24-hour dietary recall and food record method

1
24-hour dietary recall and food record method

• Chap 4 of Nutritional epidemiology
• By WALTER WILLETT

2
Introduction

• 24-hour dietary recall and food record method
  --based on foods and amounts actually consumed by
  an individual on one or more specific days
• Food frequency questionnaires(FFQ) and diet
  historiesbased on an individuals perceptions of
  usual intake over a less precisely defined period
  of time

3
Data collection methods

• 24-hour dietary recall method
• --an in-depth interview conducted by a trained
  dietary interviewer
• --Intervieweessubjects, parent,caretaker
• --The dietary interviewer solicits detailed
  information about everything the subject had to
  eat and drink from midnight to midnight to
  midnight of the previous day or over the 24-hour
  period
• --The importance of a skilled interviewer
• --Dependent on short term memory

4

• Food preparation methods, recipe ingredients,
  brand name of commercial products, use of dietary
  supplements
• Interviewer(important factors in obtaining
  complete and accurate information) nonjudgmental
  manner, a neutral attitude, use open-ended
  questions, avoid asking questions in a manner
  that might influence the subjects responses

5

• Accurate quantification of amounts of
  foodsamount estimation tool food containers,
  photographs, geometric shapes,number
• Using computer software
• The Minnesoda Nutrition Data System (NDS)
• Currently the most commonly used method for
  dietary surveys in the U.S.
• Face to face, telephone interview(saving
  traveling time, surprise effect)

6

• Food record method
• --food intake is recorded by the subject(or
  observer) at the time the foods are
  eaten-minimize reliance on memory
• --Intensive training for the subjects
• --The fields worker weighs the raw ingredients,
  as well as the individual portions of the cook
  dishesdetermine individual food intakes
• --Food quantifying tools
• --Carefully reviewed by a trained nutritionist

7
Strengths and limitations of the 24-hour dietary
recall and food record methods

• Strength (both methods)
• - based on actual intake to estimate absolute
  amount rather than relative amount of nutrients.
• - open endedhigh level of specificity
• (interviews should be sensitive to cultural
  difference)
• Limitation( both method)
• Day to day variation

8

• Strength (recall vs. food record)
• - recall does not need literacy
• - recall is less likely to alter eating
  behavior
• - relatively minimal respondent burden
• Limitation(recall vs. food record)
• - reliance on memory
• - the need for a highly trained interviewer
• Limitation(food record)
• -High level of subject motivation

9
Number of days and which days

• For group means---single day
• For the distribution of individual intake within
  the group, or relationship between individual
  intake and other variablesmore than one day(The
  number of days is dependent on the studied
  nutrient)
• Work days vs. Weekend days
• combination of days of the week and not
  consecutive days
• Estimate of long term intakeseasonal effect
• For micronutrients such as vitamin A, Cfood
  frequency questionnaire

10
Reducing error in data collection

• Factors affecting data from dietary recall Age,
  gender, intelligence, mood, attention,
  consistency of eating pattern
• Comparison of mean energy intake(Recall vs. food
  record) Table 4-1

11
Reducing error in data collection

• Reducing error from dietary recall
• Considerable training and practice of the
  interviewers.
• providing a relaxed and unhurried atmosphere for
  the subject.
• 3. asking about the previous days activities
  to help jog the subjects memory
• 4. providing a list of foods
• 5. to use an automated system for collecting the
  data in which the computer provides all of the
  prompts for describing foods at the appropriate
  level of detail.

12
-Back-
13

• Reducing error from food record
• 1. Subjects motivation
• 2. careful training of subjects in methods of
  keeping accurate records
• 3. emphasizing the importance of the subjects
  contribution to the research
• 4. Stressing the need for timely recording of
  food intake
• 5. Encouraging subjects to maintain usual eating
  habits during the recording period(for ease of
  recording, social desirability).
• 6. Providing written examples
• 7. to review the food records by trained
  nutritionist ASAP.

14

• Reducing error from both methods
• 1. Between meals snacks
• 2. Eating out details
• 3. Quantification of portion sizes
• Food weighing devices, photographs or food
  models of more than one portion size,models of
  geometric shapes, training of subjects in the use
  of food models(Table 4-2),
• 4. For telephone interview, tools(ex ruler,
  standard measuring cups or spoons, or a notebook
  of photographs of different portion sizes for
  selected foods) can be provided to each subjects
  to keep at home.

15

• Reducing error from both methods
• 5. Improvement of tools two-dimensional visuals
  portraying various sizes of geometric
  shapes,mounds, and beverage containers to
  facilitate amount estimation.
• 6. Inducing motivation
• a. convincing participants the importance of
  the research and the critical role they play
• b. Monetary or other types of incentives

16
-Back-
17
-Back-
18
Analysis of food intake data

• At individual level, by food group, by meal
  pattern(the investigation of the effects of foods
  when eating together)
• Calculation of nutrient intakes(database of food
  composition)
• The degree of specificity of food description
  detail required at the data collection level of
  coding and nutrient calculation.

19
Sources of error in nutrient calculation

• Inaccuracies in the coding of food intake
  data.data entry check of the computer system,
  intensive training and certification of coders,
  10 or 20 duplicate coding.
• Accuracy and representitive of the nutrient
  values in food composition databases.
• The use of outdated nutrient data
• Analytical methodology

20
Validation of 24-hour recalls and food records

• Reporting accuracy
• Accuracy of nutrient calculations
• Accuracy of assessing usual intake
• (how well the selected days of intake represent
  usual individual intake)

21

• Reporting accuracy
• 1. omissions of food items are more common
  than additions. (side dishes or
  add-ons)(Figure4-1)
• 2.overestimation appears to be more frequent
  than underestimation for portion sizes (Table
  4-4). (Beveragesbest estimation, add-ons-poorest
  estimation)
• 3. The underreporting of energy intake
  observed in validation studies of records and
  recalls is more likely due to omissions in
  reporting foods than to underestimate of portion
  sizes.
• 4. The greatest difficulty in estimating
  amounts was associated with items that could not
  be separate visualized.
• .

22

• Reporting accuracy
• 5. Overall, recalls tend to underestimate intake
  by about 10 compared with observed intake(vary
  considerably among individuals).
• 6. Flat slope syndrome
• Overestimaton of portion sizes tended to be
  greater among those who ate smaller portions and
  underestimamtion by those who ate larger
  portions.

23
-Back-
24
-Back-
25

• Accuracy of nutrient calculations
• 1. (Table 4-5) calculated values tend to be
  higher than the analyzed values.

26

• Accuracy of assessing usual intake
• 1. The use of biochemical indicator(Ch9)
• 2. Underreportingunderestimationobesity,athlet
  e,education,motivation,age (adolescent,elderly)---
  nutrient intakes expressed as a percentage of
  total energy intake have not been found to be
  biased,even when energy intake itself was
  under-reported)
• 3. Reproducibility
• day-to-day variation, seasonal difference
• ---reproducible at the group level, averaging of
  multiple days of intake at the individual level

27

• There are limitations and potential sources of
  error of dietary recalls and food records.---- be
  careful when using these methods to validate
  food-frequency questionnaires or diet histories
  designed for estimating long term usual dietary
  intake(CH 6).

28
-Back-
29
Summary Foods records and 24-hour recalls

1. consist of specific foods consumed by an
   individual on one or more days
2. greater specificity for describing foods and food
   preparation methods, greater flexibility for
   analyzing the data
3. seldom used as the primary method for estimating
   usual intake in large scale epidemiologic
   research
4. important in describing mean values for groups
   and in validating food frequency questionnaires
   (most commonly used method)