If you are assessing "typical" or "usual" diet:

• Ensure birthdays and other celebrations are avoided during the recording period

• Ensure subjects report everything they eat and drink.
  People tend to under-report their intake, so it is very important that records are carefully checked for completeness. Items most often omitted are spread on bread, drinks such as cups of tea/coffee, snack foods and items eaten away from home. Alcoholic drinks may be omitted, particularly if subjects are concerned that their consumption may be criticised.

• Check that sufficient detail has been recorded about the foods and drinks.
  In order to calculate nutrient intakes, you need to be able to allocate a code number to each item. For example, "bread roll" is not sufficient. You need to know whether it was white, brown or wholemeal and whether it was soft or crusty. "Chop" is not sufficient. You need to know whether this was lamb or pork, which type of chop it was and whether the fat was eaten. If the record is weighed, you need to know whether the weight included the bone. Ideally, the bone should also be weighed.

• Weighed record: ensure the weights are "reasonable"
  Unweighed record: ensure the amounts consumed are described as well and as accurately as possible.
  
  

  The following examples show the amount of detail needed:

  1. DESCRIPTION OF FOODS

  Describe each food item separately. Give as much detail as possible:-
  e.g. bread - white/brown/wholemeal
  e.g. fat spread - % fat, polyunsaturated/olive oil/ordinary, brand
  e.g. grilled/fried/boiled/baked/raw
  e.g. frozen/fresh/canned
  e.g. Brand names where possible for cannned, frozen and packaged foods.

  2. AMOUNTS OF FOODS

  Give as much detail as possible about the amounts of foods:-
  e.g. number & size of slices of bread - small/large loaf; thick/medium/thin slices.
  e.g. thick/medium/thin spread on bread - butter, margarine, jam, marmite, peanut butter, etc.
  e.g. number of tablespoons of vegetables, milk, cream, grated cheese, cottage cheese, custard, rice pudding, etc.
  e.g. large/medium/small portions - meat, fish, cheese
  e.g. number of eggs, fruits, biscuits, etc.
  e.g. pack size of packaged foods and proportion of pack eaten, e.g. crisps, ready meals, baked beans, spaghetti, etc

  3. DRINKS

  For tea/coffee, record whether you have a cup or mug and whether you add milk/cream and sugar. State the type of milk/cream and sugar, e.g. whole/semi-skimmed/skimmed milk, single/double cream and white/brown sugar. State the number of teaspoons of sugar and the number of tablespoons of milk/cream in each cup/mug.
  Record alcoholic drinks as pub measures.
  Record juices and soft drinks as glasses and give the approximate size of glass. State the type of drink, e.g. "orange juice, unsweetened", "strawberry flavour squash, low calorie", "lemonade, ordinary", etc.

  DIETARY REFERENCE VALUES

  If nutrient intakes are to be compared with Dietary Reference Values, the following information also needs to be obtained:

  • Whether the subject is male or female. If female, it is necessary to know whether the subject is pregnant or breast feeding.
  • The subject's date of birth and the start date of survey. (These are needed to calculate the subject's age.)
  • The subject's activity both at work and at leisure:
    1. Not very active
    2. Moderately active
    3. Very active BR>4. Not working

    This information is necessary in order to ensure that nutrient intakes are compared with the most appropriate reference values.

• Records should be checked for completion and entered into a computer within a few days of collection. The reason for this is, if data are incomplete or missing, subjects are unlikely to remember the additional details required if not questioned early enough.

• Ensure consistency within- and between- people who are entering records into a computer.
  Would two people entering data for one record get the same answer? Would one person entering the same record on two occasions get the same answer? If the answer to either of these questions is "no" or "don't know" then your data processing procedures need improvement.

Records can be entered into a computer, analysed and nutrient intakes compared with Dietary Reference Values using WISP. Click here for details.