HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Google Scholar Articles by Fisher, J. O Articles by Deming, D. PubMed PubMed Citation Articles by Fisher, J. O Articles by Deming, D. Agricola Articles by Fisher, J. O Articles by Deming, D.

Overestimation of infant and toddler energy intake by 24-h recall compared with weighed food records^1,2,3

¹ From the Department of Pediatrics, Baylor College of Medicine, USDA/ARS Children's Nutrition Research Center, Houston, TX (JOF, NFB, PMM, and TAW), and the School of Nursing, University of North Carolina, Chapel Hill, NC (EAH)

² Supported by the US Department of Agriculture Current Research Information System and the Gerber Products Company.

³ Reprints not available. Address correspondence to JO Fisher, USDA Children's Nutrition Research Center, 1100 Bates Street, Suite 4004, Houston TX 70030. E-mail: jfisher{at}bcm.tmc.edu.

	ABSTRACT

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Background: Twenty-four-hour dietary recalls have been used in large surveys of infant and toddler energy intake, but the accuracy of the method for young children is not well documented.

Objective: We aimed to determine the accuracy of infant and toddler energy intakes by a single, telephone-administered, multiple-pass 24-h recall as compared with 3-d weighed food records.

Design: A within-subjects design was used in which a 24-h recall and 3-d weighed food records were completed within 2 wk by 157 mothers (56 non-Hispanic white, 51 non-Hispanic black, and 50 Hispanic) of 7–11-mo-old infants or 12–24-mo-old toddlers. Child and caregiver anthropometrics, child eating patterns, and caregiver demographics and social desirability were evaluated as correlates of reporting bias.

Results: Intakes based on 3-d weighed food records were within 5% of estimated energy requirements. Compared with the 3-d weighed food records, the 24-h recall overestimated energy intake by 13% among infants (740 ± 154 and 833 ± 255 kcal, respectively) and by 29% among toddlers (885 ± 197 and 1140 ± 299 kcal, respectively). Eating patterns (ie, frequency and location) did not differ appreciably between methods. Macronutrient and micronutrient intakes were higher by 24-h recall than by 3-d weighed food record. Dairy and grains contributed the most energy to the diet and accounted for 74% and 54% of the overestimation seen in infants and toddlers, respectively. Greater overestimation was associated with a greater number of food items reported by the caregiver and lower child weight-for-length z scores.

Conclusions: The use of a single, telephone-administered, multiple-pass 24-h recall may significantly overestimate infant or toddler energy and nutrient intakes because of portion size estimation errors.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
During the first 2 y of life, infants move from frequent consumption of a single food to more adult-like meal and snack patterns (1) involving increasing numbers of foods of various forms [ie, liquid, solid (eg, apple), or solid-amorphous (eg, applesauce)] and nutrient compositions. Given its ease of use, low respondent burden, and cost-effectiveness, the 24-h dietary recall (24-h DR) is the method of choice for capturing these patterns, particularly when population-based research is concerned (2). Current estimates of energy intake (EI) among US infants and young children involve in-person and telephone-administered 24-h recalls (3, 4). The Feeding Infants and Toddlers Study (FITS) used a single, telephone-administered 24-h recall to measure the dietary intake of a nationally random sample of 3022 children aged 4–24 mo, and a second 24-h recall was collected on a random subsample (n = 703) to estimate the usual intake distribution (4). Mean EIs were 10%, 23%, and 32% higher than estimated energy requirements (EERs) among children aged 4–6, 7–11, and 12–24 mo, respectively (5). The observed discrepancies are corroborated by 24-h recall data for Special Supplemental Food Program for Women, Infants, and Children (WIC) respondents aged 0 mo–4 y who were participating in the Continuing Survey of Food Intakes by Individuals (CSFIIs from 1994–1996 and 1998) (6, 7). These data have raised concerns (4) about current EIs in young children, particularly given increases of 20% in the prevalence of overweight among children <2 y old since the 1980s (8). The degree to which apparent excesses in infant and toddler EIs reflect errors in dietary assessment, estimation of energy needs, or both, however, is unclear.

Dietary recalls approximate food records in estimating the average EIs of preschool-age children (9-11). The accuracy of the method in younger children is not well documented, and limited evidence comes from research conducted in developing countries (12). Obtaining accurate data on the intake of infants and toddlers may be complicated by the fact that persons other than the parents may be responsible for child feeding (13, 14). Other challenges to accurate assessment in infants and toddlers include frequent feeding (1) and food spillage and wastage, which may be considerable, particularly during the transition to self-feeding (5).

By following the FITS methodology (4), this research evaluated the accuracy of mean infant and toddler EIs by a single, telephone-administered, multiple-pass 24-h DR based on the reference method of 3-d weighed food records (3-d WR). Mean EIs of 1.5–4.5-y-old children based on weighed records has been validated against doubly labeled water measurements of total energy expenditure (TEE), an estimate of usual intake when energy costs of growth are minimal (15, 16). A secondary objective was to identify demographic, anthropometric, and psychosocial correlates of reporting error.

	SUBJECTS AND METHODS

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Design
A within-subjects design was used to evaluate the accuracy of reported EIs in Hispanic, non-Hispanic white, and non-Hispanic black 7–11-mo-old infants and 12–24-mo-old toddlers on the basis of a single, telephone-administered, multiple-pass, weekday 24-h DR completed by the primary caregiver. The reference method was weighed food records, made on 3 consecutive weekdays within 2 wk of the dietary recall. Infant or toddler weight-for-length z scores, maternal body mass index (BMI; in kg/m²) (17), and self-reported household income, maternal education (10), ethnicity, employment status, social desirability (17), feeding method, and childcare participation (10) were evaluated as potential correlates of reporting error by the 24-h DR.

Subjects
The study population was Hispanic, non-Hispanic white, and non-Hispanic black mothers of 7–11-mo-old infants or 12–24-mo-old toddlers who were recruited in equal numbers. Maternal participation criteria were self-identification in 1 of the 3 race-ethnicity groups and having primary responsibility for infant feeding at home. Infant inclusion criteria included study age range at the time of enrollment, term birth (37–42 wk), and weight-for-age of >5th percentile at birth (18). Infant exclusion criteria were feeding problems and chronic medical conditions or medication use.

Convenience sampling was employed to recruit mothers from the greater metropolitan Houston area by using a volunteer database maintained at the US Department of Agriculture (USDA) Children's Nutrition Research Center, flyers, and on-site recruiting at childcare classes, city-wide festivals or expos attracting families, doctors’ offices, retail stores, and churches. One child was surveyed per household.

Each mothers provided written informed consent for her own participation and that of her child. All procedures were approved by the Baylor College of Medicine Institutional Review Board.

Procedures
Potential participants were screened for inclusion by a telephone interview. Shortly after recruitment, caregivers were contacted by telephone on a scheduled weekday to recall infant dietary intake for a 24-h period. When the infant participated in childcare, caregivers were asked to contact the staff to obtain detailed information on the child's intake while in childcare. Within 1 wk of the 24-h DR, the caregiver and infant made an initial visit to the USDA Agricultural Research Service Children's Nutrition Research Center (CNRC; Houston, TX), during which initial instructions on keeping weighed records were provided. At this visit, child and caregiver anthropometric measurements were obtained and self-reported questionnaires on demographics, social desirability, and infant feeding were completed by the caregiver. During the following week, a research staff member visited the home to provide food, infant scales, and instruction on keeping weighed records of infant or toddler dietary intake. Three scheduled days of weighed food records commenced within 2 wk of the 24-h DR.

Measures
The 24-h recall
A single, telephone-administered, multiple-pass 24-h DR was used to assess infant or toddler food intake. The recall was performed by a trained registered dietitian on a scheduled weekday 1 wk after the initial recruitment of the caregiver-infant pair. Before the recall, caregivers were mailed 2-dimensional portion size aids, which were previously used in FITS (4), to facilitate reporting of the quantity consumed. Dietary intake data were collected and analyzed by using NUTRITION DATA SYSTEM for RESEARCH (NDS) software (version 2005; Nutrition Coordinating Center, University of Minnesota, Minneapolis, MN).

Assumptions used in FITS were applied to estimate the quantity of infant or toddler breastmilk intake (4). For children aged 7–11 mo who received breastmilk as the only milk source, a breastmilk intake of 600 mL was assumed; reported intake of formula, cow milk, or both (in g) was subtracted from 600 mL for children consuming multiple types of milks. For children 12 mo old, breastmilk intake was imputed according to the number of minutes of reporting: 1 fluid ounce (29.6 mL) was assumed for every 5 min of feeding. Caregivers were asked to obtain as much detail as possible from other persons (ie, childcare providers) who fed the child on the day of assessment. In a few cases, the research dietitian called the other caregivers to complete the child's recall. Recalls were reviewed for missing foods, unrealistic quantities, and supplement use. Recipes were added as records to the database. Brand names and nutrient information for commercially available foods not in the NDS database were also collected and added. Meal type and eating location were defined by the caregiver for each food reported.

Weighed food records
Weighed food records were made on 3 scheduled consecutive weekdays during a 2-wk period after the 24-h DR. The test-weighing method (19) was used to estimate breastmilk intake; caregivers were trained to use infant scales to weigh the infant to the nearest 0.1 g before and after each breastfeeding without changing the infant's clothing or diapers. A trained observer made a home visit to calibrate the portable electronic food scales [model no. SP601; Ohaus, Pine Brook, NJ (readability: 0.01 g)] and electronic infant scales [3862 MP 6; Sartorius, Goettingen, Germany (readability: 0.1 g)] and to provide detailed instruction on weighing and record completion. Food and recipe examples were given according to the child's typical diet. Caregivers were given written instructions and shown the procedures for measuring and recording foods consumed outside the home. Age-appropriate containers and cups were provided to each participant to facilitate the measurement of foods eaten away from home. The mother and, in a few instances, the research dietitian provided instruction to other caregivers (ie, childcare workers) who were responsible for feeding the child. When a set menu was to be used, the research dietitian assisted the process by indicating the food items that were to be weighed as individual, combined, or recipe foods. Meal type and eating location were defined by the caregiver for each food reported.

Infant or toddler anthropometrics
Anthropometrics included infant head circumference, length and weight measured in triplicate by trained research staff. Electronic scales [3862 MP 6; Sartorius (readability: 0.1 g)] were used to measure weight and infant length boards (Holtain Limited, Crymych, United Kingdom) were used to measure length. The 2000 growth charts of the Centers for Disease Control and Prevention were used to calculate weight-for-age, length-for-age, and weight-for-length z scores (18).

Maternal body mass index
Maternal height (Harpenden tadiometer; Holtain Limited) and weight [Doctor Scale 431/432KL series (capacity: 400 x 0.25 lb or 181.4 x 0.1 kg); Health-O-Meter, Bridgeview IL] were measured in duplicate by trained research staff according to the methods of Lohman et al (20); repeat measurements were performed in cases of nonagreement (within 0.5 kg or 0.5 cm). BMI scores were calculated.

Demographics
Demographic information on household income, maternal education, maternal employment, and childcare arrangements was obtained.

Maternal social desirability
The Marlowe-Crowne Social Desirability Scale was used to assess the extent to which bias in reporting child intake reflected socially desirable responding (21). Social desirability refers to the goal of presenting oneself in a positive or socially desirable manner. The Marlowe-Crowne scale consists of 33 items—18 representing positive behaviors that most people are not likely to exhibit and 15 representing negative behaviors that are likely to be true of the general population. The total score for social desirability is created by summing all items; higher scores indicate greater social desirability. The scale has high test-retest reliability over a 2-wk period in adults, and it has been positively associated with other social desirability measures in college-age samples (21). The internal consistency of the 33-item scale in the present sample was = 0.81.

Statistical analysis
Statistical analyses were performed with SAS software (version 9.1; SAS Inc, Cary, NC). Descriptive statistics including the mean, median, SD, and range were generated. Data are presented as means ± SDs. P < 0.05 was chosen to indicate statistical significance. Repeated-measure analysis of variance (ANOVA) was used to analyze the data; intake by each method was treated as the repeated factor, and age group (infant or toddler) was treated as a between-subjects factor. Repeated-measures analysis of covariance (ANCOVA) was also used to evaluate correlates of method bias, and child age group was used as a covariate in each model. Interactions of method and the following variables were tested in separate models: number of eating occasions and number of food items reported, maternal and child anthropometrics, family demographics, maternal social desirability, feeding method at the time of the assessments (any breast milk compared with none), and childcare participation on the day of the 24-h DR. Intraclass correlation coefficients (ICCs) were calculated to estimate the reliability of the 3-d WR measurements by using a 2-way mixed model with absolute agreement in which subjects were modeled as a random effect and EI was modeled as a fixed effect.

	RESULTS

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Participants were 157 ethnically diverse mothers (n = 56 non-Hispanic white, 51 non-Hispanic black, and 50 Hispanic) and their 7–11-mo-old infants (n = 80) or 12–24 mo-old toddlers (n = 77). Most household incomes ranged between $25 000–$50 000 (30%) and %50 000–$100 000 (35%) (Table 1). Most mothers were well-educated (84% educated beyond high school) and married (83% married or with partner). As reflected their postpartum status, mothers tended to be overweight (mean BMI: 29 ± 7); BMIs 25 and 30 were observed for 18% and 44% of mothers, respectively. More than half (59%) of the mothers reported employment, and a roughly similar proportion of mothers reported the use of childcare (60%); the largest proportion of those children were in the care of a relative (38%), and the next-largest proportion were in organized childcare (30%). On the day of the 24-h DR, 26% of infants (n = 21) and 38% of toddlers (n = 29) participated in childcare. Just under one-half of infants and toddlers (46%) were the only child in the household at the time of the study. Mothers of 94% of infants and 84% of toddlers reported ever breastfeeding; the mean duration was 3.7 ± 2.3 mo for among and 7.6 ± 4.3 mo among toddlers. On the day of the 24-h DR, 31% of caregivers with infants and 14% of caregivers with toddlers reported breastfeeding by the mother.

Table 2

Figure 1

Figure 1

View larger version (11K): [in this window] [in a new window]   FIGURE 1. Mean (±SD) daily energy intakes by 24-h dietary recall (24-h DR) and 3-d weighed records (3-d WR) with estimated energy requirements (EERs). Repeated-measures ANOVA with a between-subjects factor was used to test effects of method (24-h DR compared with 3-d WR) and age (7–11-mo-old infants compared with 12–24-mo-old toddlers) and their interaction on infant (n = 80) and toddler (n = 77) daily energy intake. Effects of method (P < 0.0001) and age (P < 0.0001) and their interaction (P < 0.0004) on energy intakes were observed. Compared with 3-d WR, the 24-h DR overestimated energy intake by 13% (P < 0.01, post hoc t test) in infants and by 29% (P < 0.0001, post hoc t test) in toddlers. Estimated energy intake by 3-d WR was within 5% of EERs. **P < 0.01, ****P < 0.0001.

Figure 2

View larger version (11K): [in this window] [in a new window]   FIGURE 2. Scatter plot of energy intakes by 24-h dietary recall (24-h DR) and 3-d weighed records (3-d WR). Points depict infant (n = 80) and toddler (n = 77) energy intakes by 24-h DR and 3-d WR dietary assessment methods. The diagonal line represents perfect agreement between methods.

Table 3

Table 4

Table 5

Intake from the grains provided the second-highest contribution to total EIs and the second-greatest contribution to overestimation by the 24-h DR (25% of energy difference between methods). The 24-h DR also produced higher estimates of EIs from fruit and vegetable food groups for both infants and toddlers and of those from the legumes group for toddlers only than did the 3-d WR. Contributing <50 kcal to total EIs, intake from the eggs, fats, and sweets groups did not differ between assessment methods.

Eating patterns
Eating most commonly occurred at home, and snacks and other types of eating occasions (primarily involving milk and other beverages) were the types of eating occasions reported most frequently (Table 6). There were no differences between methods according to the locations where feeding occurred, as reported by mothers. There also were no differences by dietary assessment method in the total number of infant or toddler eating occasions reported. However, caregivers of infants labeled a slightly greater proportion of all eating occasions by the 24-h DR than by the 3-d WR as a "breakfast" or "dinner," whereas caregivers of toddlers labeled slightly fewer eating occasions by the 24-h DR than by the 3-d WR as a "breakfast" (14% and 15%, respectively). Among infants only, caregivers labeled a greater proportion of all eating occasions by the 24-h DR than by the 3-d WR as being a "snack" but labeled fewer occasions by the 24-h DR than by the 3-d WR as fitting into an "other" category [35% and 44%, respectively (mostly milk feedings)].

Because the frequency of infant feeding and the increasing complexity of toddlers’ diets have been cited as challenges to accurate reporting, the number of eating occasions and the number of food items reported across methods were evaluated as correlates of method bias. The number of eating occasions reported was not associated with overreporting by the 24-h DR. However, overreporting increased as the average number of food items reported increased (adjusted for child age group: F = 4.6, P < 0.05).

Anthropometric, demographic, and psychosocial correlates of method bias
Differences in EIs between methods were not associated with maternal BMI. However, greater overestimation by the 24-h DR relative to the 3-d WR was associated with lower infant weight-for-length z scores (adjusted for child age group: F = 4.4, P < 0.05); this association remained statistically significant when those 8 children at <5th percentile of weight-for-length were excluded from the analysis. Attendance at childcare on the day of the 24-h DR and infant feeding method at the time of the 24-h DR were not associated with the degree of method bias. A nonsignificant trend between maternal education and method bias was observed (adjusted for child age group: F = 3.4, P = 0.07); there was greater overreporting by the less-educated mothers. Other family demographic measures (eg, maternal ethnicity and family income) and social desirability were not associated with reporting bias.

	DISCUSSION

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Given the low feasibility of labor-intensive and literacy-dependent weighed intake methods at a population level, the 24-h DR is the assessment method of choice for large quantitative studies of the diet of young children (2). The primary use of a single 24-h DR is to capture the average intake of groups (23). In the present study, a single, telephone-administered, multiple-pass 24-h DR overestimated group estimates of EIs in children aged <2 y—particularly in toddlers, among whom the mean overestimation was 29%. With the exception of small differences in meal type and the number of food items reported, the 2 methods captured similar patterns of eating. Furthermore, the absence of differences in macronutrient intakes as a percentage of total energy suggests that the relative proportions of nutrients in the diet were similar across methods. In absolute terms, however, the 24-h DR overestimated children's intake of all macronutrients and most micronutrients. Food group sources of overestimation were ordered on the basis of their contribution to total energy. Collectively, the findings suggest that overestimation by the 24-h DR was driven by errors in portion size estimation.

Dairy intake, specifically milk, provided the greatest energy to the diet and also showed the greatest overestimation by the 24-h DR. This study used FITS methodology to estimate recalled breastmilk intake by the 24-h DR; a fixed volume (600 mL/d) was assumed for infants, and time-based assumptions were made for toddlers (4). The magnitude of the overestimation of breastmilk intake by the 24-h DR was small in the full sample, which indicates that the assumptions performed reasonably well. The time-based assumptions, however, produced substantial overestimation in the small group of breastfed toddlers. Toddler intakes of "other" milks (ie, cow or soy) also were overestimated by the 24-h DR for the group as a whole. Taken together, these results suggest that bias observed in the full sample was explained to a greater extent by a general portion size overestimation than by assumptions used to estimate breastmilk.

There are few published data with which the present findings are directly comparable. As in the present research, a study of 50 underweight and normal-weight Brazilian children <2 y old found that a single interviewer-administered dietary recall overestimated mean EIs by 20% compared with a weighed record covering the same period (12). Methodologic and sociocultural differences between studies, however, limit comparisons. Other research, focused on preschool-aged and older children, showed little difference in child EIs on the basis of maternal recall compared with estimated (10, 11) and weighed (24) intakes by trained observers. Two small studies of EI among 4–7-y-old children using three 24-h DRs found little evidence of bias at the group level when total energy expenditure based on doubly labeled water was used as a reference (25, 26). However, a similarly designed study of 3–4-y-old children showed an 11% overestimation of mean energy by the three 24-h DRs (27).

Validation studies of dietary assessment methods have shown a trend toward greater energy underreporting with the increasing age of the subjects (28). In the present study, overestimation of toddler EIs by the 24-h DR was more than double that observed for infants (29% and 13%, respectively). The reasons for this difference are unclear. Unlike national trends (29), participation in any type of child-care arrangement in this sample did not increase with age (57% of infants compared with 55% of toddlers), and it was not associated with the observed bias. Mothers of toddlers also recalled one additional food item by the 24-h DR than was recorded by the 3-d WR, which explains some [67 kcal/item (1140 kcal/17 items for the 24-h DR)] but not all of the observed 255-kcal overestimation. The positive association of overestimation with the number of food items reported suggests that accurate reporting is challenged by increasing dietary complexity.

Relative child weight was a separate correlate of reporting bias, with greater overreporting by the 24-h DR in children with lower weight-for-length z scores. These findings are consistent with those of a study comparing parents’ reports of the intakes of 4–11-y-old children by multiple-pass 24-h recalls with total energy expenditure (30). In that study, parents who underreported their child's intake tended to have the heaviest children, those with the highest body fat and lowest physical fitness. It is possible that mothers in the present study may have knowingly overrepresented their child's intake. This interpretation is qualified by the fact that, unlike previous studies of reporting accuracy in adults (17), social desirability was not associated with reporting bias.

The 24-h DR method used in the present study closely paralleled that of a large, nationally random study, the FITS (4), from which population-based estimates of infant and toddler EIs have been reported. Several departures from the FITS method, however, necessitate comment. First, convenience sampling was used, given the high participant demands in keeping detailed weighed records. Although the income distribution resembled that of the FITS (4), the percentage of mothers with education beyond high school and reporting ever having breastfed (94% of infants, 84% of toddlers) were higher than in the FITS. Ever breastfeeding rates in the present study were also higher than 2004 national estimates (73.8%) (31). Whereas neither randomly chosen nor nationally representative, the sample was nevertheless composed of major racial-ethnic groups, and it varied in terms of socioeconomic indicators. A second limitation is the possibility that keeping weighed records influenced the results by altering what and how much was fed to children. That eating patterns (ie, eating frequency and location) and types of foods did not differ dramatically between methods, however, argues against an undue influence of this potential source of bias. Third, both the 24-h DR and 3-d WR were made on weekdays. Whether the same degree of bias would have been observed on weekend days is not clear. Finally, scheduling the 24-h DR and 3-d WR may have altered feeding, but it is unlikely that this potential bias would have differed by method.

The findings of this investigation suggest that efforts to improve the accuracy of the 24-h DR in children <2 y old should focus on portion size estimation. Whether error could be reduced by the use of an in-person interview mode, 3-dimensional portion size aids, or participant training (or all) remains to be seen. To date, methodologic studies evaluating the 24-h DR interview mode have found little difference between telephone and in-person administration in estimating mean EIs of women and men (33-36). Thus, a focus on participant training (35, 37) and portion size aids, particularly when milk is concerned, may prove most efficacious for improving the accuracy of the 24-h DR in estimating the EIs of young children. Because a significant percentage of infants and toddlers are routinely fed by someone other than the primary caregiver, systematic methods should be considered to maximize the quality of information obtained from multiple caregivers who are responsible for child feeding.

In conclusion, this investigation shows that the use of a single, telephone-administered, multiple-pass 24-h DR may significantly overestimate infant and toddler EIs. The nearly uniform overestimation of nutrients implies that use of a single 24-h DR may lead to erroneous dietary assessment in young children. These findings indicate that methodologic improvements to the 24-h DR for use in children <2 y old are warranted.

	ACKNOWLEDGMENTS

 
The authors’ responsibilities are as follows—JOF and NB: the design of the experiment, the analysis of data, and the writing of the manuscript; PM and EH: data collection and assistance in manuscript preparation; TW: the dietary recalls and assistance in manuscript preparation; and DD and KR: contributions to the design of the study and manuscript preparation. None of the authors a had personal or financial conflict of interest.

	REFERENCES

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Skinner JD, Ziegler P, Pac S, DeVaney B. Meal and snack patterns of infants and toddlers. J Am Diet Assoc 2004;104:S65–70.[Medline]
2. Biró G, Hulshof KF, Ovesen L, Amorim Cruz JA. Selection of methodology to assess food intake. Eur J Clin Nutr 2002;56:S25–32.
3. Dwyer J, Picciano MF, Raiten DJ, Members of the Steering Committee NHANES. Collection of food and dietary supplement intake data: What We Eat in America–NHANES. J Nutr 2003;133(suppl):590S–600S.[Abstract/Free Full Text]
4. Devaney B, Kalb L, Briefel R, Zavitsky-Novak T, Clusen N, Ziegler P. Feeding Infants and Toddlers Study: overview of the study design. J Am Diet Assoc 2004;104:S8–13.[Medline]
5. Devaney B, Ziegler P, Pac S, Karwe V, Barr SI. Nutrient intakes of infants and toddlers. J Am Diet Assoc 2004;104:S14–21.[Medline]
6. Devaney B, Kim M, Carriquiry A, Camaño-Garcia G. Assessing the nutrient intakes of vulnerable subgroups. Washington, DC: United States Department of Agriculture, Economic Research Service, 2005.
7. Institute of Medicine, Food and Nutrition Board. WIC food packages: time for a change. Washington, DC: National Academy Press, 2005.
8. Kim J, Peterson KE, Scanlon KS, et al. Trends in overweight from 1980 through 2001 among preschool-aged children enrolled in a health maintenance organization. Obesity 2006;14:1107–12.[Medline]
9. Klesges RC, Klesges LM, Brown G, Frank GC. Validation of the 24-hour dietary recall in preschool children. J Am Diet Assoc 1987;87:1383–5.[Medline]
10. Baranowski T, Sprague D, Baranowski J, Harrison JA. Accuracy of maternal dietary recall for preschool children. J Am Diet Assoc 1991;91:669–74.[Medline]
11. Basch CE, Shea S, Arliss R, et al. Validation of mothers’ reports of dietary intake by four to seven year- old children. Am J Public Health 1990;80:1314–7.[Abstract/Free Full Text]
12. Olinto MT, Victora CG, Barros FC, Gigante DP. Twenty-four-hour recall overestimates the dietary intake of malnourished children. J Nutr 1995;125:880–4.[Abstract/Free Full Text]
13. Ferguson EL, Gibson RS, Opare-Obisaw C. The relative validity of the repeated 24 h recall for estimating energy and selected nutrient intakes of rural Ghanaian children. Eur J Clin Nutr 1994;48:241–52.[Medline]
14. Persson LA, Carlgren G. Measuring children's diets: evaluation of dietary assessment techniques in infancy and childhood. Int J Epidemiol 1984;13:506–17.[Abstract/Free Full Text]
15. Lanigan JA, Wells JC, Lawson MS, Lucas A. Validation of food diary method for assessment of dietary energy and macronutrient intake in infants and children aged 6-24 months. Eur J Clin Nutr 2001;55:124–9.[Medline]
16. Davies PS, Coward WA, Gregory J, White A, Mills A. Total energy expenditure and energy intake in the pre-school child: a comparison. Br J Nutr 1994;72:13–20.[Medline]
17. Maurer J, Taren DL, Teixeira PJ, et al. The psychosocial and behavioral characteristics related to energy misreporting. Nutr Rev 2006;64:53–66.[Medline]
18. Kuczmarski RJ, Ogden C, Grummer-Strawn LM, et al. CDC growth charts: United States. Hyattsville, MD: National Center for Health Statistics, 2000.
19. Butte NF, Garza C, Smith EO, Nichols BL. Evaluation of the deuterium dilution technique against the test-weighing procedure for the determination of breast milk intake. Am J Clin Nutr 1983;37:996–1003.[Abstract/Free Full Text]
20. Lohman TG, Roche AF, Martorell R. Anthropometric standardization reference manual. Champaign, IL: Human Kinetics Books, 1988.
21. Crowne DP, Marlowe D. A new scale of social desirability independent of psychopathology. J Consult Clin Psychol 1960;24:349–54.
22. Food and Nutrition Board. Dietary reference intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein, and amino acids (macronutrients). A report of the Panel on Macronutrients, Panel on the Definition of Dietary Fiber, Subcommittees on Upper Reference Levels of Nutrients and Interpretation and Uses of Dietary Reference Intakes, and the Standing Committee on the Scientific Evaluation of Dietary Reference Intakes. Washington, DC: National Academies Press, Institute of Medicine, 2005.
23. Thompson FE, Byers T. Dietary assessment resource manual. J Nutr 1994;124(suppl):S2245–317.[Medline]
24. Klesges RC, Klesges LM, Brown G, Frank GC. Validation of the 24-hour dietary recall in preschool children. J Am Diet Assoc 1987;87:1383–5.[Medline]
25. Johnson RK, Driscoll P, Goran MI. Comparison of multiple-pass 24-hour recall estimates of energy intake with total energy expenditure determined by the doubly labeled water method in young children. J Am Diet Assoc 1996;96:1140–4.[Medline]
26. Montgomery C, Reilly JJ, Jackson DM, et al. Validation of energy intake by 24-hour multiple pass recall: comparison with total energy expenditure in children aged 5-7 years. Br J Nutr 2005;93:671–6.[Medline]
27. Reilly JJ, Montgomery C, Jackson D, MacRitchie J, Armstrong J. Energy intake by multiple pass 24 h recall and total energy expenditure: a comparison in a representative sample of 3-4-year-olds. Br J Nutr 2001;86:601–5.[Medline]
28. Livingstone MBE, Robson PJ. Measurement of dietary intake in children. Proc Nutr Soc 2000;59:279–93.[Medline]
29. Mulligan GM, Brimhall D, West J. Child care and early education arrangements of infants, toddlers, and preschoolers: 2001. Washington, DC: US Government Printing Office, 2005;
30. Fisher JO, Johnson RK, Lindquist CH, Birch LL, Goran MI. Influence of body composition on the accuracy of reported energy intake in children. Obes Res 2000;8:597–603.[Medline]
31. Centers for Disease Control and Prevention. Breastfeeding trends and updated national health objectives for exclusive breastfeeding—United States, birth years 2000-2004. MMWR Morb Mortal Wkly Rep 2007;56:760–3.[Medline]
32. Food and nutrient intakes by children 1994-96, 1998; table set 17. Beltsville, MD: Agricultural Research Service, US Department of Agriculture, 1999.
33. Casey PH, Goolsby SL, Lensing SY, Perloff BP, Bogle ML. The use of telephone interview methodology to obtain 24-hour dietary recalls. J Am Diet Assoc 1999;99:1406–11.[Medline]
34. Tran KM, Johnson RK, Soultanakis RP, Matthews DE. In-person vs telephone-administered multiple-pass 24-hour recalls in women: validation with doubly labeled water. J Am Diet Assoc 2000;100:777–83.[Medline]
35. Yanek LR, Moy TF, Raqueño JV, Becker DM. Comparison of the effectiveness of a telephone 24-hour dietary recall method vs an in-person method among urban African-American women. J Am Diet Assoc 2000;100:1172–7.[Medline]
36. Bogle M, Stuff J, Davis L, et al. Validity of a telephone-administered 24-hour dietary recall in telephone and non-telephone households in the rural Lower Mississippi Delta region. J Am Diet Assoc 2001;101:216–22.[Medline]
37. Howat PM, Mohan R, Champagne C, Monlezun C, Wozniak P, Bray GA. Validity and reliability of reported dietary intake data. J Am Diet Assoc 1994;94:169–73.[Medline]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Google Scholar Articles by Fisher, J. O Articles by Deming, D. PubMed PubMed Citation Articles by Fisher, J. O Articles by Deming, D. Agricola Articles by Fisher, J. O Articles by Deming, D.