Association between diet quality and measures of body adiposity using the Rate Your Plate survey in patients presenting for coronary angiography

Abstract

Background

Diet is a modifiable risk factor for cardiovascular disease; however, dietary patterns are historically difficult to capture in the clinical setting. Healthcare providers need assessment tools that can quickly summarize dietary patterns. Research should evaluate the effectiveness of these tools, such as Rate Your Plate (RYP), in the clinical setting.

Hypothesis

RYP diet quality scores are associated with measures of body adiposity in patients referred for coronary angiography.

Methods

Patients without a history of coronary revascularization (n = 400) were prospectively approached at a tertiary medical center in New York City prior to coronary angiography. Height, weight, and waist circumference (WC) were measured; body mass index (BMI) and waist‐to‐height ratio (WHtR) were calculated. Participants completed a 24‐question RYP diet survey. An overall score was computed, and participants were divided into high (≥58) and low (≤57) diet quality groups.

Results

Participants in the high diet quality group (n = 98) had significantly lower measures of body adiposity than did those in the low diet quality group (n = 302): BMI (P < 0.001), WC (P = 0.001), WHtR (P = 0.001). There were small but significant inverse correlations between diet score and BMI, WC, and WHtR (P < 0.001). These associations remained significant after adjustment for demographics, tobacco use, and socioeconomic factors.

Conclusions

Higher diet quality scores are associated with lower measures of body adiposity. RYP is a potential instrument to capture diet quality in a high‐volume clinical setting. Further research should evaluate the utility of RYP in cardiovascular risk‐factor control.

1. INTRODUCTION

Diet is a modifiable risk factor for cardiovascular disease (CVD), which remains the leading cause of mortality in the United States.1 Although data on the optimal diet for heart health is mixed, research shows that dietary choices influence the risk for atherosclerotic heart disease.2 The American Heart Association provides specific recommendations for intake of fruit, vegetables, whole grains, animal protein, different types of fat, sodium, and sugar.3 Healthcare providers therefore should assess a patient's diet as part of routine cardiac care. However, dietary patterns remain difficult to capture in a convenient manner. Traditional, well‐validated diet assessment methods—including interviewer‐administered 24‐hour dietary recalls, participant‐recorded food records, and food frequency questionnaires (FFQs)—are time‐consuming, require trained professionals, and rely on the use of dietary software programs.4 Thus, these methods of diet assessment are not feasible to be administered in a clinical waiting‐room setting.

Brief dietary assessment tools, or diet screeners, have been developed to quickly assess dietary patterns. However, 1‐time assessment tools come with their own limitations; many validated tools are designed specifically to capture information about a particular nutrient, and others have not been well validated in research literature. Brief tools such as the National Cancer Institute's Quick Food Scans for fruit and vegetable intake and percentage energy from fat, the National Health and Nutrition Examination Survey Brief Diet Screener, and Block Fruit/Vegetable/Fiber and Dietary Fat Screeners are useful for assessing a particular nutrient or food group, but do not capture overall dietary patterns.5, 6, 7, 8, 9 Other short screeners, like Rate Your Plate (RYP), including Rapid Eating Assessment for Patients, Starting The Conversation, and Healthy Eating Vital Signs, have been developed for use in clinical and community settings but have not been widely validated in large clinical trials.10, 11, 12, 13

RYP is a 24‐question tool validated to determine whether an individual's diet is relatively high or low in saturated fat. RYP was designed to capture qualitative information related to the intake of fat, animal protein, fruit, vegetables, whole grains, processed foods, and desserts. This ultimately provides an estimate of an individual's habitual eating patterns in a convenient, easy‐to‐score manner.10

It is well established that individuals with unhealthy measures of body adiposity, including overweight/obese body mass index (BMI), waist circumference (WC), and weight‐to‐height ratio (WHtR) are at increased risk of CVD.14, 15, 16 Although many studies have validated diet surveys with other methods of diet assessment, few have examined the efficacy of a tool designed for the rapid assessment of overall diet, such as RYP, and compared scores with risk of disease, such as measures of body adiposity.17 The objective of this analysis was to examine the association between self‐reported diet quality using RYP and measures of body adiposity in patients presenting for coronary angiography.

2. METHODS

Patients presenting for coronary angiography were enrolled in a prospective study to determine if the knowledge of their coronary artery disease status on angiogram changed their dietary habits. Participants were recruited from 2 sites at a tertiary medical center in New York City between February 2015 and February 2017. Potential participants were approached prior to their scheduled procedure, and those who had previously undergone coronary revascularization, presented emergently, or spoke a language other than English or Spanish as their primary language were excluded. All participants provided written informed consent. The study was approved by the New York University School of Medicine institutional review board and registered at http://www.clinicaltrials.gov (NCT02382250). In this analysis, baseline data obtained prior to coronary angiography were evaluated.

Participants completed an interviewer‐administered RYP survey. Gans et al. validated the RYP survey with the well‐validated Willett semiquantitative FFQ in 102 participants of a cross‐sectional household survey. Comparisons of the RYP survey and Willett questionnaires demonstrated modest but significant inverse correlations between RYP scores and intake of saturated fats, cholesterol, calories from fat, and grams of fat estimated by the Willett FFQ (r ranges from −0.37 to −0.48).10

The RYP survey takes approximately 10 minutes to complete, uses simple language, and is available in both English and Spanish. RYP is formatted into 3 columns, with column A answers including the most heart‐healthy choices, column B answers including suboptimal choices, and column C answers including the least heart‐healthy choices. Each response corresponds with a point value for the total RYP score (column A = 3 points, column B = 2 points, column C = 1 point). Total scores range from 24 to 72, with higher scores indicating better diet quality. Scores of 24 to 40 indicate that the person has much room for improvement in their diet; scores of 41 to 57 indicate that the person is making occasional heart‐healthy choices; and scores of 58 to 72 indicate that the person is making frequent heart‐healthy choices.18

The outcomes of interest included measures of adiposity (BMI, WC, and WHtR). Trained professionals measured height (Seca 216 stadiometer, model 1814009; Seca, Chino, CA), weight (Weight Watchers scale model WW24WN; Conair Corp., East Windsor, NJ), and WC (tape measure model 35‐780‐000; Mabis Healthcare, Waukegan, IL). WC was measured at the level of the iliac crest while the participant was in standing position. BMI was calculated by dividing weight in kilograms by height in meters squared (kg/m²). WHtR was calculated by dividing WC in centimeters by height in centimeters. Covariates of interest including demographics, socioeconomic factors, and medical history were obtained by direct interview, and uncertain responses were determined from the participant's electronic medical record.

2.1. Statistical analysis

Participants were divided into 2 groups based on their RYP survey scores: high diet quality (RYP score ≥ 58) and low diet quality (RYP score ≤ 57). Continuous variables are presented as median (interquartile range) and compared between the high and low diet quality groups using the Mann–Whitney test. Categorical variables are presented as n (%) and compared between the high and low diet quality groups using the Fisher exact test or χ² test. Correlations between the continuous measures of adiposity and the continuous RYP score were examined using the Spearman correlation test. The association between measures of body adiposity and RYP score were evaluated using a linear regression model. The model was adjusted for age, sex, race, and ethnicity (model 1), followed by additional adjustment for current tobacco use (defined as use within the past 6 months; model 2), and finally, additional adjustment for highest level of education, employment status, and marital status (model 3). Level of significance was set at a 2‐sided α level of 0.05. Statistical analysis was conducted using SPSS software, version 23 (IBM Corp., Armonk, NY).

3. RESULTS

Of the 478 participants who met inclusion criteria and were approached for participation in the study, 400 (84%) participants provided written informed consent and completed a RYP diet survey. Characteristics for participants in the high (n = 98) vs low (n = 302) diet quality groups are shown in Table 1. Notably, there was a significantly higher proportion of Asians and “other” races and participants who were married or in a domestic partnership in the high diet quality group. There was also a significantly lower proportion of participants of Hispanic ethnicity and who currently used tobacco in the high diet quality group. A Spanish version of the RYP diet survey was administered in 44 (11%) participants (12.3% high diet quality group vs 7.1% low diet quality group; P = 0.20).

Table 1.

Baseline characteristics of patients referred for coronary angiography and administered the Rate Your Plate diet survey

Variable	High Diet Quality (RYP score ≥ 58), n = 98	Low Diet Quality (RYP score ≤ 57), n = 302	P Value
Age, yrs	64 (55–72)	60 (52–68)	0.08
Male sex	52.0 (51)	60.6 (183)	0.16
Race			<0.001
White	50.0 (49)	56.0 (169)
Black	20.4 (20)	32.5 (98)
Asian	18.4 (18)	7.3 (22)
Other	11.2 (11)	4.3 (13)
Hispanic ethnicity	13.3 (13)	23.8 (72)	0.03
Highest education level			0.30
Less than high school	16.3 (16)	22.9 (69)
High school or some college	46.9 (46)	46.5 (140)
4‐year college or more	36.7 (36)	30.6 (92)
Employment status			0.20
Full time	38.1 (37)	33.9 (102)
Part time, looking for work, student, or homemaker	14.4 (14)	22.9 (69)
Retired or disabled	47.4 (46)	43.2 (130)
Marital status			0.018
Single and never married	13.5 (13)	27.8 (84)
Married or domestic partnership	61.5 (59)	52.0 (157)
Previously married	25.0 (24)	20.2 (61)
HTN	70.4 (69)	72.2 (218)	0.80
Hyperlipidemia	71.4 (70)	63.6 (192)	0.18
DM	35.7 (35)	35.1 (106)	0.90
CHF	5.2 (5)	9.3 (28)	0.29
CKD	5.1 (5)	5.3 (16)	1.00
Current tobacco usea	8.2 (8)	22.2 (67)	0.002
Indications for coronary angiogram			0.19
Atypical symptoms or evidence of CAD on noninvasive imaging	55.1 (54)	55.6 (168)
Stable angina	27.6 (27)	20.2 (61)
UA, MI	17.3 (17)	24.2 (73)
Severity of CAD			0.35
Angiographically normal	21.4 (21)	26.5 (80)
Mild to moderate disease	41.8 (41)	34.1 (103)
Severe disease	36.7 (36)	39.4 (119)

Abbreviations: CAD, coronary artery disease; CHF, congestive heart failure; CKD, chronic kidney disease; DM, diabetes mellitus; HTN, hypertension; IQR, interquartile range; MI, myocardial infarction; RYP, Rate Your Plate; UA, unstable angina.

Categorical data are presented as n (%) and continuous data are presented as median (IQR).

^aCurrent tobacco use is defined as use within past 6 months.

Data on BMI were missing in 1 (0.3%) participant, WC was missing in 31 (7.8%) participants, and WHtR was missing in 32 (8.0%) participants. Participants in the high diet quality group had significantly lower unadjusted measures of body adiposity than did participants in the low diet quality group (BMI: 26.9 kg/m² [23.8–30.4] vs 29.2 kg/m² [25.8–34.1], P < 0.001; WC: 99.0 cm [88.9–109.0] vs 104.6 cm [96.3–114.3], P = 0.001; WHtR: 0.59 [0.53–0.64] vs 0.62 [0.57–0.68], P = 0.001). There were small but significant inverse unadjusted correlations between diet quality score and BMI (r = −0.22, P < 0.001), WC (r = −0.23, P < 0.001), and WHtR (r = −0.20, P < 0.001).

The unadjusted and adjusted associations between measures of body adiposity and diet quality score are shown in Table 2. The RYP scores remained significantly associated with all measures of body adiposity after adjustment for demographics, tobacco use, and socioeconomic factors (BMI: R ² = 0.21, β [95% confidence interval]: −0.24 [−0.35 to −0.13], P < 0.001; WC: R ² = 0.21, β: −0.10 [−0.15 to −0.05], P < 0.001; and WHtR: R ² = 0.21, β: −15.7 [−23.7 to −7.6], P < 0.001).

Table 2.

Associations between measures of body adiposity and Rate Your Plate diet score in patients referred for coronary angiography

Measures of Body Adiposity	R2	β (95% CI)	P Value
BMI
Unadjusted	0.045	−0.24 (−0.35 to −0.13)	<0.001
Model 1a	0.152	−0.21 (−0.32 to −0.10)	<0.001
Model 2b	0.188	−0.23 (−0.34 to −0.13)	<0.001
Model 3c	0.206	−0.24 (−0.35 to −0.13)	<0.001
WC
Unadjusted	0.062	−0.12 (−0.17 to −0.07)	<0.001
Model 1a	0.163	−0.10 (−0.14 to −0.05)	<0.001
Model 2b	0.197	−0.10 (−0.15 to −0.06)	<0.001
Model 3c	0.209	−0.10 (−0.15 to −0.05)	<0.001
WHtR
Unadjusted	0.041	−16.4 (−24.6 to −8.3)	<0.001
Model 1a	0.162	−15.5 (−23.4 to −7.5)	<0.001
Model 2b	0.194	−16.2 (−24.0 to −8.3)	<0.001
Model 3c	0.206	−15.7 (−23.7 to −7.6)	<0.001

Abbreviations: BMI, body mass index; CI, confidence interval; WC, waist circumference; WHtR, waist‐to‐height ratio.

^aModel 1 adjusts for age, sex, race, and ethnicity.

^bModel 2 adjusts for age, sex, race, ethnicity, and current tobacco use.

^cModel 3 adjusts for age, sex, race, ethnicity, current tobacco use, highest education level, employment status, and marital status.

4. DISCUSSION

This study is the first to examine the use of the RYP survey administered in patients presenting for coronary angiography. Results demonstrate a significant inverse association between RYP scores and anthropometric measures. These associations remain significant after adjustment for demographics, tobacco use, and socioeconomic factors. These findings support RYP as a useful tool to quickly and conveniently capture diet quality in a high‐volume clinical setting.

Results support the association between healthy eating and improved measures of body adiposity. Prior studies have shown that RYP is an effective tool to follow changes in diet quality and BMI. Vargas et al. used a modified version of RYP to assess whether integrating nutrition education into the cardiovascular (CV) curriculum changed the eating habits of 32 medical students. Results suggested that the students' heart‐healthy eating patterns improved.19 A pilot study by Kulick et al. used RYP to evaluate diet quality at baseline and follow‐up in 61 patients with dyslipidemia. Intervention with nutrition counseling was associated with significant improvements in RYP scores, as well as statistically significant reductions in low‐density lipoprotein cholesterol and BMI.20 The current study supports these results in a larger group of participants in a clinical setting.

Other brief assessment tools have been validated with measures of body adiposity in CV patients. An analysis of the 7447 Prevención con Dieta Mediterránea (PREDIMED) trial participants found a small but significant inverse linear association between scores from a 14‐item Mediterranean diet–adherence questionnaire and measures of adiposity in adults at high CV risk. These unadjusted correlations (r) between the Mediterranean diet score and anthropometric measures ranged from −0.058 to −0.132, whereas the unadjusted correlations between RYP and anthropometric measures in the current study ranged from −0.20 to −0.23. Overall, both tools demonstrated a small but significant inverse correlation between diet quality and anthropometric measures. Although the Mediterranean diet–adherence questionnaire captures dietary information of interest, the questions are asked in terms of quantity of servings and grams, which requires a participant to be able to accurately estimate portion sizes without nutrition education. Given the time constraint in a waiting‐room setting, a survey tool with self‐explanatory serving sizes and questions related to frequency vs amount may be more appropriate.21

Similar to RYP, the Dietary Risk Assessment (DRA) is another dietary tool that was validated with a FFQ. Although the DRA was determined to be an appropriate tool to guide dietary counseling in CVD prevention programs, the questionnaire is not as useful in a time‐constrained setting, as it consists of 54 questions with open‐ended choices about number of servings of specific foods. Furthermore, the DRA was designed for a Southern US patient population and includes foods commonly found in a Southern diet rather than a more generalizable diet.22

4.1. Study limitations

There are several limitations of the current study. First, physical‐activity data, a known confounder for measures of body adiposity, were not collected. Second, interviewing participants in a waiting‐room setting prior to planned coronary angiography may have influenced responses. Third, a limitation of RYP, like any self‐reported dietary intake collection tool, is that participants tend to underreport dietary intake. Furthermore, at the time of study design and implementation, only English and Spanish versions of RYP were available; therefore, results of the current study are not applicable to individuals who do not speak English or Spanish as their primary language. Finally, a limitation of dietary‐assessment tools is that dietary recommendations continue to evolve, and adaptations to the original tools are made to reflect contemporary dietary recommendations. RYP was adapted from a tool originally designed to capture saturated fat in the American diet, and it has since been updated to be consistent with national dietary guidelines from the American Heart Association, US Department of Health and Human Services, and the US Department of Agriculture.10, 18, 23

5. CONCLUSION

Assessment of a patient's dietary patterns is necessary to determine appropriate dietary and lifestyle modifications for CV risk reduction. Education and behavioral change are imperative separate areas of focus but are based heavily on the initial accurate evaluation of dietary patterns. The RYP survey supports the growing trend for food‐based vs nutrient‐based assessments, and it may be easily and quickly administered in a high‐volume clinical setting. Results suggest that RYP scores are associated with measures of body adiposity, which is valuable information for clinicians treating patients at risk for or with known CVD, and lend support to the American College of Cardiology's recent recommendation of RYP as a useful tool in gathering baseline dietary information.24 Further research to determine whether or not changes in RYP scores are associated with changes in CV risk‐factor control is warranted.

Conflicts of interest

The authors declare no potential conflicts of interest.

Ganguzza L, Ngai C, Flink L, et al. Association between diet quality and measures of body adiposity using the Rate Your Plate survey in patients presenting for coronary angiography. Clin Cardiol. 2018;41:126–130. 10.1002/clc.22843