4. Dietary assessment methods

Name 3 retrospective dietary assessment methods

• 24-hour recall → can also be used as prospective

• Food Frequency Questionnaire (FFQ)

• Dietary history

what is retrospective

a review of past events, actions, or completed work

what is prospetive

likely to happen, expected in the future, or referring to someone who wants to become something

name 2 prospective dietary assessment methods

• Duplicate diet approach 

• Food record (weighed or estimated) → can also be used as retrospective

name 2 Measures of usual intake

• Food Frequency Questionnaire 

• Dietary history

name 3 Measure of actual intake

• Duplicate diet approach 

• 24-hour recall

• Food record (weighed, analyzed by AI from photos or estimated)

what are actual and usual intake

Actual intake = what the person ate on a given day

Usually intake = what the person normally eats (on average, estimated)

what are the Characteristics of Dietary Assessment Methods

• Timeline: cross-sectional or longitudinal (prospective or retrospective)

• Appropriateness of the method and technology used to the study design applied

  • Observational studies

    • Cross-sectional (and ecological) studies 

    • Case-control studies 

    • Cohort studies 

  • Intervention studies 

• Methodology: different methodologies of dietary assessment methods for the same type (e.g. paper format, web-based) for observational studies or inclusion of external observers

what are surveys

• Quantitative science (=the systematic study of the physical, natural, or social world using numerical data, mathematical models, and statistical analysis)

• Survey research may be informed by qualitative research, previous quantitative evidence, theory or exploratory studies 

• Surveys try to connect factors of exposure to a certain outcome

• Factor and outcome can belong to nominal, ordinal, interval and ratio type of data 

• Statements on significant differences / correlations / associations ALWAYS need to be backed up by an appropriate statistical test (gives effect size, p value)

how do you set up a questionaire

1. inform study participant about objectives of the study (why performed, why I’m part, why input is relevant, goal of study)

2. develop system for anonymous but unique identifier system 

3. make clear that participation is voluntary 

4. instruct on how to fill in

what are advantages of open questions

• answer is in own terms 

• respondents reflect upon question 

• perception of question is documented

• may open new research questions (if the Q is interpreted different)

• may be used as basis for closed Q’s in subsequent surveys

what are disadvantages of open questions

• more time-consuming for respondent and administers 

• coding more demanding

• coding rather subject to interpretation 

• some may not be codable at all

what are pre-defined alternatives and for what type of question do you use this

Instead of open Q’s use pre-defined alternatives (how long does homework take you - <1h 1-1.9h 2-2.9h etc.). And when asking for agreement/disagreement use Likert scales.

what is the likert scale and how does it work

• Example:

completely disagree - mostly disagree - slightly disagree - slightly agree - mostly agree - completely agree 

• measures the attitude to the various likert items 

• interpret the intensity or strength of experience being projected on ordinal scale 

• 2-7 levels 

• usually odd-numbered scales because they give a midpoint 

• the statements are given in ‘I / we’ form 

• Opposite meaning statements need to be inverted (wat je deed bij het experiment van Rob met Carmen en Chris) 

• Coherence of the statements of all participants can be tested with a particular test (evtl. cornbach’s alpha >0.7) 

what are the criteria of a satisfactory scale (7)

• Appropriate

  • keep in mind: does the variable represent the parameter necessary to answer the research question 

• Practicable 

  • linked to the methods used for collecting the data 

• Powerful

  • linked to the type of scale 

    • ordinal scale should be used rather than a nominal one 

    • numerical scale rather than an ordinal one 

• Clearly defined

  • operational definitions are obligatory for variables and categories 

    • value has ONLY the precision (cijfers achter de komma) of the least precise measurement 

    • calculate with one digit more than your measurement allows 

• Sufficient number of categories

  • avoid compression of data into too few categories 

• Collectively exhaustive 

  • provide sufficient classification to place EVERY subject 

• Mutually exclusive 

  • each item of information fits in only one place along the scale (you can’t to a scale from 70-80 and 80-90 because 80 fits now in both categories)

advantages of closed questions

• time- and work-efficient 

• easy to answer and process

• easy to code

• higher comprehensibility (the quality or state of being easy or possible to understand) of Q’s

• less prone to bias

disadvantages of closed questions

• loss of spontaneity / out-of-the-box answer possibility 

• categorization may be misunderstood

• respondents may not find an applicable category

what are Structured questionnaires

• Almost exclusively closed Q’s / limited in number 

• Easy to evaluate - machine readable 

• Used in large studies with high (>100-200) participants 

• Preferred in satisfaction surveys in business

what are Semi-structured questionnaires

• Mixed in closed and open Q’s

• More demanding in evaluation 

• Used in large research projects 

what are Unstructured questionnaires 

• mostly open Q’s

• tries to identify thinking patterns 

• may be used to develop new theories based on repeated patterns of individuals approaches and way of phrasing attitudes

what do you do before you collect the data

create a codebook

• Fix precisely how to code each answer for your data evaluation 

• Generate a large table that harbors all results 

• Arrange the tables (participants = rows / answers = columns)

• Logically opposite statements have to be inverted during coding

what is the Healthy Eating Index (HEI)

score for consumption of health-relevant food (groups)

• A measure of overall eating “quality” independent of quantity 

• Designed to reflect the food-based dietary guidelines 

• Scoring system with density-based standards allows study of diet quality independent of quantity applies across all levels of the food system

• Considers gaps between intakes and recommendations

•  Balances the balance (ratio) among foods

• High scores rate, with potential for improvement 

• Takes amount of money that a food is in account

• Is as simple as possible 

• Works with the simplest-possible algorithm

• Covers changes in needs over the lifespan 

• Can predict health outcome 

what is the Duplicate diet approach 

• Each portion is produced in duplicate, one portion is eaten, the other analysed 

• Use: 

  • Small studies where precision is important 

  • When analytes are to be determined that are not documented well (e.g. absorption of pesticides) 

  • Use only if you have a high-performance analytical lab

strengths of the Duplicate diet approach 

• Precise data

• free choice of analytes

• data pattern or analysis is not affected by regular, repetitive, and predictable fluctuations that happen at specific times of the year, month, or week

limitations of the Duplicate diet approach 

• Time-consuming 

• Costly

• Applicable only for a small number of study participants

what is (External) food consumption record

• All (self-determined or instructed) foods consumed are recorded by a nutritionist

• Use

  • Small studies where precision is important 

  • When controlled design of study is necessary 

  • To establish/verify another dietary intake method

strengths of (External) food consumption record 

• Reliable data

• Collected by experienced staff

limitations of (External) food consumption record 

• Possibly intimidating (subjects are never alone when preparing food)

• Costly

• Only usable for a small amount of study participants 

• Bias caused by observation probable

what is 24-hour recall

• Nutritionists interview people about their food intake during the previous 24 hours

• Use 

  • Single day collection → provide information about actual intake and usual intake 

  • Multiple day collection → provide information about common intake 

  • Number of days needed depends on the nutrient you want to investigate, seasonal variation and the study size

what are limitations and common issues of 24-hour recall

• Imprecise (estimations)

• Recall bias 

• Common issues → avoiding common issues 

  • Interviewer is biased/has presumptions → ask about details (DO NOT assume)

  • Only using closed questions and this may miss important information → use open questions 

  • Misinterpretation of portion sizes → use photos of portion size or do the interview at home so that you can see the plate 

  • Respondents may have forgotten items (recall bias) → ask what they did / doing it in the respondents house brings up the memory

what is Dietary records / food records / food diaries 

• People record their food intake for a given period (balance / standards)

• Two types: 

  • Estimated food record:

  • Weighed food record:

• Use

  • A (repeated) single day collection provides information about the respondent’s actual intake and gives information about the usual intake of a population 

  • Consecutive multiple day collection can provide information about the respondent’s usual intake - both weekdays and weekends should be included 

  • The number of days needed depends on the nutrient you want to investigate, seasonal variations and the study population

• When doing it: do not change food habits, write down everything, immediately after consuming, weigh food but do the edible part, be specific, describe preparation, remember added fat/salt/sugar during cooking

what are issues with Dietary records / food records / food diaries 

• Changing usual food intake 

• Forgetting snacks/drinks

• Including non-edible parts 

• Dining out

• Misidentification of food

what is Estimated food record

recording food and beverages in household measures 

what is Weighed food record

all food and beverage consumed should be weighed

what is Dietary history 

• Nutritionists interview individuals about their usual (averaged) food intake over the past few months on an average basis (longer period)

• Use

  • Data is collected by going step by step through the “average” day 

  • Sequence of questions can be standardized to guarantee reproducibility and to help avoiding that any question will be forgotten by the interviewer 

  • Questionnaire or face-to-face interview 

limitations of Dietary history 

• Imprecise 

• Recall bias

what is Food Frequency Questionnaire

• Questionnaire about how frequently people consume certain foods (portion size estimation) with the special form of screeners (on an average basis)

• Use

  • Form is organised due to typical meal consumption structure (culture-sensitive): breakfast, snack, lunch, snack, dinner, anything else?

  • Portion size photos can be used to achieve a good estimation of portion sizes

limitations of Food Frequency Questionnaire

• Imprecise (estimation)

• Recall bias

• Limited quantitative evaluation