![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
3 Division of Public Health Sciences and 4 Department of Family and Community Medicine, Wake Forest University School of Medicine, Winston-Salem, NC 27157; and 5 Department of Health Behavior and Health Education, School of Public Health, University of North Carolina, Chapel Hill, NC 27599
* To whom correspondence should be addressed. E-mail: jtooze{at}wfubmc.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT IntroductionSubjects and MethodsResultsDiscussionLITERATURE CITED |
|---|
|
Introduction |
|---|
|
TOPABSTRACTIntroductionSubjects and MethodsResultsDiscussionLITERATURE CITED |
|---|
Studies have investigated implausible reports of energy intake, almost always in the direction of low energy reporting, on self-report instruments of diet, including food diaries (8–13), 24-h recalls (24HR)6 (6,14–18), and FFQ (5,19,20). The factor most consistently related to low energy reporting is obesity, measured by BMI or body composition, with people who are most obese the ones most likely to be low energy reporters (LER) (6,8,12,16,20–22). Results from studies evaluating the relations between other factors and low energy reporting have been mixed. Some studies have found low energy reporting to be more common among women (6,8,16,20–23) than men, but others have found no gender differences (10,12,24,25). The relation between socioeconomic status and low energy reporting is also unclear, with some studies finding an inverse relation between socioeconomic status and low energy reporting (6,16,22,26–28), and others finding a positive relation (12,18,21), or no relation (25,29).
There are few studies on low energy reporting in African Americans and other ethnic groups. One study among African Americans found the rate of low energy reporting to be lower than among predominantly non-Hispanic white populations (10), perhaps this was due to different social norms associated with body weight and body image, whereas another found that postmenopausal African American women were more likely to report low energy intake than white women (19). Although some studies of low energy reporting included adults over the age of 65 y, and some of these have found higher rates of low energy reporting in older persons (6,12,16,18,20) than in younger participants, few studies (11,30,31) have focused on low energy reporting in the elderly. Thus, there is a need for further investigation of low energy reporting in different ethnic groups and in older adults.
The primary goal of this study was to develop a self-report dietary assessment tool for a low literacy, low-income population of rural older adults. At the completion of 6 monthly 24HR, participants completed 3 interviewer-administered questionnaires concerning their diet over the previous 6 mo. This study compares the rates of misreporting on these different assessment methods, and relates misreporting bias to personal and dietary characteristics of the participants.
|
Subjects and Methods |
|---|
|
TOPABSTRACTIntroductionSubjects and MethodsResultsDiscussionLITERATURE CITED |
|---|
Details of this study are described elsewhere (32). Briefly, the study population consisted of community-dwelling adults, 65 y of age and older, with a high school education or less, and an annual income below 150% of poverty level living in 2 rural North Carolina counties. Participants were recruited from personnel referrals at senior and low-income housing units, senior center and meal site staff, county social service agencies, and interviewers from the research counties. The sample (n = 137) was selected to be equally divided among whites, Native Americans, and African Americans. The Wake Forest University School of Medicine Institutional Review Board approved the study.
The study included participants who completed at least five of six 24HR. Participants were excluded if they reported taking a weight-loss or appetite-stimulating drug, lost or gained 
4% of their body weight over the study period, were missing height and weight because it could not be measured, or reported that they were intentionally trying to lose or gain weight. A particular recall day was excluded from the analysis if the participant had been hospitalized within the last 4 wk, if the participant was fasting on the recall day, if the recall was deemed "unreliable" by the interviewer, or if the participant reported being ill on the recall day.
Data collection
Procedure. Data collection occurred over 3 phases between February 2003 to March 2004 with approximately one-third of the participants participating in each phase. The purpose of collecting the data in phases was to ensure balance among the recalls by season. Each participant completed 6 interviewer-administered 24HR, followed by 3 dietary assessment questionnaires, including a modified Block FFQ, a Picture Sort FFQ (PSFFQ), and Meal Pattern Questionnaire (MPQ). To avoid an order effect, the dietary questionnaires were interviewer administered in random order at least 1 d after the final 24HR, and at least 5 d after the previous questionnaire.
Dietary assessment. The interviewers collected the 24HR data using the Nutrition Data System for Research (NDS-R) software [Nutrition Coordinating Center, version 4.05_(33)] on a laptop computer in the participant's home. Of six 24HR, at least 1, but no more than 2, were collected for weekend days. Food portion posters (2D Food Portion Visual, Nutrition Consulting Enterprises) were used as aids in estimating the portion sizes of foods consumed.
The FFQ used in this study is a validated, semiquantitative version of the modified version of the National Cancer Institute Health Habits and History Questionnaire (33), which queried the previous 6-mo period. The questionnaire included 94 line items, including foods that were added to represent the typical rural southern diet, such as "barbeque" in the pork line item, hush puppies, collard greens, kale, and game meat.
The PSFFQ included the same line items as the FFQ; however, labeled picture cards with a color photograph of each food or beverage category were presented to the participant. After sorting the cards and removing those that were eaten "Never or less than once per month" over the previous 6 mo, participants sorted the remaining cards onto a cardboard tray into the 8 frequency response categories of the FFQ, and then assigned a portion size category (small, medium, large). Interviewers recorded the data onto paper forms for data entry.
The MPQ was based on the diet history format. Six meal categories were queried: breakfast, morning snack, lunch, afternoon snack, dinner, and evening snack. Within each meal category, participants responded to line items from the FFQ that were typical of foods eaten for that meal or snack during the past 6 mo. The line items for all 3 snack categories were identical, as were lunch and dinner.
The interviewer asked the participants 5 questions regarding the questionnaire they just completed (the FFQ, PSFFQ, or the MPQ): how easy or difficult it was to answer the questions, how easy or difficult it was to understand the questions, how enjoyable it was to answer the questions, how interesting or boring it was to answer the questions, and how they felt about the length of time it took to answer the questions. Participants could choose a response from a 4-point scale. For each interview, the interviewer recorded the time it took the participant to complete the questionnaire.
Both the PSFFQ and MPQ were pretested with 4 older adults from a county not included in the study. Based on the comments of these adults and the interviewers, the photograph content, color, and layout of the PSFFQ picture cards were adjusted. The interview process was also adjusted based on this pretest, and further pretesting was done on 4 additional volunteers. The interviewers attended a 2-d training session and were required to submit multiple practice interviews to be certified for data collection. During the study, 5% (n = 35) of the 24HR were audiotaped and reviewed by study personnel to ensure that drift did not occur in interviewer technique and accuracy of data entry. No drift in interviewer technique was detected in any of the quality-control tape-recorded interviews. A comparison of nutrient intakes from the FFQ, MPQ, and PSFFQ to the 24HR was made in a separate analysis (32). The MPQ tended to overestimate median nutrient intakes compared with the FFQ and PSFFQ. Correlations between nutrient estimates from the 24HR and other instruments were generally significant and positive, and similar to results from other studies, although the correlations between the 24HR and FFQ tended to be somewhat lower than those of previous studies (32).
The Block 1998 Dietary Data System software was used for data entry and nutrient analysis of the 3 questionnaires (FFQ, PSFFQ, MPQ). Nutrient and food composition from the 24HR data were calculated with the NDS-R system (version 4.05_33). The mean of the six 24HR was used in all analyses, with the exception of the variety score for the calculation of the Healthy Eating Index (HEI) variables (1999–2000) (34). Because we had 6 recall days and were using the NCC food group system, we had to modify the way the variety score was computed. To determine this score, NCC subgroups from the 5 major food groups (grains, vegetables, fruit, meat, and dairy) were grouped into similar food types, regardless of fat content, resulting in 32 variety categories. Of these, 22 were considered to be essential to a balanced diet, so a maximum of 22 foods (of 0.5 servings) summed across the 6 d was assigned a score of 10, and a score of 5 was assigned a score of 0; values between 5 and 22 were linearly interpolated to a scale of 0–10. The 10 components of the HEI score (grains, vegetables, fruits, milk, meat, total fat, saturated fat, cholesterol, sodium, variety) are weighted equally to compute the total HEI score, which ranges from 0 to 100.
Reported energy intake was categorized using the Goldberg formula (35,36). In this formula, an individual is categorized as a LER [or high energy reporter (HER)] if the ratio of reported energy intake (EI) to basal metabolic rate (BMR) is below (or above) the assumed physical activity level (PAL) times the 95% CI based on within-person variability, represented in terms of the CV. The Mifflin equation (37) was used to calculate BMR. PAL was assumed to be 1.55, which is the WHO value for "light" activity, as recommended by Goldberg (35), and is very close to the mean activity level (PAL = 1.56) for adults 65 y of age from an analysis of 574 free-living subjects with total energy expenditure measured by doubly labeled water (38). The CV was calculated for the 24HR data to be 32.9%. Because we did not have repeat applications of the other questionnaires, the CV proposed by Black (36) were used. The lower 95% confidence limit for the 24HR was 1.00, and the upper 95% confidence limit was 2.40. For the questionnaires, the lower 95% confidence limit was 0.87, and the upper 95% confidence limit was 2.75.
Other measures. At the baseline visit, self-reported ethnicity, education, date of birth, household size, income data, and health status data were collected. Weight was measured at the 1st 24HR and the last 24HR, and height was measured at the 1st 24HR. Height was measured 2 times to the nearest 0.10 cm; if the difference was >0.5 cm between the 2 measures, a 3rd measure was taken, at which point they were averaged. BMI was calculated by dividing the mean of the 2 weights (in kg) by the squared mean height (in m2).
Statistical analysis
Descriptive statistics were used to summarize the demographic and health characteristics of all participants. In univariate analyses comparing the LER to the accurate reporter (AR), continuous variables (age, BMI, health status, participant assessment of questionnaire, time to complete questionnaire, HEI variables) were compared using t tests, and categorical variables (gender, ethnicity, income, education, household size) were compared using chi-square tests.
Multiple logistic regression models were used to model the probability of being a LER (vs. AR) for the 24HR, FFQ, and PSFFQ; ordinal regression was used to predict misreporting status (LER, AR, and HER) for the MPQ. All of the models included gender as a covariate as well as variables with P 
0.25 when considered one at a time. All variables included in the final models had P 0.05. All analyses were performed in SAS (SAS Institute, version 8.2), using P < 0.05 to indicate significance.
The number of foods reported and the corresponding total weight (g) were calculated for each day and then the mean of 6 recall days was computed. The most frequent foods consumed were determined by calculating the number of foods reported most often on all recalls by reporting status. We compared the number of servings of discretionary energy foods, number of foods reported, grams reported, and grams per food reported between LER and AR using t tests.
|
Results |
|---|
|
TOPABSTRACTIntroductionSubjects and MethodsResultsDiscussionLITERATURE CITED |
|---|
1 chronic health condition. Almost half of the participants were obese; only 16% had a BMI below 25 kg/m2.
|
|
|
|
Men were significantly more likely than women to be LER on the FFQ; low energy reporting status did not differ significantly by gender on the MPQ or PSFFQ (Table 4). BMI was significantly correlated with low energy reporting on the FFQ, but not the PSFFQ or MPQ, although the trends observed for these 2 instruments indicated higher LER in those who were overweight or obese than those who were normal weight. Reporting status did not relate significantly to ethnicity, education, age, household income, household size, or self-rated health. The participant assessment of the questionnaire did not differ by misreporting status, with one exception. High energy reporters on the MPQ rated the questionnaire as taking longer, in general, than LER or AR, and they generally took longer (60 min) to complete the questionnaire than LER (50 min) or AR (56 min).
|
|
In the multiple logistic regression model, the following variables were entered to compare LER with AR on all 4 instruments: gender, BMI, the total HEI score, and number of household members. The final model only included the HEI score, with lower HEI scores associated with LER. We also compared HEI subscores of these 2 groups, entering all of the HEI components into the preliminary model. In the final model, the scores for total fat and sodium differed between LER and AR, with higher HEI sodium scores (lower sodium intake) positively associated with the likelihood of LER, and total fat scores negatively associated with the likelihood of being a LER.
In addition to analyzing the HEI, we also considered the intake of discretionary energy, numbers and amounts of foods reported, the specific foods reported, and macronutrient intakes. We found that LER on the 24HR reported 3/d fewer servings of discretionary energy foods compared with accurate reporters (2.6 vs. 5.8, P < 0.01). They also reported fewer servings of grain snacks and salty snacks. We found significant differences between the number of foods and g/d on the 24HR in AR compared with LER, with AR reporting 
4 foods/d more than LER (18.8 vs. 14.4 for males, 17.9 vs. 13.0 for females) and >900 g more food than LER (3221 vs. 2272 for males, 2445 vs. 1518 for females). However, the grams of food reported did not differ significantly by reporting status. We also explored whether the most commonly reported foods on the 24HR differed by reporting status. Seven of the top 10 foods and beverages reported by each group were the same (tap water, white bread, regular coffee, sugar, banana, whole milk, and decaffeinated coffee). LER and AR on the 24HR reported similar percentages of total energy from fat (AR = 35.6%, LER = 35.8%), saturated fat (AR = 11.4%, LER = 11.6%), carbohydrates (AR = 51.1%, LER = 49.9%) and protein (AR = 14.4%, LER = 15.3%); none of these differences were significant.
|
Discussion |
|---|
|
TOPABSTRACTIntroductionSubjects and MethodsResultsDiscussionLITERATURE CITED |
|---|
Similar to other studies (6,8,16,20–22), we found BMI to be related to LER on the 24HR and FFQ, although the association was not significant for the PSFFQ and MPQ. In contrast to other studies that have found women more likely to be LER, women had lower rates of LER than men. This may be because women are more likely to be responsible for cooking in this age and cultural group. Other studies completed with this population indicates that women have the major responsibility for food preparation (40) but that men were forced to assume this role when their wife was incapacitated or died (41). Our data were supportive of this explanation as we found that, when adjusted for the number of persons in the home, the difference in reporting status between men and women on the FFQ became nonsignificant. Additionally, we did not find significant differences in reporting status by ethnicity, in contrast to other studies (10,16,19,42). Factors that are associated with ethnicity, such as lower socioeconomic status, may better explain some of the previous study results that have found differences in reporting status by ethnicity, rather than ethnicity itself.
When assessing LER, it is difficult to distinguish between misreporting of food intake and actual differences in diets. Six 24HR per person to analyze provided us with insight in the differences between diets of a LER and an AR. This was done by analyzing the subscores of the HEI, as well as considering intake of discretionary energy, numbers, and amounts of foods reported, the specific foods reported, and macronutrient intakes. In the analysis of HEI subscores of LER vs. AR on all 4 instruments and on the 24HR (Fig. 1), LER reported fewer servings in all of the food pyramid groups. Additionally, LER reported lower sodium and cholesterol intake, indicating that they may be omitting foods with these nutrients. We also found that LER on the 24HR generally reported fewer servings per day of discretionary energy foods, grain snacks, and salty snacks than AR. One interpretation of these findings is that LER are underreporting across a variety of food groups, including those considered healthy, such as vegetables, as well as discretionary energy foods.
Further support for this hypothesis was provided when considering the mean number of foods, gram weights, and gram weights per food reported on the 24HR. We found significant differences between the number of foods and g/d reported on the 24HR in AR compared with LER, with AR reporting more foods/d than LER and >900 g more food than LER. However, the grams reported per food did not differ by reporting status, which suggests that LER are reporting similar amounts of foods consumed when they are reported, but are often omitting entire foods. These results are consistent with other studies that have found LER reporting fewer eating occasions per day (24,43,44) and with a study that found rural elderly reporting lower intakes of foods that are perceived as healthful and unhealthful (31). Although we did not do an extensive analysis of dietary patterns, we explored whether the most commonly reported foods on the 24HR differed by reporting status. Seven of the top 10 foods and beverages reported by each group were the same and it was difficult to discern any pattern among the foods and beverages that differed (grits, 2% milk, sausage for LER; artificial sweetener, tea, cream for AR), although the differences may indicate that LER could be omitting items that may be added to coffee.
Our research highlights the potential impact of LER on using the HEI to assess dietary quality. It is possible that only reported dietary quality, not actual dietary quality, differs between LER and AR when the HEI is used. Macronutrient analysis indicated that the composition of the diets of LER and AR did not differ. Thus, the differences in the HEI may not reflect true differences in dietary quality between AR and LER, but appear to reflect the underestimation of all food groups by LER. Our results for this study population contrast with other studies that have found that LER reported a lower percentage of energy from fat than AR (6,10,13,22,24,29), suggesting restrained eating and perhaps a desire to omit the reporting of "unhealthy" foods. Because many food groups appear to be omitted by this population, it may be that other factors, such as memory problems, are influencing LER, rather than a social desirability bias that has been found in other studies (24,45).
The dietary questionnaires were surprisingly similar with respect to the levels of LER, with 2 exceptions. Men performed most poorly on the FFQ, compared with the other 3 tools, whereas the level of accurate reporting was similar for women on all 4 tools. The other notable exception was that there was a higher degree of HER on the MPQ, particularly for women. HER is unusual on dietary questionnaires. It may be that the diet history format of this questionnaire resulted in HER. This format also led to this being the lengthiest of the 3 questionnaires, taking 10 min longer to complete than other questionnaires, which took 45 min each. The participants did not rate the MPQ more difficult or less enjoyable than the other 2 questionnaires; however, they did accurately perceive it to be longer.
Limitations of this study warrant mentioning. These findings cannot be easily generalized to other populations because the study was specifically conducted among rural, southern, and low socioeconomic status older adults. However, exploring the accuracy of reporting in this particular population adds to the growing body of evidence that one dietary assessment technique does not fit every population. This is particularly true for this population, which we found had high levels of LER on a series of six 24HR. Although it is difficult to postulate how the findings from this study might have differed in a different study population, the current literature is consistent with higher levels of LER in low socioeconomic status populations. We would anticipate more reporting errors for this population due to their low levels of educational attainment and literacy if the questionnaires had been self-administered rather than administered by an interviewer.
Additionally, we relied upon the Mifflin equation and the Goldberg formula to assess LER, both of which may have led to the misclassification of some participants as LER. It is difficult to know how accurate these equations are in this population, as there have been few validation studies of equations to predict basal metabolic rate in the elderly or ethnically diverse populations. We chose the Mifflin equation because it is most appropriate for use in the elderly (46). Furthermore, we chose a PAL value to use in the Goldberg formula that was appropriate for the age group in this study (35). Ideally, we would have preferred to use an intake biomarker for energy expenditure, such as doubly labeled water, but that was not possible in this research setting.
Due to the importance of dietary assessment for understanding health and health-related quality of life, particularly among older adults and those who experience health disparities (minority, rural, poor), the need for reliable tools remains. This analysis indicates that existing tools remain problematic, particularly among segments of the population for which accurate tools can be most beneficial. Continued effort is required to develop new dietary assessment methods, and to validate these methods in vulnerable populations, such as the low socioeconomic status rural elderly.
|
FOOTNOTES |
|---|
2 Author disclosures: J. A. Tooze, no conflicts of interest; M. Z. Vitolins, no conflicts of interest; S. L. Smith, no conflicts of interest; T. A. Arcury, no conflicts of interest; C. C. Davis, no conflicts of interest; R. A. Bell, no conflicts of interest; R. F. DeVellis, no conflicts of interest; and S. A. Quandt, no conflicts of interest.
6 Abbreviations used: 24HR, 24-hour recall; AR, accurate reporter; HEI, healthy eating index; HER, high-energy reporter; LER, low energy reporter; MPQ, meal pattern questionnaire; PAL, physical activity level; PSFFQ, picture sort food frequency questionnaire.
Manuscript received 1 November 2006. Initial review completed 31 December 2006. Revision accepted 2 March 2007.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTIntroductionSubjects and MethodsResultsDiscussionLITERATURE CITED |
|---|
1. Elia M, Ritz P, Stubbs RJ. Total energy expenditure in the elderly. Eur J Clin Nutr. 2000;54: Suppl 3:S92–103.[Medline]
2. Ritz P. Factors affecting energy and macronutrient requirements in elderly people. Public Health Nutr. 2001;4:561–8.[Medline]
3. Blumberg J. Nutritional needs of seniors. J Am Coll Nutr. 1997;16:517–23.[Abstract]
4. Kipnis V, Subar AF, Midthune D, Freedman LS, Ballard-Barbash R, Troiano RP, Bingham S, Schoeller DA, Schatzkin A, Carroll RJ. Structure of dietary measurement error: results of the OPEN biomarker study. Am J Epidemiol. 2003;158:14–21.
5. Subar AF, Kipnis V, Troiano RP, Midthune D, Schoeller DA, Bingham S, Sharbaugh CO, Trabulsi J, Runswick S, et al. Using intake biomarkers to evaluate the extent of dietary misreporting in a large sample of adults: the OPEN study. Am J Epidemiol. 2003;158:1–13.
6. Briefel RR, Sempos CT, McDowell MA, Chien S, Alaimo K. Dietary methods research in the third National Health and Nutrition Examination Survey: underreporting of energy intake. Am J Clin Nutr. 1997;65:1203S–9S.[Medline]
7. 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;281:891–9.
8. Johnson RK, Goran MI, Poehlman ET. Correlates of over- and underreporting of energy intake in healthy older men and women. Am J Clin Nutr. 1994;59:1286–90.
9. Taren DL, Tobar M, Hill A, Howell W, Shisslak C, Bell I, Ritenbaugh C. The association of energy intake bias with psychological scores of women. Eur J Clin Nutr. 1999;53:570–8.[Medline]
10. Tomoyasu NJ, Toth MJ, Poehlman ET. Misreporting of total energy intake in older African Americans. Int J Obes Relat Metab Disord. 2000;24:20–6.[Medline]
11. Tomoyasu NJ, Toth MJ, Poehlman ET. Misreporting of total energy intake in older men and women. J Am Geriatr Soc. 1999;47:710–5.[Medline]
12. Lafay L, Basdevant A, Charles MA, Vray M, Balkau B, Borys JM, Eschwege E, Romon M. Determinants and nature of dietary underreporting in a free-living population: the Fleurbaix Laventie Ville Sante (FLVS) Study. Int J Obes Relat Metab Disord. 1997;21:567–73.[Medline]
13. Goris AH, Westerterp-Plantenga MS, Westerterp KR. Undereating and underrecording of habitual food intake in obese men: selective underreporting of fat intake. Am J Clin Nutr. 2000;71:130–4.
14. 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]
15. Johnson RK, Soultanakis RP, Matthews DE. Literacy and body fatness are associated with underreporting of energy intake in US low-income women using the multiple-pass 24-hour recall: a doubly labeled water study. J Am Diet Assoc. 1998;98:1136–40.[Medline]
16. Krebs-Smith SM, Graubard BI, Kahle LL, Subar AF, Cleveland LE, Ballard-Barbash R. Low energy reporters vs others: a comparison of reported food intakes. Eur J Clin Nutr. 2000;54:281–7.[Medline]
17. Myers RJ, Klesges RC, Eck LH, Hanson CL, Klem ML. Accuracy of self-reports of food intake in obese and normal-weight individuals: effects of obesity on self-reports of dietary intake in adult females. Am J Clin Nutr. 1988;48:1248–51.
18. Zhang J, Temme EH, Sasaki S, Kesteloot H. Under- and overreporting of energy intake using urinary cations as biomarkers: relation to body mass index. Am J Epidemiol. 2000;152:453–62.
19. Hebert JR, Patterson RE, Gorfine M, Ebbeling CB, St JS, Chlebowski RT. Differences between estimated caloric requirements and self-reported caloric intake in the women's health initiative. Ann Epidemiol. 2003;13:629–37.[Medline]
20. Johansson L, Solvoll K, Bjorneboe GE, Drevon CA. Under- and overreporting of energy intake related to weight status and lifestyle in a nationwide sample. Am J Clin Nutr. 1998;68:266–74.[Abstract]
21. Hirvonen T, Mannisto S, Roos E, Pietinen P. Increasing prevalence of underreporting does not necessarily distort dietary surveys. Eur J Clin Nutr. 1997;51:297–301.[Medline]
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. de Vries JH, Zock PL, Mensink RP, Katan MB. Underestimation of energy intake by 3-d records compared with energy intake to maintain body weight in 269 nonobese adults. Am J Clin Nutr. 1994;60:855–60.
24. Tooze JA, Subar AF, Thompson FE, Troiano R, Schatzkin A, Kipnis V. Psychosocial predictors of energy underreporting in a large doubly labeled water study. Am J Clin Nutr. 2004;79:795–804.
25. Heitmann BL. The influence of fatness, weight change, slimming history and other lifestyle variables on diet reporting in Danish men and women aged 35–65 years. Int J Obes Relat Metab Disord. 1993;17:329–36.[Medline]
26. Pryer JA, Vrijheid M, Nichols R, Kiggins M, Elliott P. Who are the ‘low energy reporters’ in the dietary and nutritional survey of British adults? Int J Epidemiol. 1997;26:146–54.
27. Stallone DD, Brunner EJ, Bingham SA, Marmot MG. Dietary assessment in Whitehall II: the influence of reporting bias on apparent socioeconomic variation in nutrient intakes. Eur J Clin Nutr. 1997;51:815–25.[Medline]
28. Kristal AR, Feng Z, Coates RJ, Oberman A, George V. Associations of race/ethnicity, education, and dietary intervention with the validity and reliability of a food frequency questionnaire: the Women's Health Trial Feasibility Study in Minority Populations. Am J Epidemiol. 1997;146:856–69.
29. Johansson G, Wikman A, Ahren AM, Hallmans G, Johansson I. Underreporting of energy intake in repeated 24-hour recalls related to gender, age, weight status, day of interview, educational level, reported food intake, smoking habits and area of living. Public Health Nutr. 2001;4:919–27.[Medline]
30. Shahar D, Shai I, Vardi H, Fraser D. Factors associated with low reported energy intake in the elderly. J Nutr Health Aging. 2005;9:300–4.[Medline]
31. Bailey RL, Mitchell DC, Miller C, Smiciklas-Wright H. Assessing the effect of underreporting energy intake on dietary patterns and weight status. J Am Diet Assoc. 2007;107:64–71.[Medline]
32. Quandt SA, Vitolins MZ, Smith SL, Tooze JA, Bell RA, Davis CC, DeVellis RF, Arcury TA. Comparative validation of standard, picture-sort, and meal-based food frequency questionnaires adapted for an elderly low socioeconomic status population. Public Health Nutr. 2006; in press.
33. Mayer-Davis EJ, Vitolins MZ, Carmichael SL, Hemphill S, Tsaroucha G, Rushing J, Levin S. Validity and reproducibility of a food frequency interview in a Multi-Cultural Epidemiology Study. Ann Epidemiol. 1999;9:314–24.[Medline]
34. Basiotis PP, Carlson A, Gerrior SA, Juan WY, Lino M. The healthy eating index: 1999–2000. U.S. Department of Agriculture, Center for Nutrition Policy and Promotion; 2002. Report No.: CNPP-12.
35. 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 identify under-recording. Eur J Clin Nutr. 1991;45:569–81.[Medline]
36. Black AE. Critical evaluation of energy intake using the Goldberg cut-off for energy intake:basal metabolic rate. A practical guide to its calculation, use and limitations. Int J Obes Relat Metab Disord. 2000;24:1119–30.[Medline]
37. Mifflin MD, St Jeor ST, Hill LA, Scott BJ, Daugherty SA, Koh YO. A new predictive equation for resting energy expenditure in healthy individuals. Am J Clin Nutr. 1990;51:241–7.
38. Black AE, Prentice AM, Coward WA. Use of food quotients to predict respiratory quotients for the doubly labelled water method of measuring energy expenditure. Hum Nutr Clin Nutr. 1986;40:381–91.[Medline]
39. Gauthier S, Reisberg B, Zaudig M, Petersen RC, Ritchie K, Broich K, Belleville S, Brodaty H, Bennett D, et al. Mild cognitive impairment. Lancet. 2006;367:1262–70.[Medline]
40. Quandt SA, McDonald J, Arcury TA, Bell RA, Vitolins MZ. Nutritional self-management of elderly widows in rural communities. Gerontologist. 2000;40:86–96.[Abstract]
41. McDonald J, Quandt SA, Arcury TA, Bell RA, Vitolins MZ. On their own: nutritional self-management strategies of rural widowers. Gerontologist. 2000;40:480–91.
42. Caan BJ, Flatt SW, Rock CL, Ritenbaugh C, Newman V, Pierce JP. Low-energy reporting in women at risk for breast cancer recurrence. Women's Healthy Eating and Living Group. Cancer Epidemiol Biomarkers Prev. 2000;9:1091–7.
43. Kant AK. Nature of dietary reporting by adults in the third National Health and Nutrition Examination Survey, 1988–1994. J Am Coll Nutr. 2002;21:315–27.
44. Huang TTK, Roberts SB, Howarth NC, McCrory MA. Effect of Screening Out Implausible Energy Intake Reports on Relationships between Diet and BMI. Obes Res. 2005;13:1205–17.[Medline]
45. Hebert JR, Clemow L, Pbert L, Ockene IS, Ockene JK. Social desirability bias in dietary self-report may compromise the validity of dietary intake measures. Int J Epidemiol. 1995;24:389–98.
46. Frankenfield D, Roth-Yousey L, Compher C. Comparison of Predictive Equations for Resting Metabolic Rate in Healthy Nonobese and Obese Adults: A Systematic Review. J Am Diet Assoc. 2005;105:775–89.[Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
