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 Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by de Castro, J. M. Search for Related Content PubMed PubMed Citation Articles by de Castro, J. M.

---

Ingestive Behavior and Neurosciences

Varying Levels of Food Energy Self-Reporting Are Associated with Between-Group, but Not Within-Subject, Differences in Food Intake

Department of Psychology, University of Texas at El Paso, El Paso, TX

¹ To whom correspondence should be addressed. E-mail: jdecastro{at}utep.edu.

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
Energy intakes reported in diet diaries are frequently significantly below the energy requirements for a weight-stable individual. To investigate the impact of low energy reporting on between-group comparisons and within-subject analyses of eating behavior, we examined the baseline 7-d food-intake diaries submitted by 365 male and 564 female free-living normal adult humans. Participants were separated into 5 different groups, based upon their levels of energy reporting relative to their predicted basal metabolic rate: EI:BMR_est = 0–0.99, 1.0–1.199, 1.2–1.399, 1.3–1.599, and 1.6. Between-group analysis revealed significant inverse relations among reporting level and body weight, BMI, cognitive restraint, positive relations with intake, meal size, and meal frequency. On the other hand, within-subject analyses suggested that, regardless of the level of energy reporting, equivalent relations are found among the amounts eaten in meals and the presence of other people, palatability, hunger, satiety, dietary energy density, contents of the stomach, time since the last meal, and time of day and correlations between daily intake and intake on subsequent days. The results suggest that comparing the intakes reported in diet diaries by different groups may be confounded by group differences in reporting levels. In contrast, the results clearly support the conclusion that diet diary data are suitable for assessing the relations between variables assessed within subjects irrespective of the level of reporting of energy intake.

KEY WORDS: • eating • meals • underreporting • undereating

Self-reports of dietary intake have been used for dietary assessment for decades, but increasingly, their accuracy is being questioned. Early reports indicate that 24-h recalls (1) and the 7-d diary are fairly reliable, with reasonable agreement among diet diary records repeated as long as 2-y later (2–6) and good agreement between diary records and actual intakes (7,8). However, the intake amounts reported in diaries are 20% below what should be ingested by subjects with average activity levels and stable body weights [(5,9–15); see (16) for review]. This reporting of low levels of energy intake appears to result from both undereating due to reactivity to the measurement procedure and underreporting of actual intake (17,18). Focus groups have suggested that the tendency toward low energy reporting is due to a combination of undereating, errors in portion size estimation, and the inconvenience of having to write every item down no matter how small (19).

Low energy reporting would not be a problem for researchers and could easily be adjusted if it were random across participants, situations, and food types. Unfortunately, low energy reporting appears to be greater in certain groups and with certain food types than others. Low energy reporting appears to be greater in subjects with higher body weight and/or BMI [(20–25); see (26) for review], in those with greater tendencies to self-restrict energy intake (27), and in those with greater needs for approval from other people (19). Low energy reporting is also selective for particular nutrients and claims lower proportions of fat and alcohol intakes and higher proportions of protein intakes [(23–25); see (16) for review]. In addition, low energy reporting is associated with reports of larger quantities of purported "good" foods, such as meats, fish, vegetables, fruits, and salads and lower quantities of purported "bad" foods, such as cakes, cookies, candies, and fried foods [(19); see (16) for review]. Consequently, low energy reporting can present interpretive difficulties for research that compares the dietary intakes of groups with different characteristics, such as degrees of obesity or dietary restraint. Confounding occurs because it cannot be discriminated whether group differences are due to their differing characteristics, their differing levels of energy reporting, or both.

Many dietary research questions, however, involve within-subject comparisons where an individual's intake under one condition is compared with that same individual's intake under a different condition, e.g., eating alone vs. eating in groups. This is done to assess environmental, social, and dietary variables upon intake (28–30). It is unclear to what extent low energy reporting affects these kinds of comparisons. If low energy reporting is selective for certain situations, then interpretation of within-subject dietary data would also be problematic. If, however, low energy reporting is relatively consistent or random across a situation, then within-subject comparisons will not be contaminated and self-reports of intakes can be considered a reasonably valid procedure for within-subject dietary assessments.

The present study attempted to address the effect of different levels of energy reporting on the interpretation of diet diary self-reports by performing both between- and within-subject analyses on baseline 7-d diary reports obtained from normal adults who participated in prior nonmanipulative studies (28–38) and who differed in levels of energy reporting. Reported intakes were compared with an estimated basal metabolic rate (BMR_est) derived from the participant's weight and with a consideration of age and gender (39). The ratio (EI:BMR_est) of the reported daily food energy intake (EI) to the BMR_est was calculated for each participant (40,41). It has been suggested that this ratio should be 1.55 for subjects with average activity levels and stable body weights (13,16). Five groups of individuals with reported intakes that fell within 5 different ranges of the EI:BMR_est were then compared on both between- and within-group measures of intake.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
    Participants. The baseline 7-d dietary data obtained from all adults recruited for a number of prior nonmanipulative studies was used (28–38). A majority of the participants, 536, were recruited by mail; they were paid $30 and also received a detailed nutritional analysis of their intake. Two-hundred twelve of the participants volunteered in response to a newspaper advertisement and solely for the nutritional analysis and 181 participants were undergraduate students who volunteered to satisfy a course requirement. The participants were 36.5 ± 14.6 y of age, weighed 68.9 ± 15.8 kg, had a height of 1.67 ± 0.09 m, and a BMI of 24.5 ± 4.5. There were no specific inclusion criteria, but participants were excluded if they were actively dieting, pregnant or lactating, on chronic medication, or alcoholic as ascertained with a demographic questionnaire. The protocols were approved by the Georgia State University Institutional Review Board.

    Procedure. Participants completing a demographic questionnaire were asked to record their current weight and the lowest and highest body weights achieved during their adulthood. Embedded in the questionnaire was the Three-Factor Eating Questionnaire (42), which measures cognitive restraint, disinhibition, and perceived hunger (43,44). The participants were given a small (8 x 18 cm) pocket-sized diary and were instructed to record, in as detailed a manner as possible, every item that they either ate or drank, the time they ate it, the amount they consumed, how the food was prepared, and the number of other people eating with them. Self-ratings were obtained of the participants' degree of hunger, thirst, food attractiveness, depression, elation, and anxiety on a 7-point Likert scale both at the beginning and end of every meal. The participants initially recorded this information for 1 d and were then contacted by the experimenter who reviewed the information, corrected any problems, and answered any questions. The participants were then asked to record their intake for 7 consecutive days. Afterwards, the experimenter reviewed the diaries with the participants, and clarified any ambiguities or missing data. For a detailed review of the diet diary method and its reliability and validity see de Castro (28–30).

    Data analysis. The foods reported in the diaries were assigned codes from a computer file developed by an experienced dietitian containing over 3500 food items. The coder was unaware of the experimental hypotheses and the participants' characteristics and did not interact directly with the participants. Meals were identified and the food energy (kJ) and nutrient compositions of the individual items comprising the meal were summed. Meals were defined based upon the amount eaten and the time from the preceding and following intakes (before- and after-meal intervals). Five different definitions of a meal were used: 15 min and 209 kJ, 45 min and 209 kJ, 45 min and 418 kJ, 45 min and 837 kJ, and 90 min and 209 kJ.

The meals were characterized by their total energy content, carbohydrate, fat, protein, alcohol content, duration of the meal, the rate of intake (kJ/min), the before- and after-meal intervals (min), the before-meal and after-meal ratings of palatability, hunger, thirst, elation, and anxiety, and the dietary energy density of the meal (kJ/g), which was calculated both including and excluding any beverages ingested. Specific food types were chosen for the study based on the following reasons: 1) to encompass a range of different foods and beverages; 2) the practical criteria with which they could easily be identified in the computer file; and 3) they were ingested frequently enough by a sufficient number of participants for meaningful analysis. The specific beverage categories that were chosen for study were: alcohol, sodas containing sugar, diet sodas, milk, coffee, tea, and fruit juices. The food categories selected were: fruit, cheese, ice cream, soup, fish, beef, chicken, turkey, pork, beans, rice, potatoes, other vegetables, nuts, bread, pasta, pastries, cookies, chips and snacks, condiments, pizza, eggs, dairy products, breakfast cereal, pancakes etc., candy, sweeteners, oils, and fats. The amounts of each food or beverage item ingested over the entire 7-d period were recorded.

The contents of the stomach at the beginning and end of the meals were estimated from the 7-d diary records with a computer model of stomach emptying. The reported intake was estimated to empty from the stomach at a rate proportional to the square root of the energy content of the stomach, S_n+1 = S_n – 5S_n, where S equals the stomach content in calories and n equals a particular minute of the day. This procedure reflects the actual measured emptying rate from the human stomach (45–47) and has been used in prior studies (37,38,48,49). The mean of each of the meal characteristics and the mean daily intakes were then calculated for each participant. These individual means were then used to calculate group means.

The within-subject analysis investigated each individual's responsiveness to variables related to the amount eaten in the meal. Social effects (36,37) were investigated by separating meals eaten alone from those eaten with 1 other person and those eaten with 2 people. In addition, individual correlations were calculated between the number of people present at the meal and the meal size. The influence of the time of day of the meal (33) was investigated by separating, for each participant, meals eaten in the morning (0600–1159), afternoon (1200–1759), and evening (1800–2400). Also, individual correlations were calculated between the time of the meal (min) and the meal size. The influence of before-meal hunger (37) was investigated by separating, for each participant, meals eaten with low (1–3), medium (4–5), and high (6–7) self-ratings of hunger. Also, individual correlations were calculated between the hunger ratings and the meal size. Similarly, the influence of after-meal satiety (37) was investigated by separating, for each participant, meals eaten with low (1–3), medium (4–5), and high (6–7) self-ratings of satiety. Also, individual correlations were calculated between the meal size and the satiety ratings.

The influence of the duration of the interval before the meal (38) was investigated by separating, for each participant, meals eaten after short intervals (<180 min), moderate intervals (181–240 min), and long intervals (>240 min). Also, individual correlations were calculated between the duration of the before-meal interval (min) and the meal size. The influence of the contents of the stomach before the meal (38,49) was investigated by separating, for each participant, meals eaten with low amounts in the stomach before the meal (<167 kJ), medium (168–0.500 kJ), and high (>500 kJ). Also, individual correlations were calculated between the before-meal stomach content (kJ) and the meal size. The influence of the energy density of the meal (32) was investigated by separating, for each participant, meals of low dietary energy density (<3.35 kJ/g), medium (3.35–4.6 kJ/g), and high (>4.6 kJ/g). Also, individual correlations were calculated between the energy density of the meal (kJ) and the meal size. Finally, the influence of the palatability of the meal (34) was investigated by separating, for each participant, meals eaten with low self-ratings of palatability (1–4), medium (5), and high (6–7). Also, individual correlations were calculated between the meal size and the palatability ratings. Correlations were also calculated between the amount eaten over a day and the amounts eaten on subsequent days (31) for each participant over the 7 d of intake recording. Group mean correlations were then calculated by averaging the individual correlations. Since correlations are not normally distributed, the coefficients were transformed to Z-scores prior to calculating group means (50). The mean correlations were then compared with 0 with a t test.

To measure the level of self-reported energy intake relative to the participant's estimated requirements, the basal metabolic rate was estimated from the participant's weight considering age and sex according to the procedure outlined by Schofield et al. (39). The ratio of the reported daily food energy intake (EI) to the BMR_est (EI:BMR_est) was calculated for each participant. Five groups were then identified based upon their EI:BMR_est; < 1.0, 1.0–1.199, 1.2–1.399, 1.4–1.599, and 1.6, respectively. These ranges of EI:BMR_est were selected because they covered the range of calculated EI:BMR_est and had reasonable numbers of participants in each category. The number of participants who fell into the 5 categories were: 68, 83, 83, 65, and 66 males (19, 23, 23, 18, and 18% of males) and 141, 103, 122, 111, and 87 females (25, 18, 22, 20, and 15% of females) for the 5 groups, respectively.

For the between-subject analysis, the participant characteristics, daily intake, and meal pattern variables were compared with a Gender by Reporting Group, 2 x 5 ANOVA and with an ANOVA over the 5 reporting groups. Individual group means were compared post hoc with t tests, only if the F-value from the ANOVA was significant. All reported results were significant at = 0.05, unless otherwise indicated.

Separate analyses were performed for males and females and there were significant gender differences in the magnitudes of the variables. But, no significant Gender by Reporting Level interaction was present. Because there were no differences between groups by meal definition, all results are presented using the 418 kJ, 15-min meal definition.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
    Characteristics of participants reporting at different levels. Participants who self-reported intakes at the 5 different levels differed primarily in body size and cognitive restraint (Fig. 1). These participants did not differ in age or height, but participants who reported intake at the 3 lowest levels had significantly higher lowest, current, and highest weights and BMIs than the high-reporting participants. Interestingly, the current weights of the low-reporting participants were significantly farther from their lowest adult weight and the range from highest to lowest adult weights were significantly greater than those of the high-reporting participants. In addition, the low-reporting participants had significantly higher levels of restraint than the high-reporting participants. But, the groups did not differ in disinhibition or perceived hunger. Hence low energy reporters were characterized by higher cognitive restraint, BMIs, and weights, which were farther from their lowest adult weights.

View larger version (28K): [in this window] [in a new window]   FIGURE 1  Current, lowest, and highest adult body weights (A), BMI (B), and cognitive restraint (C) scores from the Three-Factor Eating Questionnaire for participants self-reporting total food energy intakes at 5 different levels relative to their predicted basal metabolic rate; EI:BMREST = 0–0.99 (n = 209), 1.0–1.199 (n = 186), 1.2–1.399 (n = 205), 1.3–1.599 (n = 176), and 1.6 (n = 153). Values are means ± SEM. Means without a common letter differ, P < 0.05, as assessed with post hoc t tests when a 1-way ANOVA for each variable indicated a significant group effect.

Because the groups differed in overall levels of intake, their gram intakes of specific types of foods and beverages were expressed as a percentage of the total gram intake for the day. For the vast majority of food types there were no significant group differences. But, the lower reporting groups had significantly higher percentage intakes of vegetables, chicken, fish, and beans and lower intakes of cheese, ice cream, cookies, nuts, chips, and snack foods. In addition, the lower-reporting groups had higher percentage intakes of coffee and tea and diet sodas and lower intakes of sugared sodas, milk, and alcoholic beverages, with no differences in the intakes of water and fruit juices.

    Within- and between-subject analysis of intake relations with environmental, dietary, and subjective variables for participants reporting at different levels. The number of people present at a meal is a powerful influence on the amount eaten in the meal (35,36), and, in the present analysis, larger meals were observed when eaten with 2 other people present than with only 1 other person present, which, were in turn, larger than alone (Fig. 2). Although the 5 groups differed in meal sizes, the relation between the number of people present and meal size was not different among the groups and no interaction effect was present. In addition, the correlations between the number of people present and meal size were positive (P < 0.05) for all groups and did not differ among groups (Fig. 2). Hence, even though the 5 groups differed in the size of meals ingested, they were equivalent in their response to social facilitation.

View larger version (21K): [in this window] [in a new window]   FIGURE 2  Meal sizes for meals eaten alone, with one other person, and 2 other people for participants self-reporting total food energy intakes at 5 different levels relative to their predicted basal metabolic rate; EI:BMRest = 0–0.99 (n = 209), 1.0–1.199 (n = 186), 1.2–1.399 (n = 205), 1.3–1.599 (n = 176), and 1.6 (n = 153). Values are means ± SEM. The correlations between the number of people present at the meal and the meal size were calculated separately for each group and are presented on top of each group's set of 3 bars. Values are mean correlations and all are significantly different from zero (t tests, P < 0.05).

View larger version (22K): [in this window] [in a new window]   FIGURE 3  Meal sizes for meals eaten in the morning, afternoon, or evening for participants self-reporting total food energy intakes at 5 different levels relative to their predicted basal metabolic rate; EI:BMRest = 0–0.99 (n = 209), 1.0–1.199 (n = 186), 1.2–1.399 (n = 205), 1.3–1.599 (n = 176), and 1.6 (n = 153). Values are means ± SEM. The correlations between the time of day and the meal size were calculated separately for each group and are presented on top of each group's set of 3 bars. Values are mean correlations and all are significantly different from zero (t tests, P < 0.05).

View larger version (21K): [in this window] [in a new window]   FIGURE 4  Meal sizes for meals eaten at low, medium, and high self-reported levels of hunger at the beginning of the meal for participants self-reporting total food energy intakes at 5 different levels relative to their predicted basal metabolic rate; EI:BMRest = 0–0.99 (n = 209), 1.0–1.199 (n = 186), 1.2–1.399 (n = 205), 1.3–1.599 (n = 176), and 1.6 (n = 153). Values are means ± SEM. The correlations between the before meal hunger ratings and the meal size were calculated separately for each group and are presented on top of each group's set of 3 bars. Values are mean correlations and all are significantly different from zero (t tests, P < 0.05, n = 929).

View larger version (24K): [in this window] [in a new window]   FIGURE 5  Meal sizes for meals eaten following relatively short, moderate, or long intervals since the last meal for participants self-reporting total food energy intakes at 5 different levels relative to their predicted basal metabolic rate; EI:BMRest = 0–0.99 (n = 209), 1.0–1.199 (n = 186), 1.2–1.399 (n = 205), 1.3–1.599 (n = 176), and 1.6 (n = 153). Values are means ± SEM. The correlations between the before meal interval and the meal size were calculated separately for each group and are presented on top of each groups set of 3 bars. Values are mean correlations and all are significantly different from zero (t tests, P < 0.05).

The energy density of the food eaten is strongly associated with the amount consumed in the meal (32). In the present study, larger meals were eaten when they were relatively high in energy density than those of moderate energy density, which were, in turn, larger than when they were relatively low in energy density (Fig. 6). Again, the positive relation between the energy density of the meal and meal size was not different among the groups and no interaction effect was present. In addition, the correlations between the energy density of the meal and meal size were positive (P < 0.05) for all groups and did not differ among groups (Fig. 6). Hence, even though the 5 groups differed in the size of meals ingested, they were equivalent in their response to the energy density of the meal.

View larger version (23K): [in this window] [in a new window]   FIGURE 6  Meal sizes for meals eaten at low, medium, and high levels of energy density for participants self-reporting total food energy intakes at 5 different levels relative to their predicted basal metabolic rate; EI:BMRest = 0–0.99 (n = 209), 1.0–1.199 (n = 186), 1.2–1.399 (n = 205), 1.3–1.599 (n = 176), and 1.6 (n = 153). Values are means ± SEM. The correlations between the energy density and the meal size were calculated separately for each group and are presented on top of each group's set of 3 bars. Values are mean correlations and all are significantly different from zero (t tests, P < 0.05).

View larger version (20K): [in this window] [in a new window]   FIGURE 7  Correlations between the total amount of food energy ingested in a day and the amount ingested on the next d, 2 d later, and 3 d later for participants self-reporting total food energy intakes at 5 different levels relative to their predicted basal metabolic rate; EI:BMRest = 0–0.99 (n = 209), 1.0–1.199 (n = 186), 1.2–1.399 (n = 205), 1.3–1.599 (n = 176), and 1.6 (n = 153). Values are mean correlations ± SEM. *Indicates correlations different from zero, P < 0.05.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The present analysis also found that, although the levels of all macronutrients ingested were lower in low energy reporting groups, these groups tended to report a smaller proportion of their intake as fat and alcohol and a larger proportion as protein than higher energy reporting groups. These macronutrient differences have also been observed by others [(23–25); see (16) for review]. This may be due to the tendency of the low energy reporting groups to report proportionately greater intakes of high protein items such as chicken and fish and proportionately lower intakes of high fat foods such as cheese, ice cream, cookies, nuts, and chips as was observed presently and by others [(19); see (16) for review].

The present analysis along with the prior findings indicate that great care must be taken in interpreting results when diet diary reports of intake are used to estimate the absolute levels of intake or to compare the intakes of groups with different characteristics. Low energy reporting influences the magnitude of the estimates of intake, which can result in spuriously low estimates of population nutrient intakes. More seriously for research, however, is the fact that certain individuals, particularly overweight or participants who restrain intake, report lower energy intakes than others. Confounding occurs because it cannot be determined whether group differences are due to their differing characteristics, their differing levels of energy reporting, or both. One approach to resolving this problem may to stratify the groups on the basis of reporting level and then compare the intakes of different groups within the same strata. However, this procedure might eliminate real differences between the intakes of different groups. For example, obese participants may be underreporting energy intake, whereas normal-weight participants may be restricting intake. Both conditions would result in low levels of energy reporting. The groups would not appear to differ, when in fact, the obese participants were ingesting substantially more than normal-weight participants in the same strata. Hence, great care must be exercised in interpreting diet diary information from different groups of participants.

The present study differed from prior research in that it also investigated the associations between energy reporting levels and the relations between intake and psychological, social, and environmental variables. It is significant that the 5 reporting groups were equivalent in the associations among intake and the presence of other people, palatability, hunger, satiety, dietary energy density, contents of the stomach, time since the last meal, and time of day. With all of these variables, the relation between the levels of these variables and the size of the meal was not different among groups. This suggests that low energy reporting does not affect the results when the variables' relations with intake are assessed within subjects, comparing the relations between intake and different levels of the variables occurring for the same subject. In this case, the magnitude of the intake estimates becomes less important. Correlations are not affected by the magnitude of the variables, which can be multiplied or divided by constants without affecting the magnitude of the correlations. This is encouraging and implies that, as long as a participant's eating behavior is being compared with the same participant's behavior in different circumstances, that diet-diary assessment is an appropriate measurement procedure.

There are a number of methods of dietary assessment, including 24-h recalls, food frequency questionnaires, etc., but these techniques do not provide the detailed meal-to-meal measures needed for assessing patterns of intake and their relation to environmental, psychological, and social variables. Because there are currently no alternatives to diet diaries for these kinds of assessments, it is important to know under what circumstances reasonably straightforward conclusions can be drawn and under what circumstances great care must be exercised in data interpretation. Clearly, the impact of different reporting levels demands great caution in interpreting diet diary data when it is used to assess differences in intakes among groups, such as normal, overweight, and obese groups, groups varying in levels of dietary restraint, and ethnic or socioeconomic groups. On the other hand, the present results clearly support the conclusion that diet-diary data are suitable for assessing the relation between variables within subjects irrespective of the level of reporting of energy intake.

Manuscript received 5 October 2005. Initial review completed 30 November 2005. Revision accepted 13 February 2006.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 

1. Beaton GH, Milner J, Corey P, McGuire V, Cousins M, Stewart E, de Ramos M, Hewitt D, Grambsch PV, et al. Sources of variance in 24-hour dietary recall data: implications for nutrition study design and interpretation. Am J Clin Nutr. 1979;32:2546–59.[Free Full Text]

2. Adelson SF. Some problems in collecting dietary data from individuals. J Am Diet Assoc. 1960;36:453–61.[Medline]

3. Block G. A review of validations of dietary assessment methods. Am J Epidemiol. 1982;115:492–505.[Free Full Text]

4. Day N, McKeown N, Wong M, Welch A, Bingham S. Epidemiological assessment of diet: a comparison of a 7-day diary with a food frequency questionnaire using urinary markers of nitrogen, potassium and sodium. Int J Epidemiol. 2001;30:309–17.[Abstract/Free Full Text]

5. Livingstone MB, Prentice AM, Strain JJ, Coward WA, Black AE, Barker AE, Mckenna PG, Whitehead RG. Accuracy of weighed dietary records in studies of diet and health. BMJ. 1990;300:708–12.[Abstract/Free Full Text]

6. St Jeor ST, Guthrie HA, Jones MB. Variability of nutrient intake in a 28 day period. J Am Diet Assoc. 1983;83:155–62.[Medline]

7. Gersovitz M, Madden JP, Smicikalas-Wright H. Validity of the 24-hour dietary recall and seven-day record for group comparisons. J Am Diet Assoc. 1978;73:48–55.[Medline]

8. Krantzler NJ, Mullen BJ, Schutz HG, Grivetti LE, Holden CA, Meiselman HL. The validity of telephoned diet recalls and records for assessment of individual food intake. Am J Clin Nutr. 1982;36:1234–42.[Abstract/Free Full Text]

9. Bandini LG, Schoeller DA, Cyr HN, Dietz WH. Validity of reported energy intake in obese and nonobese adolescents. Am J Clin Nutr. 1990;52:421–5.[Abstract/Free Full Text]

10. Goran MI, Poehlman ET. Total energy expenditure and energy requirements in healthy elderly persons. Metabolism. 1992;41:744–53.[Medline]

11. Lissner L, Habicht J-P, Strupp BJ, Levitsky DA, Haas JD, Roe DA. Body composition and energy intake: do overweight women overeat and underreport? Am J Clin Nutr. 1989;49:320–5.[Abstract/Free Full Text]

12. Livingstone MB, Prentice AM, Coward WA, Strain JJ, Black AE, Davies PSW, Steward CM, Mckenna PG, Whitehead RG. Validation of estimates of energy intake be weighted-dietary record and diet history in children and adolescents. Am J Clin Nutr. 1992;56:29–35.[Abstract/Free Full Text]

13. Livingstone MB, Robson PJ, Black AE, Coward WA, Wallace JM, McKinley MC, Strain JJ, McKenna PG. An evaluation of the sensitivity and specificity of energy expenditure measured by heart rate and the Goldberg cut-off for energy intake: basal metabolic rate for identifying mis-reporting of energy intake by adults and children: a retrospective analysis. Eur J Clin Nutr. 2003;57:455–63.[Medline]

14. Mertz W, Tsui JC, Judd JT, Reiser S, Hallfrisch J, Morris ER, Steele PD, Lashley E. What are people really eating? The relation between energy intake derived from estimated diet records and intake determined to maintain body weight. Am J Clin Nutr. 1991;54:291–6.[Abstract/Free Full Text]

15. Seale JL, Rumpler WV. Comparison of energy expenditure measurements by diet records, energy intake balance, doubly labeled water and room calorimetry. Eur J Clin Nutr. 1997;51:856–63.[Medline]

16. Livingstone MB, Black AE. Markers of the validity of reported energy intake. J Nutr. 2003;133: Suppl 3:895S–920S.[Abstract/Free Full Text]

17. Goris AH, Meijer EP, Westerterp KR. Repeated measurement of habitual food intake increases under-reporting and induces selective under-reporting. Br J Nutr. 2001;85:629–34.[Medline]

18. Goris AH, Westerterp KR. Underreporting of habitual food intake is explained by undereating in highly motivated lean women. J Nutr. 1999;129:878–82.[Abstract/Free Full Text]

19. Scagliusi FB, Polacow VO, Artioli GG, Benatti FB, Lancha AH Jr. Selective underreporting of energy intake in women: magnitude, determinants, and effect of training. J Am Diet Assoc. 2003;103:1306–13.[Medline]

20. Braam LA, Ocke MC, Bueno-de-Mesquita HB, Seidell JC. Determinants of obesity-related underreporting of energy intake. Am J Epidemiol. 1998;147:1081–6.[Abstract/Free Full Text]

21. Heitmann BL, Lissner L. Dietary underreporting by obese individuals–is it specific or non-specific? BMJ. 1995;311:986–9.[Abstract/Free Full Text]

22. Price GM, Paul AA, Cole TJ, Wadsworth ME. Characteristics of the low energy reporters in a longitudinal national dietary survey. Br J Nutr. 1997;77:833–51.[Medline]

23. Samaras K, Kelly PJ, Campbell LV. Dietary underreporting is prevalent in middle-aged British women and is not related to adiposity (percentage body fat). Int J Obes Relat Metab Disord. 1999;23:881–8.[Medline]

24. Voss S, Kroke A, Klipstein-Grobusch K, Boeing H. Is macronutrient composition of dietary intake data affected by underreporting? Results from the EPIC-Potsdam Study. European Prospective Investigation into Cancer and Nutrition. Eur J Clin Nutr. 1998;52:119–26.[Medline]

25. Warwick PM, Reid J. Trends in energy and macronutrient intakes, body weight and physical activity in female university students (1988–2003), and effects of excluding under-reporters. Br J Nutr. 2004;92:679–88.[Medline]

26. Trabulsi J, Schoeller DA. Evaluation of dietary assessment instruments against doubly labeled water, a biomarker of habitual energy intake. Am J Physiol Endocrinol Metab. 2001;281:E891–9.[Abstract/Free Full Text]

27. Asbeck I, Mast M, Bierwag A, Westenhofer J, Acheson KJ, Muller MJ. Severe underreporting of energy intake in normal weight subjects: use of an appropriate standard and relation to restrained eating. Public Health Nutr. 2002;5:683–90.[Medline]

28. de Castro JM. Methodology, correlational analysis, and interpretation of diet diary records of the food and fluid intakes of free-living humans. Appetite. 1994;23:179–92.[Medline]

29. de Castro JM. Measuring real-world eating behavior. In: Guy-Grand B, Ailhaud G, Progress in obesity research: 8. Eastleigh, UK: John Libbey & Company, Ltd: 1999; 215–21.

30. de Castro JM. The control of eating behavior in free-living humans. In: Stricker EM, Woods SC, editors. Handbook of the behavioral neurobiology: volume 14, neurobiology of food and fluid intake. New York: Plenum; 2004. p. 469–504.

31. de Castro JM. Prior days intake has macronutrients specific delayed negative feedback effects on the spontaneous food intake of free-living humans. J Nutr. 1998;128:61–7.[Abstract/Free Full Text]

32. de Castro JM. Dietary energy density is associated with heightened intake in free-living humans. J Nutr. 2004;134:335–41.[Abstract/Free Full Text]

33. de Castro JM. The time of day of food intake influences overall intake in humans. J Nutr. 2004;134:104–11.[Abstract/Free Full Text]

34. de Castro JM, Bellisle F, Dalix AM, Pearcey S. Palatability and intake relationships in free-living humans: Characterization and independence of influence in North Americans. Physiol Behav. 2000;70:343–50.[Medline]

35. de Castro JM, Brewer EM. The amount eaten in meals by humans is a power function of the number of people present. Physiol Behav. 1992;51:121–5.[Medline]

36. de Castro JM, de Castro ES. Spontaneous meal patterns in humans: influence of the presence of other people. Am J Clin Nutr. 1989;50:237–47.[Abstract/Free Full Text]

37. de Castro JM, Elmore DK. Subjective hunger relationships with meal patterns in the spontaneous feeding behavior of humans: Evidence for a causal connection. Physiol Behav. 1988;43:159–65.[Medline]

38. de Castro JM, McCormick J, Pedersen M, Kreitzman SN. Spontaneous human meal patterns are related to preprandial factors regardless of natural environmental constraints. Physiol Behav. 1986;38:25–9.[Medline]

39. Schofield WN. Predicting basal metabolic rate, new standards and review of previous work. Hum Nutr Clin Nutr. 1985;39: Suppl. 1:5–41.

40. Black AE, Goldberg GR, Jebb SA, Livingstone MB, Cole TJ, Prentice AM. Critical evaluation of energy intake data using fundamental principles of energy physiology: 2. Evaluating the results of published surveys. Eur J Clin Nutr. 1991;45:583–99.[Medline]

41. Goldberg GR, Black AE, Jebb SA, Cole TJ, Murgatroyd PR, Coward WA, Prentice AM. Critical evaluation of energy intake data using fundamental principles of energy physiology: 1. Derivation of cut-off limits to indentifying under-recording Br J Clin Nutr. 1991;45:569–81.

42. Stunkard AJ, Messick S. The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger. J Psychosom Res. 1985;29:71–83.[Medline]

43. Banasiak SJ, Wertheim EH, Koerner J, Voudouris NJ. Test-retest reliability and internal consistency of a variety of measures of dietary restraint and body concerns in a sample of adolescent girls. Int J Eat Disord. 2001;29:85–9.[Medline]

44. Karlsson J, Persson LO, Sjostrom L, Sullivan M. Psychometric properties and factor structure of the Three-Factor Eating Questionnaire (TFEQ) in obese men and women. Results from the Swedish Obese Subjects (SOS) study. Int J Obes Relat Metab Disord. 2000;24:1715–25.[Medline]

45. Hopkins A. The pattern of gastric emptying: a new view of old results. J Physiol (Lond). 1966;182:144–50.[Abstract/Free Full Text]

46. Hunt JN, Knox MT. (1968) Regulation of gastric emptying. In: Code CF, Heidel W, editors, Handbook of physiology: alimentary canal, vol. 4: motility. Washington D.C.: American Physiological Society; 1968, p. 1917–35.

47. Hunt JN, Stubbs DF. The volume and content of meals as determinants of gastric emptying. J Physiol (Lond). 1975;245:209–25.[Abstract/Free Full Text]

48. Booth DA. A simulation model of psychobiosocial theory of human food-intake controls. Int J Vitam Nutr Res. 1988;58:119–34.[Medline]

49. de Castro JM. Inheritance of premeal stomach content influences on eating and drinking in free living humans. Physiol Behav. 1999;66:223–32.[Medline]

50. de Castro JM. Meal pattern correlations: facts and artifacts. Physiol Behav. 1975;15:13–5.[Medline]

51. Ballard-Barbash R. Graubard I, Krebs-Smith SM, Schatzkin A, Thompson FE. Contribution of dieting to the inverse association between energy intake and body mass index. Eur J Clin Nutr. 1996;50:98–106.[Medline]

52. Kant AK. Interaction of body mass index and attempt to lose weight in a national sample of US adults: association with reported food and nutrient intake, and biomarkers. Eur J Clin Nutr. 2003;57:249–59.[Medline]

This article has been cited by other articles:

				N. Stroebele, J. M. de Castro, J. Stuht, V. Catenacci, H. R. Wyatt, and J. O. Hill A Small-Changes Approach Reduces Energy Intake in Free-Living Humans J. Am. Coll. Nutr., February 1, 2009; 28(1): 63 - 68. [Abstract] [Full Text] [PDF]

				J. E Schebendach, L. E. Mayer, M. J Devlin, E. Attia, I. R Contento, R. L Wolf, and B T. Walsh Dietary energy density and diet variety as predictors of outcome in anorexia nervosa Am. J. Clinical Nutrition, April 1, 2008; 87(4): 810 - 816. [Abstract] [Full Text] [PDF]

				C. B. Ebbeling, M. M. Leidig, H. A. Feldman, M. M. Lovesky, and D. S. Ludwig Effects of a Low-Glycemic Load vs Low-Fat Diet in Obese Young Adults: A Randomized Trial JAMA, May 16, 2007; 297(19): 2092 - 2102. [Abstract] [Full Text] [PDF]

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 Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by de Castro, J. M. Search for Related Content PubMed PubMed Citation Articles by de Castro, J. M.