Debunking the High Cost of Healthy Diets: Consumer Behavior Predicts Dietary Energy Density in a Nationally Representative Sample of US Adults

Abstract

Purpose:

Dietary energy density (ED; kcal/g) is an established marker for diet quality and a risk factor for obesity. Previous studies have suggested that low-ED diets cost more than high-ED diets, adding an economic contribution to the obesity epidemic. This study evaluated the relationship between consumer behavior (money spent on food) and dietary energy density in a nationally representative sample of US adults.

Design, Setting and Subjects:

Data from 10,622 adult participants in the 2013-2016 NHANES were used for this study. The NHANES is a large cross-sectional survey conducted by the CDC and NCHS.

Measures:

Consumer behavior was evaluated by examining total dollars spent on food, as well as dollars spent at various categories of food stores & restaurants. Dietary ED was calculated using multiple methods.

Analysis:

Multivariate regression models were then used to evaluate the relationship between consumer behavior, defined as money spent in four categories (groceries, take-out, dining out, other food purchases) and dietary energy density.

Results:

Low-ED diets did not cost more than high-ED diets overall, though low-ED diets contained more servings of fruits (1.6 vs 0.4), vegetables (2.2 vs 0.9) and fiber (21 vs 13g), and fewer added sugars (15 vs. 18 tsp), solid fats (28 vs 39g), all p’s < 0.01. Differences in spending patterns were identified. A positive linear trend between money spent on fast food/takeout and dietary energy density (p < 0.001) was observed. Additionally, individuals in the lowest quartile of ED spent more at grocery stores per person than individuals in the highest quartile of ED ($182 vs. $150 p = 0.04).

Conclusion:

Spending pattern and consumer choices are associated with dietary ED in this cross-sectional analysis of a nationally representative population sample. Identifying eating behaviors associated with diets high in energy density may inform future investigations that intervene on dietary habit for promotion of healthy eating and prevention of weight gain.

Keywords

NHANES diet cost diet quality energy density (kcal/g)

Purpose

Obesity continues to be a problem of epidemic proportions in the United States. According to current national surveillance data, over 60% of adults in the US are considered overweight or obese (body mass index [BMI] >25 and >30 kg/m², respectively) [1]. Obesity is a risk factor for chronic disease including several types of cancer, [2] cardiovascular disease, [3-5] and Type II diabetes [6]. Researchers have established dietary energy density (ED, kcal/g) as a risk factor for obesity and the aforementioned diseases.^1,2 and national and international public health organizations, including the Dietary Guidelines for Americans have made recommendations to consume a diet low in energy density as a strategy for prevention of obesity, cancer, and as a method of weight control.³

Several factors contribute to dietary energy density. Diets comprised of foods high in water content, such as fruits and vegetables, are low in energy density (ED) due to the large gram weights of foods with a low energy contribution. In contrast, diets high in fat have higher ED [11]_. Fast food typically includes low-cost, high-ED foods; consumption of fast foods, as well as food from “fast-casual” restaurants has been is associated with greater energy intake, higher energy density and lower diet quality.⁴ Well-documented research has shown that the frequency of meals eaten away from home are associated with greater energy intake, fat intake, sodium intake, and sugar intake.^5,6 Frequently consuming foods away from home has also been associated with lower fruit & vegetable intake, and poorer diet quality in a large population study.⁷

On the other side of the energy density spectrum, previous studies conducted in small groups in the US and France have raised concerns that high-quality, low-ED diets are associated with greater cost,^8
-10 however little research has been done to investigate whether these costs are associated with consumer intake or dietary energy density at the population level. Looking beyond energy density, a large body of evidence demonstrates that health disparities may be explained in part by food, with cost of food contributing to differences in dietary intake, subsequent diet quality, and related health conditions.^9,11,12 Studies have demonstrated that adults perceive the cost of low-energy-dense foods, such as fruits and vegetables to be higher in cost and identify cost as a barrier to healthier eating,¹² which contributes to continued disparities in food selection and subsequent dietary intake. Compounding the perception of that diets low in energy density are high cost, the literature also has demonstrated that nutrient-poor, high-calorie diets cost less than nutrient-rich diets,¹³ and when comparing dietary intake with food prices, high-quality diets cost more than low-quality diets.¹⁴ The cost of a healthy diet has a robust influence on the social determinants of health. Food is a basic human need, and lower-cost diets reduce the economic burden related to this essential need, however, diet quality complicates matters dramatically. Food can be consumed for both surviving and thriving; and many health disparities that exist are related to the quality of food and the impact of diet quality on overall health. Individuals with a limited food budget must carefully select items that can meet all of their health needs. Diets that are high in energy density cost less, but also provide less food volume, and may subsequently be less satisfying as well as related to greater health risks. Diets low in energy density provide greater food volume, and are related to higher diet quality, which impacts future health.

Given the current recommendations regarding dietary energy density, as well as previous studies examining the expense of a low-ED diet it is important to understand the economic feasibility of consuming a diet low in energy density at the population level. The objective of this study was to examine the economics of low-and high-ED diets and evaluate the relationship between consumer behavior (money spent on food) and dietary energy density in a nationally representative sample of US adults.

Methods

Design

The National Health and Nutrition Examination Survey (NHANES) is a large, cross-sectional survey conducted by the National Center for Health Statistics. NHANES and its related nutritional component What We Eat In America (WWEIA) is designed to monitor the health and nutritional status of non-institutionalized civilians in the US. Nationally representative survey and physical data are collected on a continual basis and released in two-year increments. Complete details regarding the NHANES sampling methodology, data collection, and interview process are available on the NHANES website; written consent was obtained from all subjects (http://www.cdc.gov/nchs/nhanes.htm).

Sample

The present study sample was composed of a nationally representative sample who participated in the 2013-2016 NHANES, which was the most recent survey with relevant eating behavior, spending, and household food security data. Individuals under age 18 (n = 6,545), as well as those with incomplete dietary, spending or demographic data (n = 2,979) were excluded, resulting a final n of 10,622. The NHANES study was approved by the National Center for Health Statistics (NCHS) Research Ethics Review Board (ERB), Protocol #2011-17.

Measures

Dietary Intake, Calculation of Dietary Energy Density & Food Component Intake

Adults who participated in NHANES provided dietary recall data including all foods and beverages during their visit to the mobile examination unit as part of the What We Eat in America Study.¹⁵ The dietary data is collected accounting for sample variability and intake day of week. Specific status codes were provided in the NHANES data set to indicate the quality, reliability, and completeness of the dietary data. Two days of dietary intake are collected during the NHANES; one day is collected in-person by trained interviewers using the automated multi-pass method of 24-hour recall (24HR), Assessment of portion size during the in-person interview are completed using standard measuring guides (i.e.: plates, cups, bowls, and portion-sized circular tools). A second day is completed 3-10 days later over the phone for a slightly smaller group of participants. The complex sampling strategy used for collecting dietary data allows for data from one day of 24HR to represent the mean of population’s usual intake. The present study includes only data collected during the in-person interview for maximum accuracy, data consistency and to maximize sample size. Though a second day of 24HR is also collected 3-10 days later via phone; the present study.

The USDA Food and Nutrient Database for Dietary Studies was used to process NHANES dietary data. While there are several methods to calculate dietary ED, there is no standardized method for selecting which foods and/or beverages to include in the calculation. Current literature has suggested that using a food-only method for calculating energy density may provide the most robust results, since beverages can contribute disproportionately to overall dietary ED due to their high gram weight and high water content^16,17 and can mask relationships between foods in the diet and markers of disease.¹⁷ In preliminary analyses, dietary ED was calculated two ways: using all foods and beverages, and using foods only. For each method, ED was calculated dividing the energy content (kcal) by weight of foods or beverages (g) consumed. USDA food codes were used to identify which items were foods, and which were beverages (e.g. differentiating between milk used in cereal vs. milk consumed as a beverage). During these initial analyses, the disproportionate contribution of beverages to overall ED was observed and further analyses were conducted using the food-only method, controlling for beverage ED; furthermore, no relationship between beverage energy and body weight status was observed. Overall dietary ED was calculated for each individual by totaling the food-only energy intake (kcal) and dividing by the total gram weight of foods consumed. In addition to utilization of the FNDDS, USDA Food Patterns Equivalents Database (FPED) were used to process dietary data.¹⁸ The FPED converts 7,000+ individually reported food and beverage items into 37 disaggregated USDA Food Patterns components (e.g.: added sugars, solid fats, total fruit, dark vegetables, etc.) allowing for comparison to recommendations of the Dietary Guidelines for Americans, as well Healthy People guidelines.

Consumer Behavior

Consumer behavior was assessed at the family level during an in-home interview as part of the NHANES Family Questionnaire as part of the Flexible Consumer Behavior Survey (FCBS) module using the Computer-Assisted Personal interview (CAPI) system. Participants responded questions pertaining to food expenditure regarding the dollar amount spent on food for the household during the past 30 days in one of the following categories: at supermarkets or grocery stores; money spent dining out (including money spent at cafeterias and vending machines at school and work); money spent on foods carried out or delivered, and money spent on foods obtained from all other sources. In each category, participants were given instructions to prevent counting of food expenditures in more than one category. For the purposes of the present analysis, the three categories are referred to as “Grocery spending,” “Money Spent Dining Out,” “Take-Out Spending” and “Other Spending” in the analysis. Since supermarkets frequently carry a variety of non-food items, participants were also asked about the amount of money they spent on non-food items at grocery stores or supermarkets. This amount was then deducted to accurately assess the money spent on food alone. In order to determine the amount spent per family member, the total amount for each category was divided by the number of family members.

Other Demographic Characteristics

During the NHANES, height and weight were measured by trained examiners using standardized protocols and calibrated equipment during the physical examination component of the study. BMI was calculated from these measurements of height (in cm) and weight (in kg). Data regarding smoking status (current, former, never-smoker), age, education, race/ethnicity, marital status, sex, household size, and physical activity (calculated as metabolic equivalent minutes, or MET/min) were also included in the NHANES and were used as covariates in various models. The NHANES also includes a questionnaire module assessing Food Security using U.S. Food Security Survey Module. Participant responses to either 18 or 10 questions (in households with and without children, respectively) are used to identify households into one of the following levels of food security: full food security, marginal food security, low food security, very low food security; these categorical levels of food security are provided by the NHANES dataset and are used in the present analysis.

Analysis

All data were analyzed using SAS version 9.4 (SAS Institute, Cary, NC). Specific survey procedures were used in the analysis to account for sample weights, unequal selection probability, and clustered design. Multivariate regression models were then used to evaluate the relationship between consumer behavior, defined as money spent in four categories (groceries, take-out, dining out, other food purchases) and dietary energy density. Total family food spending models were adjusted for age, sex, race/ethnicity, education, household size, smoking status, and socioeconomic status (assessed by Poverty:Income Ratio, PIR) with significance determined at P < 0.05; models assessing spending per person were not adjusted for household size. Dietary intake and food pattern intake models were adjusted for age, sex, race/ethnicity, education, smoking status, beverage ED, and physical activity with significance determined at P < 0.05. Models that did not include food security as an independent variable also were adjusted for household food security status. For outcomes with more than two categories (e.g., food security), test for trend was conducted by evaluating continuous ED data in a linear model. In order to accurately quantify food group servings, all data regarding mean servings of each food group has been energy adjusted.

Results

In this nationally representative sample of 10,622 adults >18y, roughly half the weighted sample were between 18 and 50 y, with the majority of participants identifying as identifying as non-Hispanic white. Demographic characteristics of the study sample are presented in Table 1.

Table 1.

Demographic Characteristics of a Nationally Representative Sample of Adults From the 2013-2016 NHANES.

	n	weighted percent
Sex
Female	5521	51.6
Male	5101	48.4
Educational Attainment
HS or less	2178	14.2
High School Grad/GED	2249	21.6
Some College or AA degree	3070	33.1
College Graduate or above	2562	31.0
Smoking Status
Non-Smoker	2374	56.6
Current Smoker	2014	43.4
Race/Ethnicity ^a
NH-White	4078	64.5
NH-Black	2202	11.4
Hispanic / Latino	2861	15.2
NH-Asian	1111	5.6
Other inc. multi-racial	370	3.3
	Mean	SE
Age in years at screening	46.9	0.4
Ratio of family income to poverty	2.9	0.1
Energy (kcal)	2123.4	14.1
Total number of people in the Household	3.1	0.04
Total number of people in the Family	2.9	0.04
Weekly physical activity (MET-min) ^b	406.1	4.5

Table 1 Legend: Sample characteristics are presented as n, as well as percentage adjusted for complex survey sample design and subject weights to represent a nationally representative population. ^a Race is self-reported by NHANES participants; NH indicates “Non-Hispanic,”^b MET-Min represent weekly metabolic equivalents of metabolic activity.

Dietary Intake by Energy Density Quartile

Consistent with earlier studies, we found dietary energy density to be inversely related to consumption of total vegetables, fruits, fiber, and protein foods (including meats, poultry, seafood, and legumes), and positively associated with intake of total fat, saturated fat, solid fats, added sugars, and grains (both total and refined) after adjusting for energy intake, as shown in Table 2. No differences in dairy consumption or dietary cholesterol intake were observed across quartile groups. Diets in the lowest quartile of energy density also contained greater variety, with a mean of approximately 16 foods reported per day compared with diets in the highest quartile of energy density, which had a mean of 13 foods reported per day. The observed dietary pattern (high veg, low fat) associated with low-ED diets is consistent with reports of other low-ED diets, as well as the recommendation of the US DGA.

Table 2.

Adjusted Mean Dietary Intake by Energy Density Quartile.

	Lowest Quartile Energy Density N = 2838		2nd Quartile Energy Density N = 2779			3rd Quartile Energy Density N = 2449			Highest Quartile Energy Density N = 2550
Nutrient Intake	Mean	SE	Mean	SE	Sig	Mean	SE	Sig	Mean	SE	Sig
Total energy (kcal)	1975	32	2117	35	***	2176	32	***	2134	49	**
Total protein (g)	88	2.3	85	1.1		83	1.1		71	1.6	***
Total carbohydrate (g)	265	2.7	250	2.6	**	250	2.6	***	250	2.6	***
Total fat (g)	74	0.8	80	0.9	***	85	0.8	***	91	1.1	***
Total saturated fat (g)	23	0.4	26	0.5	**	27	0.5	***	29	0.5	***
Dietary Cholesterol	311	14.9	317	11.6		308	9.8		280	9.9
Solid Fats (g)	28	0.7	32	0.9	**	34	0.9	***	39	0.8	***
Added sugars	15	0.5	15	0.6		15	0.6		18	0.7	***
Dietary fiber (g)	21	0.6	18	0.6	***	16	0.6	***	13	0.3	***
Food Group Intake
Total Number of Foods	16.2	0.3	16.0	0.3		14.9	0.2	**	13.5	0.2	***
Total Vegetable Servings (cup equiv.)	2.2	0.1	1.6	0.1	***	1.3	0.1	***	0.9	0.1	***
Total Fruit Servings (cup equiv.)	1.6	0.1	1.0	0.1	***	0.7	0.1	***	0.4	0.1	***
Total Grain Servings (oz. equiv)	6.1	0.3	6.6	0.2	*	6.8	0.3	***	6.9	0.2	***
Refined Grain Servings (oz. equiv)	5.1	0.2	5.6	0.2	*	6.0	0.2	***	6.2	0.2	***
Total Meat / Poultry / Seafood Servings (oz. equiv)	5.6	0.3	5.5	0.1		5.2	0.2		4.1	0.2	***
Total Vegetable Protein Servings (oz equiv)	0.6	0.09	0.6	0.05		0.6	0.05		0.2	0.1	***
Total Dairy Servings (cup equiv.)	1.3	0.1	1.3	0.1		1.2	0.1		1.2	0.1

Table 2 legend: Adjusted mean dietary intake by sex-specific quartile of energy density. Nutrient intakes are adjusted for age, sex, race/ethnicity, education, income, smoking status, physical activity. Food group intakes are also energy-adjusted. Statistical comparisons are made using the lowest ED quartile as the referent category. Superscripts indicate statistical difference between groups in the following manner: *p < 0.05, **p < 0.01, ***p < 0.0001.

Food Spending by Energy Density Quartile

Despite containing more vegetables, fruits, dairy, and protein foods, individuals with low-ED diets did not spend more on food each month per family than individuals with high-ED diets ($267 vs. $279 per person, p = 0.86), nor were any differences in spending observed between any quartiles of energy density, Figure 1. However, when separating out where spending was done, differences in spending patterns were noted, and are shown in Table 3. Those with the lowest-ED diets spent approximately $30 more per person at grocery stores (mean $181, 95% CI $148-215) compared with family spending for those in the highest quartile of energy density (mean $150; 95% CI $138-162), an amount that was statistically significant (p = 0.04). Additionally, individuals in the lowest quartile of energy density spent significantly less per family on dining out than individuals in the highest quartile ($116 vs. $149, p = 0.04). Finally, a significant inverse relationship was observed between dietary energy density and spending on takeout and carryout foods: individuals with high ED diets spent 22% more each month on take-out & fast food than those with low ED diets, and this trend was observed across quartiles. No differences in spending on foods obtained from other sources were observed between energy density categories.

Figure 1.

Mean total family spending ($USD) on food during the past 30 days. A, Average amount spent on food is calculated by totaling participants reports of amount spent on food for their family at grocery stores/supermarkets (subtracting reported amount spent on non-food items), amount spent on fast food/takeout, amount spent on dining out, and amount spent on food from all other sources. Mean amounts spent are adjusted for age, sex, race/ethnicity, educational attainment, income, smoking status and household size. Participants are categorized into sex-specific energy density quartiles for analysis. B, Mean total spending ($USD) per family member on food during the past 30 days. Average amount spent on food per family member is calculated by totaling participants reports of amount spent on food at grocery stores/supermarkets (subtracting reported amount spent on non-food items), amount spent on fast food/takeout, amount spent on dining out, and amount spent on food from all other sources divided by the total number of family members. Mean amounts spent are adjusted for age, sex, race/ethnicity, educational attainment, income and smoking status. Participants are categorized into sex-specific energy density quartiles for analysis.

Table 3.

Spending Patterns by Dietary Energy Density Quartile.

	Grocery Store Spending			Dining Out Spending			Take Out Spending			Spending on Food at Other Stores
	Mean $	SE	p-value	Mean $	SE	p-value	Mean $	SE	p-value	Mean $	SE	p-value
Spending Per Family^a
Lowest Quartile of Dietary ED	452.62	42.81	ref	115.65	19.85	ref	22.52	1.55	ref	106.97	6.25	ref
2nd Quartile of Dietary ED	422.83	27.38	0.29	138.53	13.55	0.06	26.16	1.76	0.02	115.76	7.53	0.32
3rd Quartile of Dietary ED	400.51	26.38	0.02	216.49	48.05	0.14	28.96	2.78	0.03	114.17	6.33	0.33
Highest Quartile of Dietary ED	385.36	24.75	0.004	149.03	9.45	0.04	27.96	2.03	0.01	112.10	6.07	0.44
Spending Per Person^b
Lowest Quartile of Dietary ED	181.67	16.35	ref	53.64	5.82	ref	9.42	0.96	ref	37.29	3.25	ref
2nd Quartile of Dietary ED	161.15	7.56	0.26	64.89	4.07	0.09	10.01	1.47	0.08	39.68	3.99	0.58
3rd Quartile of Dietary ED	172.84	9.01	0.61	71.29	5.53	0.03	12.44	1.98	0.048	40.78	3.43	0.34
Highest Quartile of Dietary ED	150.08	5.70	0.04	65.05	4.97	0.09	11.44	1.44	0.007	41.08	4.59	0.39

Table 3 Legend: Adjusted mean amount (in USD) spent on food at grocery stores/supermarkets (subtracting reported amount spent on non-food items), amount spent on dining out, amount spent on fast food/takeout, and amount spent on food from all other sources. Mean amounts spent are adjusted for household size, age, sex, race, education, income, smoking status, and physical activity level. Participants are categorized into sex-specific energy density quartiles for analysis.

^a Mean spending per family is the total amount reported spending over the past 30 days.

^b Mean spending per individual is total amount reported spending over the past 30 days divided by the number of people in the family.

Dietary Energy Density by Spending Category

In order to evaluate the relationship between food spending and diet quality, energy density was calculated for each quartile of food spending and compared between quartiles. The mean dietary energy density was not significantly different between those who spent the most or the least on food each month (1.43 vs. 1.31, p = 0.91). Though we did not observe a difference in dietary ED among spending categories, the average monthly spending on food was still quite high indicating that the quartiles may not address the true differences in diet costs / spending ability. In order to address this, the relationship between dietary energy density and household food security level was calculated and presented in Table 4. Food security status was not associated with dietary energy density; individuals who were identified as “food secure” did not have a lower dietary ED than individuals who were identified as “very low food security” (1.81 vs. 1.86, p = 0.22), nor did they differ in spending on takeout / fast food each month per family ($21 vs. $22, p = 0.93).

Table 4.

Dietary Energy Density by Food Security Status.

Household Food Security Category	Mean Dietary ED	SE
Food Secure	1.81	0.02
Marginal Food Security	1.86	0.02
Low Food Security	1.92	0.05
Very Low Food Security	1.86	0.04

Table 4 legend: Food Security Status was assessed during the NHANES using U.S. Food Security Survey Module. Participant responses to either 18 or 10 questions (in households with and without children, respectively) are used to identify households into one of four levels of food security. Mean dietary ED is adjusted for age, sex, race, smoking status, education level, and physical activity level.

Discussion

Contrary to previous reports indicating that diets low in ED cost more than high-ED diets^10,19 this nationally representative data suggests that individuals consuming low-ED diets have equivalent spending on food overall compared to individuals consuming diets high in energy density. The present study supports prior findings that low-ED diets are higher in diet quality,^8,9,20 as the low-ED diets in this nationally representative sample had the greatest amounts of total servings of vegetables, fruits, fiber, protein foods, and dairy, while simultaneously having the lowest amounts of total, saturated, and solid fats, added sugars and refined grains. Additionally, the low-ED diets observed in the present study were also higher in food variety.²¹ These findings have important implications for nutrition education and translation of nutrition messaging for public consumption. Most importantly, low-ED diets can be obtained regardless of amount spent on food, socioeconomic status, or even food security status, and messaging regarding low-cost strategies to incorporate low-ED foods into the diet may be a potentially successful strategy for obesity prevention at all income levels.

The current findings stand in direct contrast to a previous review which found that individuals of lower SES were more likely to select high-ED, low-cost foods for consumption,⁹ as well as a more recent study that indicates that food spending may determine purchase of healthy foods.¹⁷ The review demonstrates the contribution of food cost to related health disparities, and concludes that nutrition scientists must identify food patterns associated with lower cost and higher diet quality in order to prevent health disparities. The present analysis seeks to add insight to a potential solution. In the present analysis, cost associated with high- and low-ED diets were related to food spending behaviors, indicating that a low-ED diet can be obtained regardless of food spending.

The comparable costs of a low-ED diet versus a high-ED diet is echoed in a study completed in Scotland where price of healthy foods were not perceived to be a barrier for healthy eating.²² Individuals consuming a low-ED diet did not report higher overall spending on food each month; low-ED consumers spent the most each month on food at grocery stores and supermarkets, and the least on fast food & takeout and dining out. Studies evaluating the presence of foods within a specific contexts (i.e., the food microenvironment) have identified the mere presence of fresh fruit is associated with lower body weight. Therefore, increasing the availability of low-ED food for purchase at a variety of stores, for example: stocking frozen vegetables at a neighborhood market, may also serve as the foundation for public health strategies to educate consumers about methods for reducing the risk of obesity without a high associated cost. Highlighting methods of improving diet quality without increasing diet cost is a critical hurdle to create targeted strategies aimed at vulnerable populations that are more likely to rely on a high-ED diet, such as fast food and takeout . One implication of such a finding, however, is that although spending is not related to diet health, accessibility of low-ED options likely influences food choice. Limited access to healthy foods and lack of grocery stores are well-established risk factors for obesity and poor diet quality.^23,24 As such, the findings from the present study indicate that low-ED diets are possible regardless of money spent, the caveat remains that intake of healthy foods is dependent on the opportunity of individuals to obtain them. With that said, it is of equal importance to note that individuals with the highest ED diets did not spend any less overall, and that only the increase of spending within the fast food/takeout category was associated with increased dietary ED. This finding dispels the common myth that increased intake of high-ED convenience foods is due to lower cost, though availability may still be an issue.

There are several strengths to the present analysis. First, the results are generalizable to the US population and the dietary data are collected under tightly controlled conditions. In contrast to previous studies, this study used a novel approach to evaluate the association between money spent on food and dietary energy density. This allows for a simplified public health message—it’s possible to eat healthy at any price point or food security level. There are also several possible limitations to this research. The nutritional data within the NHANES study are self-reported and may be subject to recall bias; additionally the study uses one day of dietary recall data, which represents the mean of the population’s usual intake—however, a single day of recall may not be good representations of each individual’s usual intake or dietary pattern. The data collection methods for the 24-hour diet recalls are state-of-the art with quality control procedures in place during the data collection phase and within our analyses (e.g., excluding implausible data) help to address this potential concern. Finally, the cross-sectional survey design of NHANES allows for evaluation of population-wide associations but prevents evaluation of causality. Additionally, as mentioned earlier, though these data demonstrate that low-ED diets can be obtained regardless of money spent on food, the present analysis does not take into consideration the local food environment of participants, or availability of low-ED foods.

Given the established relationship between dietary energy density and body weight status^1,16 as well as diet variety and BMI²¹ it is important to educate the public regarding low-cost strategies to consume a low-ED diet. Strategies to educate consumers regarding low-cost “healthy” food options and dispelling misconceptions that high-ED diets are “cheaper” may be a successful method of slowing the obesity epidemic.

So What?

What is already known on this topic?

Dietary energy density (ED, kcal/g) is marker for diet quality; a diet high in energy density is an established risk factor for obesity, and has been shown to cost less than a diet low in energy density.

Low ED diets contain more fruits and vegetables, and fewer refined grains; some studies have suggested that this type of diet costs more than a high-ED diet.

Health disparities related to dietary intake may be related to dietary cost, particularly in spending per person.

What does this article add?

Our findings indicate that despite common misconceptions, individuals with low-ED diets do not spend significantly more on food than individuals with high-ED diets. Instead, individuals with low-ED diets spend more at the grocery store whereas individuals with high-ED diets spend more on take-out, fast food, and on dining out.

What are the implications for health promotion practice or research?

Strategies to educate consumers regarding low-cost “healthy” food options and dispelling misconceptions that high-ED diets are “cheaper” may be a successful method of slowing the obesity epidemic.

Nutrition education related to the high “health cost” of fast foods may help to decrease diet-related health disparities.

Footnotes

Acknowledgments

JAV designed the study, contributed to drafting the manuscript, data analysis, and data interpretation. RD contributed to drafting the manuscript, data analysis, and critical input. Both authors approve of the final manuscript.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Jacqueline A. Vernarelli

References

Vernarelli

Mitchell

Rolls

Hartman

. Dietary energy density is associated with obesity and other biomarkers of chronic disease in US adults. Eur J Nutr. 2015;54(1):59–65.

Perez-Escamilla

Obbagy

Altman

, et al. Dietary energy density and body weight in adults and children: a systematic review. J Acad Nutr Diet. 2012;112(5):671–684.

Millen

Abrams

Adams-Campbell

, et al. The 2015 dietary guidelines advisory committee scientific report: development and major conclusions. Adv Nutr. 2016;7(3):438–444.

Bowman

Vinyard

. Fast food consumption of U.S. adults: impact on energy and nutrient intakes and overweight status. J Am Coll Nutr. 2004;23(2):163–168.

Kant

Whitley

Graubard

. 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

Mancino

Todd

Guthrie

Biing-Hwan

. How Food Away From Home Affects Children’s Diet Quality. U.S. Dept. of Agriculture, Econ. Res. Serv.; 2011 October, 2010. Report No.: EER-104 Contract No.: EER-104.

Seguin

Aggarwal

Vermeylen

Drewnowski

. Consumption frequency of foods away from home linked with higher body mass index and lower fruit and vegetable intake among adults: a cross-sectional study. J Environ Public Health. 2016;2016:3074241.

Drewnowski

Monsivais

Maillot

Darmon

. Low-energy-density diets are associated with higher diet quality and higher diet costs in French adults. J Am Diet Assoc. 2007;107(6):1028–1032.

Darmon

Drewnowski

. Contribution of food prices and diet cost to socioeconomic disparities in diet quality and health: a systematic review and analysis. Nutr Rev. 2015;73(10):643–660.

10.

Monsivais

Drewnowski

. Lower-energy-density diets are associated with higher monetary costs per kilocalorie and are consumed by women of higher socioeconomic status. J Am Diet Assoc. 2009;109(5):814–822.

11.

Hardin-Fanning

Rayens

. Food cost disparities in rural communities. Health Promot Pract. 2015;16(3):383–391.

12.

Chapman

Goldsbury

Watson

, et al. Exploring perceptions and beliefs about the cost of fruit and vegetables and whether they are barriers to higher consumption. Appetite. 2017;113:310–319.

13.

Andrieu

Darmon

Drewnowski

. Low-cost diets: more energy, fewer nutrients. Eur J Clin Nutr. 2006;60(3):434–436.

14.

Rehm

Monsivais

Drewnowski

. The quality and monetary value of diets consumed by adults in the United States. Am J Clin Nutr. 2011;94(5):1333–1339.

15.

U.S. Department of Agriculture ARS, Beltsville Human Nutrition Research Center, Food Surveys Research Group, U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics. What we eat in America NHANES 2013-2014. 2011 U.S. Department of Agriculture ARS, Beltsville Human Nutrition Research Center, Food Surveys Research Group (Beltsville, MD) and U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics.

16.

Vernarelli

Mitchell

Hartman

Rolls

. Dietary energy density is associated with body weight status and vegetable intake in U.S. children. J Nutr. 2011;141(12):2204–2210.

17.

Johnson

Wilks

Lindroos

Jebb

. Reflections from a systematic review of dietary energy density and weight gain: is the inclusion of drinks valid? Obes Rev. 2009;10(6):681–692.

18.

Bowman

Clemens

Friday

Lynch

Moshfegh

. Food patterns equivalents database 2013-14: methodology and user guide [Online]. Accessed September 10, 2019. 2013. http://www.ars.usda.gov/ba/bhnrc/fsrg

19.

Pechey

Monsivais

. Socioeconomic inequalities in the healthiness of food choices: exploring the contributions of food expenditures. Prev Med. 2016;88:203–209.

20.

Ledikwe

Blanck

Khan

, et al. Low-energy-density diets are associated with high diet quality in adults in the United States. J Am Diet Assoc. 2006;106(8):1172–1180.

21.

Roberts

Hajduk

Howarth

Russell

McCrory

. Dietary variety predicts low body mass index and inadequate macronutrient and micronutrient intakes in community-dwelling older adults. J Gerontol A Biol Sci Med Sci. 2005;60(5):613–621.

22.

Mc Morrow

Ludbrook

Macdiarmid

Olajide

. Perceived barriers towards healthy eating and their association with fruit and vegetable consumption. J Public Health (Oxf). 2017;39(2):330–338.

23.

Larson

Story

Nelson

. Neighborhood environments: disparities in access to healthy foods in the U.S. Am J Prev Med. 2009;36(1):74–81.

24.

Hilmers

Dave

. Neighborhood disparities in access to healthy foods and their effects on environmental justice. Am J Public Health. 2012;102(9):1644–1654.