Coding
Coding is generally carried out using an electronic database which is searched by a ‘coder’ to match a food or food code on the system with food items reported in a diet record (Conway et al, 2004). The process of coding to produce estimates of nutrient intake is an important component of dietary assessment, but inaccuracies can occur at several points. Some of the potential sources of error include:
- Difficulties in interpreting written details in food diaries or 24 hour recalls
- Practical example: Food diaries completed by young children may be difficult to read if the clarity of writing and spelling is poor.
- Practical example: Information obtained during 24 hour recall poorly documented by interviewer. Information should be coded as documented (not as recalled), therefore accuracy in documentation is crucial.
- Poorly chosen food codes which do not best match the food or beverage reported in the diary.
- Practical example: Coding a pot of yoghurt as a standard yoghurt when low-fat yoghurt was reported in the food diary. The accuracy of data entry also relies on the respondent providing sufficient detail in the diet record.
- Human error; making mistakes during the coding of food items
- Practical example: Number displacement. Mistakenly coding 200g of pasta rather than the quantity reported in the diary e.g. 100g
- Practical example: Forgetting to code spread (e.g. butter or margarine) twice i.e. for each slice of bread when a sandwich is coded.
- Practical example: When it has been reported that half a cup of tea has been drunk, halving only the weight of tea consumed and not the weight of milk and sugar.
It is inevitable that coders will have to make numerous judgements when coding diet records. Irrespective of the skill and experience of the coder some degree of estimation and ‘guesswork’ is involved in the coding process. Research groups have endeavoured to minimise the errors involved with coding by providing training for dietary coders and implementing quality control checks (Welch et al, 2001). Attempts have also been made to standardise the coding process (Gibson, 2005). For example, a code book was designed for use in the INTERMAP study (Conway et al, 2004) which acted as a ‘rule book’. Its aim was to remove the need for coders to make subjective decisions. The number of errors during coding may also be minimized if ‘coding rules’ are established to deal with incomplete or ambiguous entries in the food record (Conway et al, 2004). The dietary coding manual developed for use in the Infant Feeding Peer Support Trial, UCL, provides standard infant portion sizes for commonly consumed food and drinks as well as listing default entries to aid coders in the coding process.
Variability in coding might also result from differences in the personal characteristics of the food coder (Braakhuis et al, 2003). Factors such as age, nationality, familiarity with cooking, views on portion sizes, and personal eating habits may influence a coder’s judgement when interpreting a food record. Although the use of standard procedures is designed to reduce coding errors, a coder’s familiarity (or lack of familiarity) with the population he or she is coding may introduce bias into the dietary assessment.
Key considerations
Braakhuis et al (2003) provide a set of recommendations designed to help coders reduce their coding errors. Similarly, Conway et al (2004) address efforts to standardise the coding of diet records. Key considerations to take in to account in the coding process include:
- Familiarise oneself with the types of foods and supplement products consumed by the study population.
- Become accustomed with the workings of the dietary analysis program and its food composition database.
- Keep the database up-to-date. This will include updating nutritional composition data on foods already listed on the database and adding new foods.
- Keep an inventory of food composition information for foods that are not included in the food composition database. These may need to be added if a suitable substitute is not already on the database. This may be particularly the case for ‘niche’ foods such as sports products or low-fat varieties. Baby and toddler foods may also have to be dealt with in the same way.
- Asking respondents to keep packaging of e.g. ready meals and unusual processed foods to return with the record can help with identification and coding.
- Adopt a standardised protocol for handling the coding of each new/substiute food item (‘food rules’) so that all coders deal with these items in a consistent manner.
- Develop a ‘code book’ or set of default rules to deal with missing information for certain types of foods and beverages e.g. for unknown types of cooking fat or an unknown type of milk added to a bowl of cereal.
- Implement quality control procedures such as routine spot-checks. It is good practice to check all entries keyed by new coders until the error rate drops to acceptable levels. Thereafter, a pre-agreed % error may be checked on a regular basis as a means of ongoing quality control.
- Edit checks in software to prevent gross data entry errors so you can't enter 1000g of pasta instead of 100g.
Coding programs
Once dietary intake information has been collected a wide range of computer systems (of varying quality) are now available to help with nutritional analysis (see Coding Programs).