Association of raisin and raisin‐containing food consumption with nutrient intake and diet quality in US children: NHANES 2001‐2012

Abstract Background Raisins are a commonly consumed dried fruit and given their nutrient profile may offer nutritional and health benefits. Objective To examine the association between consumption of raisins and raisin‐containing foods with nutrient intake and dietary quality in children. Methods National Health and Nutrition Examination Survey (NHANES) data for 2001–2012 in those 2–18 years of age (n = 20,175) were used. Consumers of raisins (n = 154, 51.6% female) and raisin‐containing foods (n = 1,993, 52.5% female) were defined as reporting any consumption of raisins and raisin‐containing foods, respectively, during the first 24‐hr diet recall. Diet quality was assessed using the Healthy Eating Index (HEI)‐2010. Regression analyses were conducted comparing consumers and nonconsumers using appropriate sample weights and adjusted for demographic and lifestyle covariates with significance set at p < 0.01. Results Regarding “nutrients of public health concern/shortfall nutrients” and “nutrients to limit,” raisin consumers had higher intakes of dietary fiber (23%), potassium (16%), magnesium (12%) with lower intakes of added sugars (−19%) than nonconsumers. Similarly, consumers of raisin‐containing foods also had higher intakes of dietary fiber (15%), potassium (5%), magnesium (11%), iron (6%), vitamin A (10%), and vitamin E (13%) and lower intake of sodium (−5%). Consumers of raisin and raisin‐containing foods had higher intakes of fruits (60%, 16%, respectively), whole fruits (119%, 23%, respectively) and whole grains (44%, 93%, respectively) and had a better diet quality as per higher total HEI 2010 scores (22%, 8%, respectively) than nonconsumers. Conclusion In conclusion, consumption of raisins or raisin‐containing foods was associated with better nutrient intake and diet quality in American children.


Increased intake of fruit and vegetables is an unequivocal recommendation by Dietary Guidelines worldwide. The Dietary
Guidelines for Americans 2015-2020 (DGA) recommended adequate consumption of fruits and vegetables as part of healthy eating pattern [DHHS/USDA, 2015]. Dietary recommendations from the American Academy of Pediatrics for children and adolescents also include eating more fruits and vegetables and reducing added sugar [Gidding et al., 2006]. MyPlate (ChooseMyPlate.gov) recommends one to two cup equivalents of fruit per day for children depending on age, gender, and physical activity [USDA, 2010].
Despite these recommendations, about 25% of 1-3-year-old toddlers, about 40% of 4-8-year-old young children, 76% of 9-13-year-old children, and over 85% 14-18-year-old teens in the United States do not consume amount of fruit recommended [NCI, 2014a]. Fruits are naturally low in fat and sodium and can be important sources of potassium, dietary fiber, vitamin C, and folate.
Raisins (dried grapes) are a commonly consumed dried fruit [Keast, O'Neil, & Jones, 2011]. Raisins provide dietary fiber and important minerals and are low-medium energy dense. A serving (43 g) of raisins provides 129 calories with 1.6 g dietary fiber, 0.2 g total fat, 25 g total sugar, 14 mg magnesium, 322 mg potassium, and 0.8 mg iron [USDA, 2016]. Raisins also contain various phytochemicals including flavonoids (catechins, quercetin, kaempferol, and rutin), hydrocinnamic acids (coutaric and cafterics), epicatechins, phytoestrogens (genestein and daidzein), and resveratrol [Karadeniz, Durst, & Wrolstad, 2000;Williamson & Carughi, 2010], but the physiological consequences of consumption of these compounds remain to be elucidated. However, intervention studies examining nutritional and health effect of raisin consumption have been mostly limited to adults [Williamson & Carughi, 2010;Anderson & Waters, 2013;Kanellos et al., 2014;Anderson, Weiter, Christian, Ritchey, & Bays, 2014;Bays, Weiter, & Anderson, 2015;Kanellos et al., 2013;Esfahani, Lam, & Kendall, 2014;]. Satiating effects of raisins have been reported in a couple of studies with children Patel, Luhovyy, Mollard, Painter, & Anderson, 2013]. Using NHANES data, we have recently reported that raisin consumption in adults was associated with better diet quality, and better nutrient intake, with lower risks of having metabolic syndrome and being obese and (Fulgoni, Painter, & Carughi, 2017). Epidemiological studies evaluating the effect of raisins on diet quality and nutrient intake in children are very limited. In one cross-sectional study in children aged 2-19 years, the intake of grapes and grape products (juice and raisins) was associated with a healthier overall dietary pattern and higher intake of certain nutrients [McGill, Keast, Painter, Romano, & Wightman, 2013]. The objective of this study was to evaluate the association of raisin or raisin-containing foods consumption with nutrient intakes and diet quality in a large, nationally representative sample of US children.

| Subjects
Data from children age 2-18 years participating in the National Health and Nutrition Examination Survey (NHANES) 2001-2002, 2003-2004-2006, 2007, 2009[CDC, 2015 were used for these analyses to improve sample size. Exclusion criteria included those with incomplete or unreliable 24-hr recall data and those pregnant and/or lactating. Written informed consent was obtained for all participants or proxies (i.e., parents or guardians for children under 6 years; those 6-11 years had assistance from parents or guardians), and the survey protocol was approved by the Research Ethics Review Board at the NCHS.  (Fulgoni et al., 2017), intake of raisins were assessed using only two USDA food codes [USDA 2015a]: plain raisins (USDA food code 62125100) and cooked raisins (USDA food code 62125110). Intake of raisin-containing foods was assessed using 147 USDA food codes that contained raisins as an ingredient. A list of top 40 foods containing raisins and their food codes is provided in Table 1. Consumers of raisins (n = 154) were defined as those consuming raisins in any amount during the 24-hr recall.

| Estimation of intake
Consumers of raisin-containing foods (n = 2,059) were defined as those consuming any quantity of foods that contained raisins during the 24-hr recall. Energy and nutrient intake were determined similar to that of Fulgoni et al., 2017 [Bowman, Friday, & Moshfegh, 2008;USDA, 2015b] which provides servings of major food groups and corresponding subgroups. The MyPlate food group intakes per day were calculated by aggregating the number of servings for all the foods reported in the 24-hr recall.

| Estimation of diet quality
USDA's Healthy Eating Index (HEI)-2010 was used to estimate diet quality. HEI-2010 has 12 components, and each component represents a particular aspect of diet quality .
Dietary intake data (day 1) and the SAS code downloaded from the USDA website [NCI, 2014b] were used to determine HEI-2010 scores.

| Statistical analysis
SAS 9.2 (SAS Institute, Cary, NC) and SUDAAN 11 (RTI, Research Triangle Park, NC) were used for all analysis. The complex sample design of NHANES was considered using primary sampling units and strata with appropriate survey weights (day one dietary sample weights) in all intake analyses. Similar to our approach used previously (Fulgoni et al., 2017), least square means (LSMs) and standard errors (SEs) were generated via regression analyses for intakes of energy, nutrients, and food groups and diet quality in consumers and nonconsumers of raisins and raisin-containing foods. Analyses of intakes of food groups and nutrient intakes were adjusted for age, ethnicity, gender, poverty income ratio, physical activity level, current smoking status, and energy intake. Analyses of energy intake and diet quality were adjusted for the same covariates but without energy intake; HEI scores are

| RE SULTS
Approximately, 0.92% children (n = 154, 51.6% females) were consumers of raisins and 10.3% children (n = 1,993, 52.5% females) were consumers of raisin-containing foods. Consumers of raisins as well as of raisin-containing foods were younger than nonconsumers. Consumers of raisins were less likely to be Mexican American and smokers while the consumers of raisin-containing foods were more likely to be non-Hispanic White and less likely to be of other ethnicity (Table 2).
Nutrient intakes differed significantly between the consumers of raisins and nonconsumers as well as between consumers and nonconsumers of raisin-containing foods (  Table 3).

| D ISCUSS I ON
This is the first report to explore relationships of consumption of raisins and raisin-containing foods with diet quality and nutrient intakes among US children using a large nationally representative sample.
We  have higher intake of several key nutrients including dietary fiber, vitamin A, vitamin B6, vitamin C, calcium, potassium, and magnesium compared to nonconsumers. DGA has identified most of these nutrients (fiber, potassium, magnesium, and vitamin E) as-"shortfall nutrients" as these are currently underconsumed [DHHS/USDA, 2015]. Additionally, DGA has termed fiber and potassium as "nutrients of public health concern" as current The consumers of raisins and also of raisin-containing foods also had higher intakes of fruits, whole fruit, and whole grain. The higher intake of these food groups is associated with healthier diets. HEI-2010 is a validated measure of diet quality and is indicative of compliance with dietary recommendations. It has 12 components (nine components address adequacy intake and three address moderation of intake) each of which relate to a specific DGA recommendation [DHHS/USDA, 2011]. HEI has been previously used to assess whether diet quality changes over time [Juan, Guenther, & Kott, 2008], the effectiveness of dietary interventions, and to validate research tools and nutrition indices [Fulgoni, Keast, & Drewnowski, 2009]. It has also been effectively used to evaluate diets of population subgroups [Hiza, Casavale, Guenther, & Davis, 2013] and food environments [Reedy, Krebs-Smith, & Bosire, 2010]. Several recent studies have also used HEI to assess relationships between intakes of nutrients, specific foods, and dietary patterns with health-related outcomes [Fulgoni et al., 2017;Nicklas, O'Neil, & Fulgoni, 2012;Chiuve et al., 2012;Reedy et al., 2008;O'Neil, Nicklas, Rampersaud, & Fulgoni, 2011].
Total and many subcomponent HEI-2010 scores of consumers of raisins and raisin-containing foods were significantly higher than those of nonconsumers. Consumers of grapes and grape products (juice and raisins) were reported to have higher diet quality as compared to nonconsumers [McGill et al., 2013]. In adults, we previously found that HEI-2010 total scores and scores of many subcomponents were higher in raisin consumers than nonconsumers [Fulgoni et al., 2017].
One limitation of this study is the use of cross-sectional data, which cannot confer casualty. Additionally, given 24-hr dietary recalls rely on participants' memory to self-report dietary intakes, they are subject to misreporting or reporting bias. Data used in this study were based on a single 24-hr dietary recall and for some participants relied on intakes reported by parents/caregivers. Major strengths of this study included the use of a nationally representative large sample and the use of numerous covariates to adjust data in an attempt to remove potential confounding. However, residual confounding may still exist that could explain some of the results reported.
In conclusion, the results from this study suggest that consumption of raisins/raisin-containing foods is associated with better diet quality and nutrient intake in American children.

CO N FLI C T O F I NTE R E S T
JP consults with Sun-Maid Raisin Growers of California, the National Pasteurized Egg Board, Paramount Pistachios, and the National Dairy Council. AC works for Sun-Maid Raisin Growers of California.
VLF provides analyses of NHANES for members of the food/beverage industry. Informed Consent: Written informed consent was provided by all participants or proxies (i.e., parents or guardians).

R E S P O N S I B I LITI E S
VLF and AC designed of the study. VLF conducted analysis; VLF, JP, and AC reviewed results. VLF developed the first draft of the manuscript; JP and AC provided edits. All authors read the final version.