Dietary Assessment Methods

Dietary Assessment Aim

• Establish the extent to which nutritional needs are being met

Dietary Assessment Applications

• Individual assessments in clinical situations
• Estimate intakes of populations and establish associations between nutrition and disease in epidemiology which is then used for developing dietary guidelines and health policy
• Used in research

Ideal Assessment Method

• Low burden on respondents and investigators
• Rapid and cheap
• Accurate, precise data with minimal measurement error

Levels of Dietary Assessment

• National
• Household
• Individual

National Level Assessment

• Food balance sheets can be used
• A comprehensive food supply list for each specific country for a specific time

Household Level Methods

• Food account method
• Inventory method
• Household record method
• List recall methods

Individual Dietary Assessment Factors

• Current food and fluid intake
• Increase or decrease in appetite and duration and severity of these changes
• Presence of factors affecting food and fluid intake
• Alterations in meal pattern or timings
• Changes in food choice or food consistency

Dietary Intake Measurement Types

• Retrospective measures (24-hour recalls, diet history, food frequency questionnaires (FFQs))
• Current measures (weighed or estimated food records)
• The quantity of food consumed may be weighed directly or estimated using household measures (cups, spoons, photos, standard units, food models/replicas), particularly in clinical practice

Nutrient Measurement Duration

• The number of days of report varies depending on the day to day variability of nutrient consumption
• Three days of record (one weekend day) is optimal for estimating energy intake (Ma et al., 2009)
• Nutrients with greater variability (alcohol, vitamin C, riboflavin, iron) require between 4 and 14 days of records
• Fewer days of data collection are required to capture habitual diet in children (Lanigan et al., 2004)

24-Hour Recall

• Determines intake during the preceding 24 hours recorded on paper or using interactive computerised software (Slimani et al., 1999)
• Day to day variability in nutrient intake is large and a single day will not categorise individuals correctly within a distribution of intake
• Particular issues are the capture of infrequently eaten foods (fish, liver)
• Single 24 HRs are better used for group assessments than estimates for individuals
• Multiple 24 HRs can be used to overcome this problem

Diet History

• An interview to establish usual eating pattern over a 1-week period, followed by a frequency questionnaire
• Interview administered only
• Provides a representative pattern of usual intake
• Traditionally used in clinical practice taking up to 1.5 hours to complete

Food Frequency Questionnaire (FFQ)

• Consists of a list of specific foods or food types associated with frequency of consumption to record average consumption over the previous month or year
• Semi quantitative if portions are included
• Provides estimates of habitual intake
• Widely used in nutritional epidemiology
• Needs to be specific to a population group to ensure coverage of important foods
• The population must be literate and numerate to calculate relative frequencies
• Factors influencing response are length and complexity of the food list and current diet

FFQ Advantages

• Moderate burden
• Postal survey can be done
• Large number of participants can be used

FFQ Disadvantages

• Must be population specific
• Poor precision
• Dependent on memory

Weighed Food Records

• All food and drink consumed is weighed and recorded (food type, method of preparation) over a period of days
• Portable scales need to be supplied
• Leftover food should be weighed and deducted
• A recommended time period for records is 4-7 days or more weekend days

Weighed Intake Advantages

• Good information on individual intake
• Meal pattern recognized

Weighed Intake Disadvantages

• Burden on subject
• Compliance is low
• Limited time period taken, the week might be an exceptional one

Estimated Food Records

• All foods consumed over a period of days are recorded, with details of food type and method of preparation and estimated (not weighed) portion sizes using spoons, cups etc.
• Better compliance than weighed record
• Greater error than weighed

WFR and EFR

• High respondent burden
• Need cooperative, literate respondents
• Respondents require training in the level of detail needed to describe foods
• Respondents may change usual eating patterns to simplify the process of the record
• The method may also be tedious

Duplicate Sample Method

• Duplicate samples of all foods consumed are made and the nutrient content analysed at the lab
• Provides greater accuracy
• Feasible for research centres and metabolic units only
• Expensive, time consuming and needs lab technical support

Method Choice (Clinical Practice)

• Rapid assessments of nutritional intake is needed in order to prescribe dietary change or to improve nutritional status
• Traditionally, 24 HRs of usual intake or diet histories have been used
• Food frequency questionnaires and weighed or estimated food records are not generally used due to increased respondents' burden and limit time available

Method Choice (Research Purposes)

• FFQs to multiple 24 HRs can be used (method depends on the study/resources/participants/time)

Structured 24 Hour Recall

• Follow a structure to help the client recall what he ate
  • Breakfast
  • Mid-morning
  • Lunch
  • Mid-afternoon
  • Evening meal
  • Other foods eaten
• Include the food and drinks, their description and amount

Diet Recall Advantages

• Can be done in a single short interview

Diet Recall Disadvantages

• Depends on memory
• Estimated portion size
• Is the last 24hrs representative

Supplements

• Sources of nutrients
• Vitamin and mineral supplements can contain considerable amounts of nutrients
• Should be taken into account when assessing total nutrient intakes

Errors in Dietary Assessment

• Sources of measurement error are many and may occur both during data collection and processing
• Bias is the modification of a method of measurement by a factor that influences the measurement in one or more directions
• Measurement error may consist of one or more types of error
• Interviewer bias, recall bias and social desirability bias have been identified as different types of measurement errors