... 25 REFERENCES... 33 Appendix A INFORMED CONSENT... 46 B HEALTHY EATING INDEX-2010 COMPONENTS AND STANDARDS FOR SCORING... 55 C SHORT FORM 12 HEALTH SURVEY AS USED IN THE HANDLS STUDY... 57 D REGRESSION MODELING FOR HEI-2010 SCORES... 60 E REGRESSION MODELING FOR SF-12 MENTAL COMPONENT SUMMARY SCORES... 65 F REGRESSION MODELING FOR SF-12 PHYSICAL COMPONENT SUMMARY SCORES... 70 G DEMOGRAPHIC FACTORS FOR BREAKFAST CONSUMERS AND SKIPPERS... 76 H FURTHER ANALYSIS OF EDUCATION AND BREAKFAST GROUP CHI-SQUARE... 77 I POST-HOC ANALYSIS OF DEMOGRAPHIC DIFFERENCES AMONG HOME BREAKFAST CONSUMERS, AFH BREAKFAST CONSUMERS, AND BREAKFAST SKIPPERS... 80 iv

Table 1 Table 2 Table 3 Table 4 Table 5 Table 6 Table 7 Table 8 Table 9 LIST OF TABLES Number and percent of participants of the Healthy Aging in Neighborhoods of Diversity across the Life Span study who report eating breakfast at home, eating breakfast away from home (AFH), or skipping breakfast on each day of recall... 26 Comparison of demographics for Healthy Aging in Neighborhoods of Diversity across the Life Span study participants by breakfast group... 26 Comparison of nutrient profiles of breakfasts eaten at home or away from home (AFH) by participants in the Healthy Aging in Neighborhoods of Diversity across the Life Span study... 27 Comparison of nutrient profiles per kilocalorie of breakfasts eaten at home or away from home (AFH) by participants in the Healthy Aging in Neighborhoods of Diversity across the Life Span study... 27 Comparison of Healthy Eating Index-2010 total and component scores of Healthy Aging in Neighborhoods of Diversity across the Life Span participants by breakfast group... 28 Factors influencing the Healthy Aging in Neighborhoods of Diversity across the Life Span study participants Healthy Eating Index-2010 scores by multiple regression... 29 Comparison of Short Form 12** summary and health concept scores of the Healthy Aging in Neighborhoods of Diversity across the Life Span study sample by breakfast group... 30 Factors influencing the Healthy Aging in Neighborhoods of Diversity across the Life Span study participants Mental Component Summary scores by multiple regression... 31 Factors influencing the Healthy Aging in Neighborhoods of Diversity across the Life Span study participants Physical Component Summary scores by multiple regression... 31 v

LIST OF FIGURES Figure 1 HANDLS study participant flowchart... 32 vi

ABSTRACT Background Observational studies have documented decreased diet quality and health-related quality of life with breakfast skipping. These outcomes may be connected to the increased risk of obesity and chronic disease compared to breakfast consumers. Away from home (AFH) foods have also been correlated with lower diet quality; however, the relationship between away from home foods and health-related quality of life is unknown. AFH breakfasts have a larger impact on diet quality than other meals; however, to our knowledge no one has compared the effects of eating away versus not eating breakfast at all or has assessed the relationship between healthrelated quality of life and breakfast location. Objective To characterize the breakfast habits of a diverse, urban population and to compare diet quality and health-related quality of life when breakfasts are eaten at home, eaten away from home, or skipped. Subjects Participants of the prospective cohort Healthy Aging in Neighborhoods of Diversity across the Life Span (HANDLS) study. Participants were African American and White adults residing in Baltimore, MD. Methods Dietary data (two 24-hour recalls) from Wave 3 of the HANDLS study were used to divide participants into three groups: home breakfast consumers, AFH breakfast consumers, and breakfast skippers. Those who reported different breakfast habits across the two days were labeled breakfast switchers and were not included. The Healthy Eating Index (HEI)-2010 was used as a diet quality measure. The Short Form 12 (SF-12) Health Survey was the measure of health-related quality of life. This survey gives two composite scores: the Mental Component Summary (MCS) score and the Physical Component Summary (PCS) score. vii

Statistical Analysis Only subjects who completed two 24-hour recalls and reported either no breakfast or consumption of breakfast at home or AFH were eligible (n=2115). Descriptive statistics were calculated for demographic and dietary data. After removing breakfast switchers, 1834 cases remained. One-way ANOVA and Chi-square tests were used to compare dietary and demographic data among the three groups. Independent samples T-tests were used to compare breakfast nutrient intake among home and AFH breakfast consumers. Three regression models were run to examine the relationship between the outcome measures (HEI-2010, MCS score, and PCS score) and breakfast group after controlling for covariates (age, sex, race, poverty status, BMI, marital status, employment, education, smoking, literacy, and income). Results 2115 subjects were included in analysis of breakfast characteristics; 60.8% were breakfast consumers. When divided into breakfast groups, 41.2% of subjects ate breakfast at home both days, 6.3% of subjects ate both breakfasts AFH, and 13.3% were breakfast switchers. For the HEI analysis, 1834 subjects were included. The home breakfast consumers had the highest HEI-2010 scores, whereas eating breakfast AFH or skipping breakfast was associated with a decrease of 3.59 and 4.66 HEI points, respectively. After adjustment, neither mental nor physical HRQoL scores were associated with breakfast group. Conclusion When counseling clients, dietitians should consider breakfast habits and recommend eating breakfast at home; however, if clients cannot eat at home, eating breakfast away from home was related to higher diet quality than skipping breakfast. In meal pattern research, location of meals should be considered. HRQoL, as a long-term measure of health, did not appear to be related to daily breakfast habits. viii

Chapter 1 DEFINITIONS Breakfast Breakfast is often understood as the first meal of the day; however, there is no consistent definition used by researchers. 1 A proposed definition for breakfast is the first meal of the day that breaks the fast after the longest period of sleep and is consumed within 2 to 3 hours of waking; it is comprised of food or beverage from at least one food group, and may be consumed at any location. 2 In research practice, many papers use either the participant-identified approach, which defines breakfast according to the subject s naming of the meal, or the time-frame approach, which includes any food eaten within a certain time frame (often 6am-10am) as breakfast. 3 When the two definitions were applied to whole day intake, there was no clear advantage of either definition, 3 and both are used in the literature. The inconsistent breakfast definition has, so far, limited conclusions about its benefits, 1 3 as has the large variety in breakfast foods. The most recent proposed definition of a quality breakfast includes (1) choices from at least three food groups, (2) 15-25% of recommended total calories, (3) meets at least 10% of the Daily Value for as many nutrients as possible, (4) considers the amounts of sodium and saturated fats, (5) portion sizes appropriate for age, sex, and daily energy requirements, and (6) is based on nutrient dense foods. 2 For this analysis, the participant-identified definition of breakfast was applied. 1

Although the importance of breakfast is well-known, approximately 20% of people report skipping breakfast. 4 6 African American 5,7,8 and low-income populations 8 are more likely to skip breakfast than other groups. The prevalence of breakfast skipping is so high, the 2010 Dietary Guidelines for Americans included eating a nutritious breakfast as a goal. 9 Away From Home Foods Away from home (AFH) foods can be described as foods that are eaten or obtained away from the home. 10 When defining AFH foods, some researchers use the where obtained specification, which counts AFH food as food obtained from places outside the household, regardless of where eaten. Although a common source is fastfood, AFH foods can come from full-service restaurants, school or worksite cafeterias, and many other places. Using this definition, fast food that is brought home to eat is still considered AFH food. Other studies use the where eaten definition, which counts AFH foods as food eaten away from home, regardless of where obtained. Under this definition, food prepared and packed at home but eaten away would be considered AFH. As of now, there is no consensus on which definition is correct. 10 For this analysis, the where eaten definition of AFH foods was applied. AFH foods are common among US adults, especially African Americans 11,12 and those with low education 11 or low nutrition knowledge. 12 In 2007-2008, 36% and 27% of people reported consumption of fast food and full-service restaurant food on their National Health and Nutrition Examination Survey (NHANES) recall, respectively. 2 Another study estimated 3.9 AFH meals per week. 13 2

Diet Quality Despite the lack of a standard definition for diet quality, it is generally regarded as a whole-diet approach to measuring intake; it often compares true intake to a pre-defined diet standard. Various measures of diet quality have been linked to important health outcomes. 14 The Healthy Eating Index -2010 (HEI-2010) is a valid and reliable measure of diet quality, which measures adherence to the Dietary Guidelines for Americans 2010. 15 This measure has been inversely associated with cardiovascular, cancer, and all-cause mortality 16,17 and type 2 diabetes mellitus. 18 Using HEI-2005 (a measure based on previous US dietary guidelines), HEI scores were inversely related to body mass index (BMI), waist circumference, diastolic blood pressure, C-reactive protein, total cholesterol, LDL cholesterol and metabolic syndrome. 19 The HEI-2010 has also been used to measure national progress towards federal goals. Using NHANES data from 1999-2000 and 2011-2012, Wilson et al calculated a national mean HEI score of 49 (out of 100) in 1999-2000, which increased to a 59 in 2011-2012. 20 Although the average total HEI score increased, the trajectory of the increase suggests that the nation will not reach the Healthy People 2020 goals for diet. Furthermore, although most of the individual components improved over time, the sodium component moved further from goals. 20 This finding suggests that national improvements in diet quality are required. Diets of low socioeconomic, diverse populations are often low quality, 21 and may be targeted for improvement. Given the negative health effects associated with low quality diets, studies on factors that influence diet quality are needed. Diet quality can be influenced by eating patterns, such as meal skipping and eating away from home. 22 Meal skipping has been 3

shown to decrease quality; most of the literature focuses on breakfast skipping in particular. 1 Diet quality may also change with meal location at home or away. 10 Health-Related Quality of Life Health-related quality of life (HRQoL) is defined as those aspects of overall quality of life that are shown to affect health, such as physical and mental health perceptions. 23 Health-related quality of life can be measured with several tools, including the Short Form 12 Health Survey (SF-12). 24 The SF-12 is reported as two scores: the Physical Component Summary (PCS) score and the Mental Component Summary (MCS) score. Generally, as people get older, the PCS score decreases and the MCS score increases. 25 Those with higher BMI often have lower PCS scores, but not MCS scores. 26 Recently, measures of HRQoL have gained popularity as predictors of mortality and morbidity in older adults. 27 29 Low HRQoL scores are associated with chronic diseases such as diabetes, 23,30 metabolic syndrome, 30 breast cancer, arthritis, and hypertension, 23 and with risk factors such as smoking, physical inactivity, 23,31 inflammation 31 and BMI. 23 4

Chapter 2 REVIEW OF THE LITERATURE Breakfast and Diet Quality The variability in the types of breakfasts has limited general conclusions about diet quality and breakfast consumption. Using NHANES data from 2001-2008, 11 breakfast consumption patterns were observed, including Grain/100% Fruit Juice and Coffee/Teas. Seven of these breakfast consumption patterns (including Grain/100% Fruit Juice ) had better diet quality, but the rest (including Coffee/Teas ) had equal or lower diet quality scores than breakfast skippers. 4 Often, ready-to-eat (RTE) breakfast cereals are studied as optimal breakfasts. These cereals are associated with micronutrient and whole grain intake in older Americans. 32 Using NHANES 1992-2002 data, it was found that RTE cereal consumers had higher intakes of fiber, total fruits, dairy products, and a higher overall diet quality score than other breakfast consumers. 5 Oatmeal has also been studied, and oatmeal consumers were shown to have higher diet quality scores; however, this study included oatmeal intake throughout the day, not just at breakfast. 33 Despite these differences, observational studies suggest that breakfast is related to diet quality. 1,5,7 Breakfast skippers often have lower quality diets which include more added sugars, 4,5,34 fewer shortfall micronutrients, 5 and fewer servings of fruit and whole grains. 8 Breakfast skippers are less likely to meet the Recommended Dietary Allowances (RDAs) for nutrients, 35 and the low nutrient intake is independent of total number of eating occasions, meaning that the lack of nutrients cannot be made up for by snacks. 36 This change in diet quality among breakfast skippers may be related to chronic stress; higher levels of stress were related to a higher empty calorie 5

and added sugars score on the HEI for breakfast skippers but not breakfast consumers. 34 AFH Foods and Diet Quality Studies find that AFH foods are related to lower diet quality. 10,37,38 They are associated with increased daily intake of total fat, 10,11 total energy, and sodium. 10,11,39 It is estimated that spending $10 more per week on AFH food can lead to an increase of 1.4g fat and a decrease of 0.2g fiber per day, 12 and each AFH meal adds 134 kcals and lowers HEI-2005 scores by 2 points. 37 Fast food consumption in particular increases sugar intake and decreases fiber intake. 11 In fact, when the HEI-2005 scoring system was applied to five popular fast food chain menus, not one had a total score of 50 or above. 40 Dark green/orange vegetables received some of the lowest component scores, 40 but adding vegetables to the menu may not be the answer. One additional cup of AFH tomatoes contributes an average 363 kcal, mostly because tomatoes are often added as sauces for energy-dense foods. 41 AFH tomatoes also have more sodium than these vegetables prepared at home. 41 The effect of restaurant food on energy may also vary by restaurant type: Thai, Greek, Japanese, Vietnamese, and Mexican foods were shown to have less energy per entrée, whereas American, Chinese, and Italian entrees had the most energy. 42 Despite these differences, the average energy per entrée was 1,205 kcal. 42 Nutrition standards for these menus do not exist, although recommendations were recently published for AFH food standards. 43 Menu labeling of nutrition information was also shown to decrease AFH energy consumption in a recent meta-analysis. 44 People often eat larger portions of AFH foods than when eating at home; in fact, when restaurant food was eaten AFH, people ate larger portions 45 and more 6

cholesterol and sugar 11 than if they brought it home to eat. Unfortunately, this effect is larger in overweight and obese subjects than those of a healthy weight. 46 So far, studies do not suggest that people compensate for eating AFH by eating healthier the rest of the day. 38 Breakfast and HRQoL Breakfast skipping has been related to lower HRQoL, especially mental (rather than physical) quality. 47 This relationship may be due to the fact that breakfast skipping is often clustered with many unhealthy behaviors, such as smoking, alcohol use, and physical inactivity. 8,48 Skipping breakfast has been hypothesized to directly decrease physical activity in the morning, especially among obese. 49 Other characteristics of breakfast skippers include lower levels of education, male sex, and higher intake of fast food. 7 Also, there is an association between breakfast skipping and greater risk for chronic diseases including hypertension, metabolic syndrome, 7 type 2 diabetes mellitus, 7,18 and obesity. 7,50 These diseases could possibly decrease health-related quality of life. AFH Foods and HRQoL The lower nutritional quality of AFH foods can lead to negative health outcomes, which could decrease health-related quality of life. Consumption of AFH foods is positively associated with BMI 51 and negatively associated with HDL cholesterol, vitamin E, vitamin B12, and folate levels. 13 Increased AFH food consumption is also associated with greater risk for vitamin D deficiency. 13 AFH food consumption has also been linked to other health-compromising characteristics such as breakfast skipping, 7,8 low education 11 and low nutrition knowledge. 12 Although the 7

association between AFH foods and HRQoL has not been studied, it is likely that AFH food consumption would be related to lower health-related quality of life. Diet Quality and HRQoL Increased HRQoL is positively associated with compliance to nutrition guidelines developed in France 52 and Australia. 53 The Mediterranean diet and its relationship with HRQoL has also been studied in several populations. There was no association between adherence to a Mediterranean diet and HRQoL in a study of Spanish older adults, 54 although another Spanish study found a direct association. 55 In a study of Italian adults, the Mediterranean diet score was associated with increased PCS and MCS scores, but this was not significant after controlling for total antioxidant content and fiber. 56 So far, to the author s knowledge there are no studies of the association between American dietary guidelines (namely the HEI-2010) and HRQoL. AFH Breakfasts, Diet Quality, and HRQoL Prevalence of AFH breakfast consumption varies by study population, from 8% 37 to 25%. 35 Although less common than other meals, AFH breakfasts have larger negative effects on diet quality than other AFH eating occasions. 37,38 An AFH breakfast was estimated to decrease HEI score by 4.5 points. 37 AFH breakfasts are significantly higher in energy, percent total fat, percent saturated fat, and lower in fiber, and are associated with increased risk of obesity. 57 A fast-food style breakfast was also found to decrease HDL-C and increase triglycerides and some markers of oxidative stress. 58 Considering the unhealthful attributes and outcomes associated with AFH breakfasts, it is likely that the HRQoL would also be decreased. 8

Chapter 3 PURPOSE OF THE STUDY Although both skipping breakfast and AFH breakfast are associated with decreased diet quality, to the knowledge of the author, no one has compared these effects. Furthermore, although breakfast skipping is related to lower health-related quality of life, the effect of AFH breakfasts on HRQoL appears unknown. The purpose of this study is to characterize the breakfast habits of a racially and socioeconomically diverse, urban population and to compare diet quality and health-related quality of life when breakfasts are eaten at home, eaten away from home, or skipped. By knowing the breakfast habits of a diverse urban population, recommendations for improvement can be made. By comparing diet quality and health-related quality of life, identification of the most healthful breakfast choices can be achieved. Research Question 1: What are the breakfast habits of the HANDLS study population? Research Question 2: How do the three habits breakfast at home, breakfast away from home, or no breakfast relate to diet quality? Research Question 3: How do the three habits relate to health-related quality of life (HRQoL)? 9

Chapter 4 METHODS HANDLS Study The Healthy Aging in Neighborhoods of Diversity across the Life Span (HANDLS) study is a prospective longitudinal study of African American and White adults who live in Baltimore, Maryland. The study sample includes participants from 13 neighborhoods using a factorial cross of age, sex, race, and socioeconomic status (SES). At baseline the age of participants was between 30 and 64 years old. This study was designed to examine the influence of race and SES on selected health outcomes in an aging cohort. 59 Several waves of data have been collected; for this analysis, Wave 3 data will be used (see consent forms in Appendix A). Wave 3 data were collected between June 2009 and July 2013. During data collection, participants came to a Mobile Research Vehicle (MRV), where several measures of health were collected, including a 24-hour dietary recall, medical history, physical examination, and cognitive tests. Between 4-10 days after their MRV visit, participants performed another 24-hour dietary recall over the phone. 60 Dietary Recalls The United States Department of Agriculture s (USDA) Automated Multiple- Pass Method (AMPM) was used to complete the dietary recalls. 61 Trained interviewers used this computer-assisted interview technique which includes five steps (quick lists, forgotten food, time and occasion, detailed description, and final probe) to collect all of the foods and beverages consumed within the previous 24-hour day, from midnight to midnight. Food models and booklets were used by participants to estimate portion 10

sizes. During the recall, participants were asked to name each eating occasion (as breakfast, lunch, dinner, brunch, supper, snack, drink, or extended consumption) and where it was eaten (home or away). 61 In the AMPM, a meal s location is based on where the first bite was eaten. For example, if a subject ate part of breakfast at home and then the rest in the car, the meal would be considered eaten at home. For this analysis, the location where meals were eaten was used to define foods as away from home or not, and participant naming of the meal was used to define breakfast. Each participant was asked to complete two dietary recalls. After interviews were collected, trained coders used SurveyNet to match foods to their codes from the Food and Nutrient Database for Dietary Studies (FNDDS, Version 5.0). 62 From these codes, energy, macronutrients, and micronutrient intakes were calculated. Diet Quality Measures The Healthy Eating Index (HEI)-2010 was used to assess diet quality. The HEI-2010 is a valid and reliable measure of diet quality, and compares intake to federal diet standards, namely the Dietary Guidelines for Americans. Foods are first divided into 12 components (nine adequacy and three moderation components) (see Appendix B). For the adequacy components, a higher score is given for a higher intake; however, in components where moderation is key (such as sodium), a higher score is given for a lower intake. Components vary in their maximum scores from a possible five points to 20 points; for all components, energy is accounted for by calculating intake per 1,000 kcals. The scores from each category are added up to give a total HEI score, which ranges from 0 to 100 (100 being the highest quality diet). 15 HEI scores were calculated from the 24-hour recall data for both days and then 11

averaged. For more information on HEI-score calculations, see the HANDLS website (http://handls.nih.gov/06coll-w01hei.htm). Health-Related Quality of Life Measures The Short Form 12 Health Survey (SF-12) was used to measure health-related quality of life (HRQoL) (see Appendix C). The SF-12 measure is a briefer version of the Short Form 36 Health Survey (SF-36), which was developed from the Medical Outcomes Study. 63 The original survey has 36 questions and, although reliable and valid, its length limits its use in large research studies. The SF-12 was created based on the physical and mental components of the SF-36, which were shown to explain 90% of the variance in SF-36. 63 The validity and reliability of the SF-12 have been shown to be acceptable in the general population, 63 in those with obesity, 26 in older adults, 64,65 and in different disease groups. 24 Although more efficient, the 12-question survey is not as valid and reliable as the SF-36, but this limitation is less important in studies with larger sample sizes. 63 HANDLS participants completed the survey during their MRV visit. Survey items asked how often in the past four weeks the subject had difficulty performing certain tasks due to mental and physical health. Then, the answers were used, following the SF-12 scoring guidelines, to calculate two scores: the Physical Component Summary (PCS) score and Mental Component Summary (MCS) score. Each score ranges from 0-100 (100 being the best quality), and scoring is defined such that a 50 is equal to the average score for Americans, with a standard deviation of ±10 points. 66 This scoring system allows results to be comparable between study groups. The SF-12 can also be used to calculate eight health concept scores. These scores are derived from one or two questions each and cover the following health concepts: 12

physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. 66 Statistical Analysis Analysis included only participants who completed both 24-hour dietary recalls (n=2140). Some participants reported having two or more breakfasts on a single recall. If all of these breakfasts were in the same location, they were included; however, if a participant had breakfasts in more than one location on a single recall, they were excluded from analysis. Some participants did not report where they had breakfast and were excluded as well. A total of 25 participants were removed, resulting in a final study sample of 2115 (Figure 1). Descriptive analysis was performed to determine demographics and breakfast habits of the entire sample. Then, subjects were divided into home breakfast consumers, AFH breakfast consumers, and breakfast skippers. If a subject switched breakfast groups between the two recalls (ie, ate breakfast at home one day and then either skipped or ate AFH the other day; breakfast switchers ), they were excluded from further analysis. Descriptive statistics were calculated for demographic and dietary data for the three groups. Independent samples t-tests were used to compare breakfast nutrients between home breakfast consumers and AFH breakfast consumers. Chi-square and one-way ANOVA tests with Tukey HSD post hoc analysis were used to compare demographics, HEI component scores, and SF-12 health concept scores among the three breakfast groups. Regression models were used to compare the three breakfast groups with HEI total score and MCS and PCS scores after adjusting for potential confounders (see 13

Appendices D-F). Age, sex, race, poverty status, a race by poverty status interaction, BMI, marital status, employment, education, smoking, literacy, and income were considered. All demographic variables were collected during the MRV visit. Literacy was assessed using the Wide Range Achievement Test 3 (WRAT-3). Income category was collected; to run in the regression, the variable was made continuous by assigning each subject a numerical value (the midpoint of each category). Any variable that was not significant in a particular model was removed from that model, with the exception of race, poverty status, and their interaction, which were retained because of the overall HANDLS study aims. All model assumptions, including normality and linearity, were evaluated. Outliers were screened for; only one which was judged to be impossible was removed (a BMI of 760). A p-value 0.05 was considered significant. 14

Chapter 5 RESULTS Sample Population and Breakfast Habits In Wave 3, 2115 subjects completed two days of dietary recall with known breakfast locations and only one location per recall day. Of this sample, 58.8% were female, 38.3% were White, and 60% were above 125% of the 2004 federal poverty guidelines. The mean age (±SE) was 53.14 ± 0.19 years and the mean BMI (±SE) was 30.71 ± 0.17. All subjects who consumed breakfast on day 1 also consumed breakfast on day 2. Overall, 60.8% were breakfast consumers and 39.2% were breakfast skippers. When comparing breakfast skippers with breakfast consumers (either home or away), breakfast consumers were older, with higher income, literacy, and a higher percentage were nonsmokers (data shown in Appendix G). Breakfast skippers had lower education levels than consumers, although there was no difference between the lowest education levels ( less than high school and some high school, no degree ) (Appendix H). On day 1 and day 2, 77.4% and 80.1% of breakfasts were eaten at home, respectively. However, only 41.2% of subjects ate breakfast at home both days and 6.3% of subjects ate both breakfasts AFH. The remaining 13.3% were breakfast switchers who were removed from regression analyses, leaving an analytical sample of 1834 (Table 1). When categorized into the three breakfast groups (skippers, home consumers, or AFH consumers), home breakfast consumers were significantly older than the other groups. AFH breakfast consumers had a significantly higher average BMI than the 15

other groups, and a higher percentage of this group was employed. This group had the lowest rates of smoking, the highest literacy scores, and the highest average annual income of the three groups. Compared to the other groups, a higher percentage of breakfast skippers were below 125% of the 2004 federal poverty guideline and smokers (Table 2; post hoc analysis Appendix I). Foods eaten at home were mainly obtained from grocery stores (96.5%), with a small portion from fast food restaurants (1.9%). Foods eaten away from home were obtained from grocery stores (45.5%), fast food restaurants (23.3%), and full service restaurants (10.2%). Breakfasts eaten at home contained less energy, and lower amounts of carbohydrates and fat. Both home and AFH breakfasts contained minimal amounts of fiber and omega-3 fatty acids (Table 3). After adjusting for energy intake, home breakfasts had slightly more protein and fiber and slightly less total fat per 100 kcal than AFH breakfasts (Table 4). Diet Quality and Breakfast Consumption Unadjusted results for the HEI total and component scores are shown in Table 5. The overall mean (±SE) HEI score for the analytical sample was 46.04 ±0.28. The total HEI scores were significantly different across breakfast groups. Home breakfast consumers had the highest mean (±SE) HEI score of 49.0 ± 0.4 and breakfast skippers had the lowest score of 43.0 ± 0.4. Home breakfast consumers had significantly higher whole-day greens and beans, whole grain, whole dairy, and refined grains scores and a lower seafood and plant protein score than AFH breakfast consumers. Home breakfast consumers also had significantly higher scores for all components except fatty acids and refined grains 16

when compared to breakfast skippers. AFH breakfast consumers had higher total fruit, whole fruit, seafood and plant proteins, fatty acid, and empty calorie scores than those who skipped breakfast; however, they had a significantly lower refined grains score. After adjusting for covariates, eating breakfast at home was associated with higher HEI scores than eating breakfast AFH or skipping breakfast (Table 6). Eating breakfast AFH was associated with a 3.59 point decrease in HEI score, and skipping breakfast was associated with a 4.66 point decrease in HEI score. Additionally, being African American, above the poverty line, older, female, with higher literacy, and lower BMI was associated with increases in HEI score. The largest increases in HEI score were related to having a degree beyond a high school diploma (associated with a 7.49 point increase) and being a nonsmoker (associated with a 6.03 point increase). Marital status, employment, and income were not significant, and were not included in the final model. Health-Related Quality of Life and Breakfast Consumption SF-12 Health Concept Scores The unadjusted SF-12 health concept scores show differences between breakfast groups (Table 7). AFH breakfast consumers had significantly higher scores in all concepts except vitality and bodily pain (this group had significantly less bodily pain) when compared to home breakfast consumers and breakfast skippers. Home breakfast consumers and breakfast skippers were not significantly different. Mental Component Summary Score The overall unadjusted mean (±SE) MCS score for the sample was 47.89 ± 0.28. The unadjusted results for the Mental Component Summary score by breakfast 17

group are shown in Table 6. AFH breakfast consumers had significantly higher scores than breakfast skippers; home breakfast consumers did not differ from either group. After adjustment for covariates, breakfast group was not significantly associated with MCS score, and therefore was not included in the final model (Table 8). However, being African American, above 125% of the poverty line, older, male, employed, and having a degree beyond a high school diploma, higher income, and higher BMI were associated with higher MCS scores. Marital status, smoking, and literacy were not significant and were not included in the final model. Physical Component Summary Score The overall unadjusted mean (±SE) PCS score for the sample was 44.93 ± 0.29. The unadjusted results for the Physical Component Summary score by breakfast group are shown in Table 7. AFH breakfast consumers had higher scores than either home breakfast consumers or breakfast skippers; home breakfast consumers and breakfast skippers did not differ. After adjustment for covariates, breakfast group was not significantly associated with PCS score, and therefore was not included in the final model (Table 9). Being African American, above 125% of the poverty line, younger, with lower BMI, higher income and a nonsmoker were associated with modest increases in PCS scores. The largest increase in PCS score was associated with being employed (increase of 6.3 points). Marital status, literacy, and sex were not significant and were not included in the final model. 18

Chapter 6 DISCUSSION To the knowledge of the author, this study was the first analysis to explore the relationship of breakfast habits (including consumed at home, consumed AFH, and skipped) and diet quality in a diverse urban population. In this analysis, participants who ate breakfast at home had higher average HEI-2010 diet quality scores than those who ate breakfast away from home. However, those who ate breakfast away from home still had higher diet quality scores than those who skipped breakfast. In this analysis, an AFH breakfast was related to a 3.59 point decrease in HEI score. This difference was slightly smaller than that shown in Mancino et al. (a decrease of 4.5 points for AFH breakfast). 37 That study used national intake data from 1994-1996 and 2003-2004, so it is possible that AFH breakfasts have improved in diet quality over the past decade. However, that study used the HEI-2005, which differs slightly from the 2010 version, for diet quality calculations. 37 Additionally, the Mancino et al study did not compare AFH breakfasts with skipped breakfasts, so further comparison with the current analysis is not possible. The association between breakfast habits and changes in diet quality may be related to the types of breakfast foods available in different locations. For example, in a study by O Neil et al, 4 a breakfast of presweetened ready-to-eat cereal with lower-fat milk was shown to increase diet quality compared to skipping. This type of breakfast may be more likely to be available at home. In the same study, a breakfast of coffee, cream and sugar, and sweets was not significantly different than skipping breakfast in terms of diet quality. Perhaps this latter type of breakfast may be more available AFH. The association between breakfast habits and diet quality may also be due to 19

health behavior patterns. As in previous studies, in this analysis breakfast skipping was clustered with smoking, lower levels of education, and being male. Stress may also cause differences in diet quality between breakfast groups; in breakfast skippers, but not breakfast consumers, stress has been associated with empty calories and evening intake of added sugars. 34 Perceived stress was not measured in Wave 3 of this study so this relationship could not be explored. Further research on health behavior patterns including stress may help explain the association between breakfast habits and diet quality. The study findings suggest that meal location should be considered in counseling and in nutrition policy. In counseling, clients wishing to increase their diet quality could be encouraged to eat breakfast at home. If time and resources do not allow, eating breakfast AFH may still provide for higher diet quality than skipping breakfast altogether. Thus consumption of breakfast should be encouraged and perhaps more education on healthier food selections from menus when AFH is warranted. Analysis of HEI scores for five common fast food restaurants suggests that improvements are needed in AFH menus; 40 perhaps more effort should be made to increase HEI components which were lower in AFH breakfast consumers (namely greens and beans, whole grain, whole dairy, and refined grains scores). Researchers could use information about meal location to learn more about meal patterns and their relationship to health. For example, there is variation in research on breakfast and weight, with some studies finding that eating breakfast is related to lower weight 7,8,50,57 while others report breakfast is unrelated to weight (for example, the study by Dhurandhar et al 67 ). Understanding the role of meal location 20

might provide insights into the relationship of weight management and dietary patterns. The breakfast patterns of the HANDLS sample provided new information about diets of socioeconomically and racially diverse urban groups which are typically understudied due to recruitment challenges. 59 Nearly 40% of participants skipped breakfast on both days of recall, which was higher than previous studies which reported around 20% of NHANES subjects skipping breakfast. 4 6 This finding was to be expected considering that the demographics of the HANDLS study population (high percent African American and low income) more closely match previous profiles of breakfast skippers than the general NHANES sample. In the HANDLS sample, about 12% of subjects ate breakfast AFH on one of their recalls. This percentage is much lower than in Nicklas et al (25%) 35, which may be related to the fact that Nicklas et al used a sample of young adults (19-28 years old), who may have different breakfast habits than the HANDLS sample of adults (32-70 years old). The mean HEI-2010 score of the HANDLS study population was nearly 13 points lower than the NHANES 2011-2012 average for the population (score of 59). 68 As noted by Kuczmarski et al, 69 education has an important influence on diet quality in this population; in this analysis, those who obtained a degree past high school had a distinct advantage in diet quality compared to those with less than a high school education. Using baseline data from the HANDLS study, Kuczmarski et al 69 reported there was no difference in total diet quality between the races. Compared to the baseline HANDLS study, there was an increase in HEI-2010 scores of HANDLS participants in Wave 3. This difference may explain the small increase in diet quality, especially for African Americans, seen in the current study. In other studies, African 21

Americans have been shown to have lower diet quality when compared to Whites. 70 Diets of low socioeconomic, diverse populations are often low quality, 21 and improvement in these populations is necessary to increase national diet quality scores. Considering the results of this analysis, one way to improve diet quality may be to encourage people to eat breakfast at home or, if they are limited by time, to eat breakfast AFH rather than skip it. No difference was found in the health-related quality of life measures by breakfast group. It may be that because HRQoL is affected by long-term health, daily breakfast patterns (which may change over time) do not affect scores. However, breakfast skipping was associated with lower mental HRQoL in a population of adults in Taiwan. 47 This analysis may have had different results due to variations in population-wide breakfast habits or the inclusion of breakfast location in this analysis. Similar to the findings with diet quality, African Americans were found to have higher HRQoL scores than Whites. This contrasts with an analysis of the 2005-2006 National Health Measurement Study 71 which found that African Americans, particularly women, had lower HRQoL scores than Whites. Overall, the sample had SF-12 scores close to the national average of 50. As shown in other studies, higher mental health and lower physical health scores were associated with increased age. One highlight was the importance of employment in this sample, which added four mental health points and six physical health points. It may be that employment adds to a person s perception of quality of life, or it may be that people who are of better health are more likely to be able to work. In this analysis, a participant-identified approach was used to define breakfast and breakfast location. This approach avoids imposing a researcher-defined time 22

frame for breakfast consumption. Efforts to reach a consensus on a breakfast definition in research would further enable researchers to draw conclusions about breakfast and health. This analysis also used the where eaten definition of AFH foods. This definition was chosen because foods from one meal could be obtained in many places, but participants were asked to define a single place (home or away) for the consumption of the meal. However, a substantial percentage (45%) of foods eaten AFH were obtained from the grocery store in this analysis. Furthermore, grocery stores were defined to include traditional grocery stores, warehouse stores, bakeries, and other food retailers. Future researchers could further examine the where obtained definition to explore this topic. Strengths of this study include the use of a racially and socioeconomically diverse, urban sample which is often underrepresented in research. Also, the use of two 24-hour recalls is a strength, as the HEI-2010 scores were then based on an average and only participants who had the same breakfast habits on both days of recall were included, increasing the strength of the groups. Limitations include the possibility of underreporting due to the participantbased recall used in the study. If one breakfast group was more likely to underreport than others (the AFH breakfast group, for example, had a higher average BMI, which has been linked to underreporting), this could alter results; however, the USDA s Automated Multiple Pass Method has been shown to reduce bias in energy collection 72 and considering all three groups had an average BMI in the obese category, it is unlikely that one group differed from the others. Also, because this analysis was crosssectional, no conclusions can be drawn about causation. Due to the changing nature of breakfast consumption over the 10-year difference between HANDLS baseline and 23

Wave 3, as well as the small sample size in the AFH group cross-section, it was not deemed appropriate to use a longitudinal design in this study. Future research with additional HANDLS study waves, yielding larger sample sizes and more frequent measurements of breakfast habits, could surmount this limitation. 24

Chapter 7 CONCLUSION To the knowledge of the author this is the first study to consider breakfast habits and their relation to diet quality and health-related quality of life in a sample of diverse adults. Breakfasts eaten at home were associated with higher diet quality than breakfasts eaten away from home; however, breakfasts eaten away from home were associated with higher diet quality than when breakfast was skipped. If clients are counseled to eat breakfast, preferably at home, diet quality scores may increase. The difference in diet quality related to breakfast location may help to explain previous variance in breakfast research results. There was no change in health-related quality of life in association with different breakfast patterns. It may be that the daily habit of breakfast may not affect this longer-term measure of perceived health. 25

Table 1 Number and percent of participants of the Healthy Aging in Neighborhoods of Diversity across the Life Span study who report eating breakfast at home, eating breakfast away from home (AFH), or skipping breakfast on each day of recall Day 1 (n=2115) Day 2 (n=2115) Ate breakfast at home 995 (47.0%) 1030 (48.7%) Ate breakfast AFH 291 (13.8%) 256 (12.1%) Skipped Breakfast 829 (39.2%) 829 (39.2%) Table 2 Comparison of demographics for Healthy Aging in Neighborhoods of Diversity across the Life Span study participants by breakfast group Home Breakfast Consumers (n=872) AFH Breakfast Consumers (n=133) Breakfast Skippers (n=839) P- value* x ± SEM Age, yr 54.7 ± 0.3 51.2 ± 0.8 52.0 ± 0.3 <0.001 Sex (% male) 40.7 39.1 41.6 0.885 Race (% White) 39.6 36.8 37.5 0.794 Poverty Status (% below 125% of 2004 federal poverty level) 42.4 26.3 44.0 <0.001 BMI (kg/m 2 ) 30.3 ± 0.3 33.0 ± 0.8 30.9 ± 0.3 0.002 Marital Status (% 0.106 married) 31.8 38.1 30.2 Employed (%) 36.3 87.5 47.1 <0.001 Education <0.001 Less than high school 7.1 4.8 7.5 Some high school, no degree 56.4 48.8 65.6 Graduated high school 19.3 31.2 18.3 Higher degree 17.1 15.2 8.6 Smoking (%) 42.3 33.9 53.9 <0.001 Literacy (WRAT-3 score) 42.4 ± 0.3 44.2 ± 0.6 41.6 ± 0.3 Annual Income ($) 25,500 ± 28,200 ± 900 38,600 ± 2200 900 *P-value from one-way ANOVA (continuous) or Chi square test (categorical) <0.001 <0.001 26

Table 3 Comparison of nutrient profiles of breakfasts eaten at home or away from home (AFH) by participants in the Healthy Aging in Neighborhoods of Diversity across the Life Span study Home Breakfast Consumers (n=872) AFH Breakfast Consumers (n=133) P- value* x ± SEM Energy (kcals) 120 ± 2.3 155 ± 9.1 0.001 Protein (g) 4.5 ± 0.1 5.2 ± 0.4 0.111 Total Carbohydrates (g) 15.5 ± 0.3 18.8 ± 1.0 0.002 Sugar (g) 7.5 ± 0.2 9.0 ± 0.6 0.019 Fiber (g) 0.9 ± 0.03 0.9 ± 0.1 0.514 Total Fat (g) 4.6 ± 0.1 6.6 ± 0.6 0.001 Saturated Fat (g) 1.6 ± 0.1 2.3 ± 0.2 0.005 Omega-3 Fatty Acids (g) 0.1 ± 0.003 0.1 ± 0.01 0.043 *P-value from Independent samples T-test Table 4 Comparison of nutrient profiles per 100 kcal of breakfasts eaten at home or away from home (AFH) by participants in the Healthy Aging in Neighborhoods of Diversity across the Life Span study Home Breakfast Consumers (n=872) AFH Breakfast Consumers (n=133) P- value* x ± SEM Protein (g/100 kcal) 3.7 ± 0.04 3.3 ± 0.1 0.002 Total Carbohydrates (g/100 kcal) 13.5 ± 0.2 13.0 ± 0.4 0.248 Sugar (g/100 kcal) 6.5 ± 0.1 6.2 ± 0.3 0.416 Fiber (g/100 kcal) 0.8 ± 0.03 0.7 ± 0.05 0.001 Total Fat (g/100 kcal) 3.7 ± 0.06 4.0 ± 0.1 0.020 Saturated Fat (g/100 kcal) 1.3 ± 0.02 1.4 ±0.05 0.089 Omega-3 Fatty Acids (g/100 kcal) 0.1 ±0.004 0.1 ± 0.007 0.858 *P-value from Independent samples T-test 27

Table 5 Comparison of Healthy Eating Index-2010 total and component scores of Healthy Aging in Neighborhoods of Diversity across the Life Span participants by breakfast group Home Breakfast Consumers (n=872) AFH Breakfast Consumers (n=133) P- value a Breakfast Skippers (n=829) P- P- value b value c x ± SEM x ± SEM Adequacy Total Vegetables 2.8 ± 0.1 2.6 ± 0.1 0.279 2.6 ± 0.1 0.004 0.990 Greens and Beans 1.5 ± 0.1 1.0 ± 0.1 0.004 1.1 ± 0.1 <0.001 0.981 Total Fruit 2.1 ± 0.1 1.9 ± 0.2 0.732 1.4 ± 0.1 <0.001 0.001 Whole Fruit 1.8 ± 0.1 1.7 ± 0.2 0.891 1.1 ± 0.1 <0.001 <0.001 Whole Grains 2.8 ± 0.1 1.8 ± 0.2 <0.001 1.5 ± 0.1 <0.001 0.336 Total Dairy 4.6 ± 0.1 3.8 ± 0.2 0.002 3.6 ± 0.1 <0.001 0.787 Total Protein Foods 4.4 ± 0.03 4.4 ± 0.1 0.932 4.2 ± 0.04 <0.001 0.011 Seafood and Plant Proteins 1.9 ± 0.1 2.2 ± 0.2 0.042 1.6 ± 0.1 0.007 <0.001 Fatty Acids* 5.2 ± 0.1 5.6 ± 0.2 0.389 5.3 ± 0.1 0.774 0.607 Moderation Sodium 3.9 ± 0.1 4.1 ± 0.2 0.735 4.5 ± 0.1 <0.001 0.202 Refined Grains 6.7 ± 0.1 5.6 ± 0.2 <0.001 6.8 ± 0.1 0.839 <0.001 Empty Calories 11.2 ± 0.2 10.9 ± 0.4 0.733 9.3 ± 0.2 <0.001 0.004 Total Score 49.0 ± 0.4 45.8 ± 0.9 0.011 43.0 ± 0.4 <0.001 0.028 a= P-value from Tukey HSD for home breakfast consumers and AFH breakfast consumers b= P-value from Tukey HSD for home breakfast consumers and breakfast skippers c= P-value from Tukey HSD for AFH breakfast consumers and breakfast skippers *= ANOVA was not significant, P>0.05 28

Table 6 Factors influencing the Healthy Aging in Neighborhoods of Diversity across the Life Span study participants Healthy Eating Index-2010 scores by multiple regression Factor Estimate Standard Error P-value Intercept 38.503 3.252 <0.001 Age, yr 0.157 0.034 <0.001 Sex (Ref: Female) -1.920 0.625 0.002 Race (Ref: White) 1.755 0.663 0.008 Poverty Status (Ref: Above 125% of 2004-1.708 0.645 0.008 federal poverty level) Race*Poverty status -0.051 1.278 0.968 BMI (kg/m 2 ) -0.120 0.040 0.003 Smoking (Ref: Nonsmoker) -6.030 0.665 <0.001 Literacy (WRAT-3 score) 0.156 0.044 <0.001 Breakfast Group (Ref: Home breakfast consumer) AFH Breakfast consumer -3.594 1.167 0.002 Breakfast skipper -4.665 0.643 <0.001 Education (Ref: Less than high school) Some high school, no degree 1.435 1.095 0.191 Graduated high school 1.760 1.276 0.168 Higher degree 7.485 1.380 <0.001 29

Table 7 Comparison of Short Form 12** summary and health concept scores of the Healthy Aging in Neighborhoods of Diversity across the Life Span study sample by breakfast group Home Breakfast Consumers (n=666) AFH Breakfast Consumers (n=113) P- value a Breakfast Skippers (n=654) P- P- value b value c x ± SEM x ± SEM Health Concepts Social Functioning 45.4 ± 0.5 49.3 ± 0.9 0.003 45.8 ± 0.5 0.801 0.010 Vitality* 49.0 ± 0.4 50.5 ± 0.9 0.307 48.8 ± 0.4 0.945 0.230 Bodily Pain 44.1 ± 0.5 48.6 ± 1.1 0.001 44.3 ± 0.5 0.938 0.003 Mental Health 48.4 ± 0.4 51.0 ± 0.9 0.050 47.5 ± 0.4 0.276 0.004 Role- Emotional 45.0 ± 0.4 49.0 ± 0.9 0.002 45.1 ± 0.4 0.984 0.002 Role-Physical 44.5 ± 0.4 50.0 ± 0.8 <0.001 45.6 ± 0.4 0.149 <0.001 Physical Function 45.6 ± 0.4 49.5 ± 0.9 0.003 45.8 ± 0.5 0.961 0.004 General Health Perceptions 44.0 ± 0.4 47.2 ± 0.9 0.017 44.1 ± 0.5 1.000 0.018 Summary Scores Mental Component Summary score 48.0 ± 0.4 50.2 ± 0.9 0.100 47.4 ± 0.4 0.639 0.030 Physical Component Summary score 44.2 ± 0.4 48.8 ± 0.8 <0.001 44.8 ± 0.4 0.495 0.001 a= P-value from Tukey HSD for home breakfast consumers and AFH breakfast consumers b= P-value from Tukey HSD for home breakfast consumers and breakfast skippers c= P-value from Tukey HSD for AFH breakfast consumers and breakfast skippers *= ANOVA was not significant, P>0.05 **= Ware JJ, Kosinski MM, Keller SSD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care. 1996;34(3):220-233. doi:10.2307/3766749. 30

Table 8 Factors influencing the Healthy Aging in Neighborhoods of Diversity across the Life Span study participants Mental Component Summary scores by multiple regression Factor Estimate Standard P-value Error Intercept 27.512 2.761 <.001 Race (Ref: White) 2.729 0.676 <.001 Poverty Status (Ref: Above 125% of 2004-0.862 0.752 0.252 federal poverty level) Race*Poverty status -1.557 1.345 0.247 Age, yr 0.180 0.037 <.001 Sex (Ref: Female) 1.777 0.656 0.007 Education (Ref: Less than high school) Some high school, no degree 3.595 1.183 0.002 Graduated high school 2.828 1.346 0.036 Higher degree 3.771 1.421 0.008 Annual Income ($) 0.00005 0.00001 0.003 BMI (kg/m 2 ) 0.082 0.041 0.044 Employment (Ref: Unemployed) 4.263 0.695 <.001 Table 9 Factors influencing the Healthy Aging in Neighborhoods of Diversity across the Life Span study participants Physical Component Summary scores by multiple regression Factor Estimate Standard P-value Error Intercept 64.114 2.848 <.001 Race (Ref: White) 1.315 0.665 0.048 Poverty Status (Ref: Above 125% of 2004-0.635 0.767 0.408 federal poverty level) Race*Poverty status -0.266 1.370 0.846 Age, yr -0.225 0.038 <.001 Annual Income ($) 0.00008 0.00001 <.001 BMI (kg/m 2 ) -0.395 0.043 <.001 Employment (Ref: Unemployed) 6.326 0.704 <.001 Smoking (Ref: Nonsmoker) -2.605 0.734 <.001 31

Figure 1 Healthy Aging in Neighborhoods of Diversity across the Life Span study participant flowchart 32

REFERENCES 1. Leech RM, Worsley A, Timperio A, McNaughton SA. Understanding meal patterns: definitions, methodology and impact on nutrient intake and diet quality. Nutr Res Rev. 2015:1-21. doi:10.1017/s0954422414000262. 2. O Neil CE, Byrd-Bredbenner C, Hayes D, Jana L, Klinger SE, Stephenson- Martin S. The role of breakfast in health: Definition and criteria for a quality breakfast. J Acad Nutr Diet. 2014;114(12):S8-S26. doi:10.1016/j.jand.2014.08.022. 3. Leech RM, Worsley A, Timperio A, McNaughton SA. Characterizing eating patterns: A comparison of eating occasion definitions1. Am J Clin Nutr. 2015;102(5):1229-1237. doi:10.3945/ajcn.115.114660. 4. O Neil CE, Nicklas TA, Fulgoni VL. Nutrient intake, diet quality, and weight/adiposity parameters in breakfast patterns compared with no breakfast in adults: National health and nutrition examination survey 2001-2008. J Acad Nutr Diet. 2014;114(12):S27-S43. doi:10.1016/j.jand.2014.08.021. 5. Deshmukh-Taskar PR, Radcliffe JD, Liu Y, Nicklas T a. Do breakfast skipping and breakfast type affect energy intake, nutrient intake, nutrient adequacy, and diet quality in young adults? NHANES 1999-2002. J Am Coll Nutr. 2010;29(4):407-418. doi:10.1080/07315724.2010.10719858. 6. Kant AK, Andon MB, Angelopoulos TJ, Rippe JM. Association of breakfast 33

energy density with diet quality and body mass index in American adults: National Health and Nutrition Examination Surveys, 1999-2004. Am J Clin Nutr. 2008;88(5):1396-1404. 7. Odegaard AO, Jacobs DR, Steffen LM, Van Horn L, Ludwig DS, Pereira MA. Breakfast frequency and development of metabolic risk. Diabetes Care. 2013;36(10):3100-3106. doi:10.2337/dc13-0316. 8. Kant A, Graubard B. With-in person comparison of eating behaviors, time of eaiting, and dietary intake on days with and without breakfast: NHANES 2005-2010. Am J Clin Nutr. 2015;In this is. doi:10.3945/ajcn.115.110262.1. 9. Usda, Dhhs. Dietary Guidelines for Americans, 2010. 7Th Ed. 2010:76. doi:10.1016/s0300-7073(05)71075-6. 10. Lachat C, Nago E, Verstraeten R, Roberfroid D, Van Camp J, Kolsteren P. Eating out of home and its association with dietary intake: A systematic review of the evidence. Obes Rev. 2012;13(4):329-346. doi:10.1111/j.1467-789x.2011.00953.x. 11. An R. Fast-food and full-service restaurant consumption and daily energy and nutrient intakes in US adults. Eur J Clin Nutr. 2015;(August 2014):1-7. doi:10.1038/ejcn.2015.104. 12. Beydoun M a, Powell LM, Wang Y. Reduced away-from-home food 34

expenditure and better nutrition knowledge and belief can improve quality of dietary intake among US adults. Public Health Nutr. 2009;12(3):369-381. doi:10.1017/s1368980008002140. 13. Kant AK, Whitley MI, Graubard BI. Away from home meals: associations with biomarkers of chronic disease and dietary intake in American adults, NHANES 2005-2010. Int J Obes (Lond). 2015;39(5):820-827. doi:10.1038/ijo.2014.183. 14. Schwingshackl L, Hoffman G. Diet quality as assessed by the Healthy Eating Index, the Alternate Healthy Eating Index, the Dietary Approaches to Stop Hypertension score, and health outcomes: A systematic review and metaanalysis of cohort studies. J Acad Nutr Diet. 2015;115(5):780-800. 15. Guenther PM, Kirkpatrick SI, Reedy J, et al. The Healthy Eating Index-2010 is a valid and reliable measure of diet quality according to the 2010 Dietary Guidelines for Americans. J Nutr. 2014;144(3):399-407. doi:10.3945/jn.113.183079. 16. Harmon BE, Boushey CJ, Shvetsov YB, et al. Associations of key diet-quality indexes with mortality in the Multiethnic Cohort: The dietary patterns methods project. Am J Clin Nutr. 2015;101(3):587-597. doi:10.3945/ajcn.114.090688. 17. Reedy J, Krebs-Smith S, Miller P, et al. Higher diet quality is associated with decreased risk of all-cause, cardiovascular disease, and cancer mortality among older adults. J Nutr. 2014;144(6):881-889. doi:10.3945/jn.113.189407. 35

18. Bi H, Gan Y, Yang C, Chen Y, Tong X, Lu Z. Breakfast skipping and the risk of type 2 diabetes: a meta-analysis of observational studies. Public Heal Nutr. 2015;18(16):3013-3019. doi:10.1017/s1368980015000257. 19. Nicklas TA, O Neil CE, Fulgoni VL. Diet quality is inversely related to cardiovascular risk factors in adults. J Nutr. 2012;142(12):2112-2118. doi:10.3945/jn.112.164889. 20. Wilson MM, Reedy J, Krebs-Smith SM. American Diet Quality: Where It Is, Where It Is Heading, and What It Could Be. J Acad Nutr Diet. 2016;116(2):302-310. doi:10.1016/j.jand.2015.09.020. 21. Hiza HAB, Casavale KO, Guenther PM, Davis CA. Diet Quality of Americans Differs by Age, Sex, Race/Ethnicity, Income, and Education Level. J Acad Nutr Diet. 2013;113(2):297-306. doi:10.1016/j.jand.2012.08.011. 22. Woodruff SJ, Hanning RM. Effect of meal environment on diet quality rating. Can J Diet Pract Res. 2009;70(3):118-124. doi:10.3148/70.3.2009.118. 23. Centers for Disease Control and Prevention. No Title. Health-related quality of life concepts. http://www.cdc.gov/hrqol/concept.htm. Published 2011. Accessed April 28, 2016. 24. Busija L, Pausenberger E, Haines TP, Haymes S, Buchbinder R, Osborne RH. Adult measures of general health and health-related quality of life: Medical 36

Outcomes Study Short Form 36-Item (SF-36) and Short Form 12-Item (SF-12) Health Surveys, Nottingham Health Profile (NHP), Sickness Impact Profile (SIP), Medical Outcomes Study Sh. Arthritis Care Res (Hoboken). 2011;63(S11):S383 S412. doi:10.1002/acr.20541. 25. Utah Dept of Health. Interpreting the SF-12 Health Survey. Utah Heal Status Surv. 2001:17. http://health.utah.gov/opha/publications/2001hss/sf12/sf12_interpreting.pdf. 26. Wee CC, Davis RB, Hamel MB. Comparing the SF-12 and SF-36 health status questionnaires in patients with and without obesity. Health Qual Life Outcomes. 2008;6:11. doi:10.1186/1477-7525-6-11. 27. Tsai S-Y, Chi L-Y, Lee C-H, Chou P. Health-related quality of life as a predictor of mortality among community-dwelling older persons. Eur J Epidemiol. 2007;22(1):19-26. doi:10.1007/s10654-006-9092-z. 28. Dorr DA, Jones SS, Burns L, et al. Use of health-related, quality-of-life metrics to predict mortality and hospitalizations in community-dwelling seniors. J Am Geriatr Soc. 2006;54(4):667-673. doi:10.1111/j.1532-5415.2006.00681.x. 29. Brown DS, Thompson WW, Zack MM, Arnold SE, Barile JP. Associations between health-related quality of life and mortality in older adults. Prev Sci. 2015;16(1):21-30. doi:10.1007/s11121-013-0437-z. 37

30. Slagter SN, Van Vliet-Ostaptchouk J V., Van Beek AP, et al. Health-Related quality of life in relation to obesity grade, type 2 diabetes, metabolic syndrome and inflammation. PLoS One. 2015;10(10). doi:10.1371/journal.pone.0140599. 31. Jiang Y, Hesser JE. Associations between health-related quality of life and demographics and health risks. Results from Rhode Island s 2002 behavioral risk factor survey. Health Qual Life Outcomes. 2006;4(1):14. doi:10.1186/1477-7525-4-14. 32. Albertson AM, Wold AC, Joshi N. Ready-to-eat cereal consumption patterns: The relationship to nutrient intake, whole grain intake, and body mass index in an older American population. J Aging Res. 2012;2012. doi:10.1155/2012/631310. 33. Fulgoni VL, Chu Y, O Shea M, Slavin JL, DiRienzo MA. Oatmeal consumption is associated with better diet quality and lower body mass index in adults: the National Health and Nutrition Examination Survey (NHANES), 2001-2010. Nutr Res. 2015;35(12):1052-1059. doi:10.1016/j.nutres.2015.09.015. 34. Widaman AM, Witbracht MG, Forester SM, Laugero KD, Keim NL. Chronic Stress Is Associated with Indicators of Diet Quality in Habitual Breakfast Skippers. J Acad Nutr Diet. 2014;116(11):1776-1784. doi:10.1016/j.jand.2016.03.016. 38

35. Nicklas TA, Myers L, Reger C, Beech B, Berenson GS. Impact of breakfast consumption on nutritional adequacy of the diets of young adults in Bogalusa, Louisiana: ethnic and gender contrasts. J Am Diet Assoc. 1998;98(12):1432-1438. doi:10.1016/s0002-8223(98)00325-3. 36. Kerver JM, Yang EJ, Obayashi S, Bianchi L, Song WO. Meal and snack patterns are associated with dietary intake of energy and nutrients in US adults. J Am Diet Assoc. 2006;106(1):46-53. doi:10.1016/j.jada.2005.09.045. 37. Mancino L, Todd J, Lin BH. Separating what we eat from where: Measuring the effect of food away from home on diet quality. Food Policy. 2009;34(6):557-562. doi:10.1016/j.foodpol.2009.09.003. 38. Todd JE, Mancino L, Lin B-H. The Impact of Food Away from Home on Adult Diet Quality. SSRN elibrary. 2010;(90). http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1557129. 39. Nguyen BT, Powell LM. The impact of restaurant consumption among US adults: Effects on energy and nutrient intakes. Public Health Nutr. 2013;17(11):2445-2452. doi:10.1017/s1368980014001153. 40. Kirkpatrick SI, Reedy J, Kahle LL, Harris JL, Ohri-Vachaspati P, Krebs-Smith SM. Fast-food menu offerings vary in dietary quality, but are consistently poor. Public Health Nutr. 2013;17(4):924-931. doi:10.1017/s1368980012005563. 39

41. Lin B-H, Wendt M, Guthrie JF. Impact on energy, sodium and dietary fibre intakes of vegetables prepared at home and away from home in the U.S.A. Public Health Nutr. 2013;16(11):1937-1943. doi:10.1017/s1368980013001602. 42. Urban LE, Weber JL, Heyman MB, et al. Energy Contents of Frequently Ordered Restaurant Meals and Comparison with Human Energy Requirements and US Department of Agriculture Database Information: A Multisite Randomized Study. J Acad Nutr Diet. 2016;116(4):590-598.e6. doi:10.1016/j.jand.2015.11.009. 43. Cohen DA, Bhatia R. Nutrition standards for away-from-home foods in the USA. Obes Rev. 2012;13(7):618-629. doi:10.1111/j.1467-789x.2012.00983.x. 44. Littlewood JA, Lourenço S, Iversen CL, Hansen GL. Menu labelling is effective in reducing energy ordered and consumed: a systematic review and meta-analysis of recent studies. Public Health Nutr. 2015;19(12):1-16. doi:10.1017/s1368980015003468. 45. Powell LM, Nguyen BT, Han E. Energy intake from restaurants: Demographics and socioeconomics, 2003-2008. Am J Prev Med. 2012;43(5):498-504. doi:10.1016/j.amepre.2012.07.041. 46. de Castro JM, King GA, Duarte-Gardea M, Gonzalez-Ayala S, Kooshian CH. Overweight and obese humans overeat away from home. Appetite. 2012;59(2):204-211. doi:10.1016/j.appet.2012.04.020. 40

47. Huang C-J, Hu H-T, Fan Y-C, Liao Y-M, Tsai P-S. Associations of breakfast skipping with obesity and health-related quality of life: evidence from a national survey in Taiwan. Int J Obes (Lond). 2010;34(4):720-725. doi:10.1038/ijo.2009.285. 48. Keski-Rahkonen A, Kaprio J, Rissanen A, Virkkunen M, Rose RJ. Breakfast skipping and health-compromising behaviors in adolescents and adults. Eur J Clin Nutr. 2003;57(7):842-853. doi:10.1038/sj.ejcn.1601618. 49. Betts JA, Chowdhury EA, Gonzalez JT, Richardson JD, Tsintzas K, Thompson D. Is breakfast the most important meal of the day? Proc Nutr Soc. 2016;(September):1-11. doi:10.1017/s0029665116000318. 50. Timlin MT, Pereira MA. Breakfast Frequency and Quality in the Etiology of Adult Obesity and Chronic Diseases. Nutr Rev. 2008;65(6):268-281. doi:10.1111/j.1753-4887.2007.tb00304.x. 51. Bezerra IN, Curioni C, Sichieri R. Association between eating out of home and body weight. Nutr Rev. 2012;70(2):65-79. doi:10.1111/j.1753-4887.2011.00459.x. 52. Germain L, Latarche C, Kesse-Guyot E, Galan P, Hercberg S, Briançon S. Does Compliance with Nutrition Guidelines Lead to Healthy Aging? A Quality-of- Life Approach. J Acad Nutr Diet. 2013;113(2):228-240. doi:10.1016/j.jand.2012.10.015. 41

53. Milte CM, Thorpe MG, Crawford D, Ball K, McNaughton SA. Associations of diet quality with health-related quality of life in older Australian men and women. Exp Gerontol. 2015;64:8-16. doi:10.1016/j.exger.2015.01.047. 54. Pérez-Tasigchana RF, León-Muñoz LM, López-García E, Banegas JR, Rodríguez-Artalejo F, Guallar-Castillón P. Mediterranean diet and healthrelated quality of life in two cohorts of community-dwelling older adults. PLoS One. 2016;11(3). doi:10.1371/journal.pone.0151596. 55. Muñoz M-A, Fíto M, Marrugat J, Covas M-I, Schröder H. Adherence to the Mediterranean diet is associated with better mental and physical health. Br J Nutr. 2009;101(12):1821-1827. doi:10.1017/s0007114508143598. 56. Bonaccio M, Di Castelnuovo A, Bonanni A, et al. Adherence to a Mediterranean diet is associated with a better health-related quality of life: a possible role of high dietary antioxidant content. BMJ Open. 2013;3(8):e003003-. doi:10.1136/bmjopen-2013-003003. 57. Ma Y, Bertone ER, Stanek EJ, et al. Association between eating patterns and obesity in a free-living US adult population. Am J Epidemiol. 2003;158(1):85-92. doi:10.1093/aje/kwg117. 58. Devaraj S, Wang-Polagruto J, Polagruto J, Keen CL, Jialal I. High-fat, energydense, fast-food-style breakfast results in an increase in oxidative stress in metabolic syndrome. Metabolism. 2008;57(6):867-870. 42

doi:10.1016/j.metabol.2008.02.016. 59. Evans M, Lepkowski JM, Powell N, LaVeist T, Kuczmarski MF, Zonderman AB. Healthy Aging in Neighborhoods of Diversity across the Life Span (HANDLS): overcoming barriers to implementing a longitudinal, epidemiologic, urban study of health, race, and socioeconomic status. Ethn Dis. 2010;20:267-275. 60. HANDLS Research Protocol. HANDLS Research Protocol. https://handls.nih.gov/pdf/02protocol-wave03-2013-05-17.pdf. Published 2013. Accessed June 1, 2016. 61. Raper N, Perloff B, Ingwersen L, Steinfeldt L, Anand J. An overview of USDA s Dietary Intake Data System. J Food Compos Anal. 2004;17(3-4):545-555. doi:10.1016/j.jfca.2004.02.013. 62. Agricultural Research Service; Food Surveys Research Group. USDA Food and Nutrient Database for Dietary Studies. http://www.ars.usda.gov/services/docs.htm?docid=12089. 63. Ware JJ, Kosinski MM, Keller SSD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care. 1996;34(3):220-233. doi:10.2307/3766749. 64. Jakobsson U. Using the 12-item Short Form health survey (SF-12) to measure 43

quality of life among older people. Aging Clin Exp Res. 2007;19(6):457-464. http://www.ncbi.nlm.nih.gov/pubmed/18172367. 65. Resnick B, Nahm ES. Reliability and validity testing of the revised 12-item Short-Form Health Survey in older adults. J Nurs Meas. 2001;9(2):151-161. 66. SF-12 : How to Score the SF-12 Physical and Mental Health Summary Scales. 2016;(January 1998). 67. Dhurandhar EJ, Dawson J, Alcorn A, et al. The effectiveness of Breakfast recommendations on weight loss: A randomized controlled trial. Am J Clin Nutr. 2014;100(2):507-513. doi:10.3945/ajcn.114.089573. 68. USDA Center for Nutrition Policy and Promotion. Healthy Eating Index. https://www.cnpp.usda.gov/healthyeatingindex. Accessed December 19, 2016. 69. Kuczmarski MF, Beydoun MF, Cotugna N, et al. Literacy contributes to greater higher diet quality in a socioeconomically diverse urban prospective cohort. Top Clin Nutr. 2016;31(1):47-58. doi:10.1097/tin.0000000000000057. 70. Hsiao PY, Mitchell DC, Coffman DL, et al. Dietary patterns and diet quality among diverse older adults: The university of Alabama at Birmingham study of aging. J Nutr Heal Aging. 2013;17(1):19-25. doi:10.1007/s12603-012-0082-4. 71. Pereira C, Palta M, Mullahy J, Fryback D. Race and preference-based healthrelated quality of life measures in the United States. Qual Life Res. 44

2011;20(6):969-978. doi:10.1007/s11136-010-9813-3.race. 72. Moshfegh AJ, Rhodes DG, Baer DJ, et al. The US Department of Agriculture Automated Multiple-Pass Method reduces bias in the collection of energy intakes. Am J Clin Nutr. 2008;88(2):324-332. doi:88/2/324 [pii]. 45

Appendix A INFORMED CONSENT 46

47

48

49

50

51

52

53