Cambridge University
EPIC-Norfolk

EPIC-Norfolk: Nutritonal Methods




Menu:







Introduction to dietry assessment in EPIC-Norfolk

EPIC-Norfolk has used three different methods to record a participant’s diet:


  1. 24 hour Diet Recall (24hDR) – paper version

  2. Food Frequency Questionnaire (FFQ)

  3. 7 day diet diary (7dDD)

Trulli
Trulli


The FFQ is country-specific and was developed by all of the EPIC centres and these are used when research results from the international cohort are presented. The country-specific FFQs were calibrated using 24hDR, which were computer-assisted face-to-face interviews on a subset of the participants using EPICsoft. The 7dDD is our main tool for studying the association between diet and illness (or healthy aging) in the EPIC-Norfolk study. This instrument is combined with the UK version of the FFQ and a paper-based version of the 24hDR. Suites of programs and databases (DINER, DINERMO, CAFÉ and FETA) have been developed for the EPIC-Norfolk study to convert dietary information into nutrient and food group data. These are described here.



The validity of our dietary assessment methods was measured using biological markers (biomarkers). Relative validity was assessed by comparing the dietary assessment instruments against each other. The papers discussing these are outlined in the validation and calibration sections. For further information about our nutritional methods, see the following references:


  1. Bingham SA, Welch AA, McTaggart A, et al. (2001) Nutritional methods in the European Prospective Investigation of Cancer in Norfolk. Public Health Nutr. 4, 847–58.

  2. Welch AA, McTaggart A, Mulligan AA, et al. (2001) DINER (Data Into Nutrients for Epidemiological Research) - a new data-entry program for nutritional analysis in the EPIC-Norfolk cohort and the 7-day diary method. Public Heal. Nutr 4, 1253–1265.

  3. Lentjes MAH, McTaggart A, Mulligan AA, et al. (2014) Dietary intake measurement using 7 d diet diaries in British men and women in the European Prospective Investigation into Cancer-Norfolk study: a focus on methodological issues. Br. J. Nutr. 111, 516–26.

  4. Welch AA, Luben R, Khaw KT, et al. (2005) The CAFE computer program for nutritional analysis of the EPIC-Norfolk food frequency questionnaire and identification of extreme nutrient values. J Hum Nutr Diet 18, 99–116.

  5. Mulligan AA, Luben RN, Bhaniani A, et al. (2014) A new tool for converting food frequency questionnaire data into nutrient and food group values: FETA research methods and availability. BMJ Open 4, e004503.

  6. Lentjes MAH, Bhaniani A, Mulligan AA, et al. (2011) Developing a database of vitamin and mineral supplements (ViMiS) for the Norfolk arm of the European Prospective Investigation into Cancer (EPIC-Norfolk). Public Health Nutr. 14, 459–71.



24-hr Diet Recall (24hDR)

A self-administered 24hDR was sent out to all potential EPIC participants with the invitation to join the study. A second 24hDR was sent out with the invitations to the second health check three years later. This short questionnaire asked for a record of all food and drink, including snacks, taken on the previous day. The 24hDRs were coded using DINER, see the 7-day diet diary data processing section.


Participants completed 24hDRs which they returned by post. Most of these people participated in a first health check and went on to complete a 7dDD. The 24hDR is the only dietary instrument for those who did not attend a health check.



Food Frequency Questionnaire (FFQ)

The FFQ used in the EPIC-Norfolk study was based on the FFQ used in the Nurses’ Health Study and was designed to measure a participant's usual food intake during the previous year. It is also possible to obtain food intake data from the FFQ.


The questionnaire is a 10-page A4 document, consisting of two parts:


FFQ-Part 1

The main part contains a list of 130 commonly and less commonly consumed foods. For each item on the list, participants are asked to indicate their usual frequency of consumption choosing from nine frequency categories. The categories range from "never or less than once/month" to "6 times per day". The servings are specified in terms of units or common portions (e.g. one apple, one slice of bread) or household measures (e.g. glass, cup, spoon). An average portion size was assigned to each questionnaire item (this portion size is the same for all participants, regardless of their sex or age).


FFQ-Part 2

This part includes a set of additional questions on type and brand of breakfast cereal; type of fat used in frying, roasting, grilling or baking; and the amount of visible fat on meat. These questions are linked to relevant items on the list and used to help categorise breakfast cereals and total fat and fatty acid consumption respectively. A further question on milk is also found in part 2, requesting information on the type and quantity of milk consumed.


FFQ Entry & Calculation Process

The FFQ data were originally entered into an Oracle-based entry system, and processed using the CAFE program (Compositional Analyses from Frequency Estimates), which was written in SAS. More recently we developed a new open source, freely available tool, FETA, which is based on the CAFÉ program. The FETA software, which uses a comma-separated values input file, was written in C and C++ languages, enabling faster processing times than SAS and the C/C++ software can also be used from the command line. The step-based graphical wizard for running FETA was written in Perl.



These nearly identical programs (see references for details) convert the frequency category into a portion multiplier (e.g. once a week = 1/7 = 0.14). After multiplication with the portion size, an average daily food weight for each of the 130 FFQ items is obtained. These weights are multiplied with the nutrient composition per gram to obtain the nutrient composition of the actual amount eaten. After summing all FFQ items for a participant, an average daily nutrient intake is obtained. Individuals with more than 10 missing lines of data are excluded. The top and bottom 0.5% of the ratio of energy intake to estimated basal metabolic rate are flagged as extreme outliers of nutrient intakes.



For further information on the nutritional analysis of the EPIC FFQ, see the CAFE and FETA publications:


  1. Welch AA, Luben R, Khaw KT, et al. (2005) The CAFE computer program for nutritional analysis of the EPIC-Norfolk food frequency questionnaire and identification of extreme nutrient values. J Hum Nutr Diet 18, 99–116.

  2. Mulligan AA, Luben RN, Bhaniani A, et al. (2014) A new tool for converting food frequency questionnaire data into nutrient and food group values: FETA research methods and availability. BMJ Open 4, e004503.


Please note that it is also the user’s responsibility to code and process the FFQs in order to obtain nutrient data.


Download: FFQ data-entry program

The entry of FFQ data may be done using commonly available computer programs. Please click here to download the FFQ data-entry program.


Download: FFQ processing program

FETA - a free web-based FFQ entry and processing program based on this FFQ is now available on our website. Although it is free to download, we do encourage you to register with us so we can let you know about software updates and provide support from our team of research nutritionists. Please click here to go to our new FETA download page.


7-Day Diet Diary (7dDD)

Free diaries available.


The EPIC-Norfolk study has printed 7-day diet diaries for use by other studies/projects. There is space left on both the front and back of the A5 booklet to personalise the diary with your own logo, and/or notification of sponsors etc. We can provide the diaries for free but we will require you to pay postage costs. The diaries can be used with any data-entry program. Portion sizes of the photographs displayed can be provided.


Please note that there is currently no funding for diaries to be entered at the study centre in Cambridge. The DINER program will enable you to convert the free text into a structured data format; DINERMO (currently not made publicly available) was developed to calculate nutrient & food group content.


The EPIC-Norfolk 7dDD was based on the diary as used in the British Birth Cohort, and follows a similar structure. 7dDDs form the primary dietary assessment used in EPIC-Norfolk. 7dDD are issued as A5, 50-page booklets. Participants record everything that they have eaten for seven consecutive days (covering weekdays and weekend days) describing amounts and how foods are prepared. The pages for each day have eight sections, seven meal slots for recording food and drinks taken from midnight to midnight (i.e. before breakfast to the last thing in the evening), and a checklist and section for recording additional snacks and other information.


Click here to see a page from a 7dDD


7dDD - instruction pages


The front pages of the 7dDD include suggestions for describing and quantifying food and drinks consumed. Participants are encouraged to include full information including how foods are cooked, the type of fat or oil used in preparation or frying, brand names of commercial products and details of recipes used. The amount an individual consumes of each food item or beverage is crucial information and detailed instructions are also available on how different foods may be usefully quantified. For example, there are colour photographs showing different portions of a range of different foods. Participants can use these where appropriate to help them quantify foods.


At the 1HC, the diary layout was explained to participants by a nurse and standardised instructions given on how to fill it in. To help this process the participant was asked to recall the previous day's intake from waking to going to bed, and the details written into the first day of the diary. The following six days were completed at home and diaries returned in a Freepost envelope.


At further time-points, diaries were posted out or handed to participants at a health check, with reminders as to what information was needed, to encourage adequate records.



Data processing using DINER and DINERMO: introduction

A series of programs and databases that constitute the DINER program has been developed at EPIC-Norfolk to convert 7dDD or 24hDR information into structured data for dietary analysis. Trained data enterers code the data from food diaries according to specified guidelines by selecting items from the extensive food list within DINER (nearly 11,000 items) and entering the associated portion size and any other information. DINER allows food intake and the amount consumed to be translated exactly as recorded in the original diary.


DINERMO consists of further programs and databases which translate the structured data into weights of food and nutrient data for analysis.


When the EPIC-Norfolk study began in 1993, there were no data entry and nutrient analysis systems suitable for a large-scale study with a long period of data collection. A system was needed that would be robust enough to deal with the changes that would occur within the food supply over time and also to investigate the many hypotheses of interest. Systems available in 1993 were unable to record sufficient detail, as they were based on the limited number of food types available within the available national food composition databases. The range of published portion sizes was also limited. Extensive databases have had to be developed and maintained to support DINER and DINERMO. All databases can be updated and extended as new data becomes available without the re-coding of food records.


Each food item in the 7dDD or 24hDR is represented by one line of data in the final structure. A sandwich is entered as bread, spread and filling, a cup of tea is entered as black tea and milk (and sugar where taken).


After a coder has made the initial entry for a food diary, they run a checking program. The program checks that the correct number of days and meal slots have been entered (or noted as empty) and identifies potentially unreasonable amounts of foods and that the appropriate portion types have been used for different food items.



Nutrient and food group calculation: DINERMO

Each food item in a food diary is converted from the entered data into a weight of that food and the calculation of the nutrients contained in the weight consumed. Intake for each day is calculated and the total intake for 7 days divided by the number of days to provide the average intake for each individual.


There are many stages involved in achieving the final set of data. The DINERMO program first checks the entered diary data and also the underlying databases. The individual diary data is checked for completeness with regard to the number of meals and days entered as well as the appropriateness of food choices and correct use of portion sizes.


To be able to calculate the nutrient data, all databases (e.g. for portions, food densities or cooking losses) are checked for completeness as they must contain the data appropriate to the particular foods and portion sizes selected in each diary. There may be new foods which arise in a set of diary data and these must be added to the DINER program with data in the associated databases before nutrient data can be calculated. After all these stages are complete and checks are cleared, the nutrients can be calculated.


The calculation program uses a nutrient database based on the 2200 foods in the UK food composition database (UKFCD). For new foods, nutrient data are derived by matching the known composition of the new food to proportions of up to four existing items in the UKFCD, to create the best possible nutritional equivalent.


A series of in-filling programmes have enabled us to complete missing values in the UKFCD. The design of DINER allows the range of nutrients processed to be extended.


The final data produced by this process is the nutrient intake at day level and the diary mean for each individual.


The DINERMO data can also be used to calculate food intake. i.e. rather than calculating vitamin C intake, we may be interested in the weight of the sum of all fruits and vegetables combined. Below follow three examples representing the entire EPIC-Norfolk cohort.











For further information about the DINER/DINERMO programs, see these publications:


Welch AA, McTaggart A, Mulligan AA, et al. (2001) DINER (Data Into Nutrients for Epidemiological Research) - a new data-entry program for nutritional analysis in the EPIC-Norfolk cohort and the 7-day diary method. Public Heal. Nutr 4, 1253–1265.


Lentjes MAH, McTaggart A, Mulligan AA, et al. (2014) Dietary intake measurement using 7 d diet diaries in British men and women in the European Prospective Investigation into Cancer-Norfolk study: a focus on methodological issues. Br. J. Nutr. 111, 516–26.



Vitamin and mineral supplement intake in the 7dDD (ViMiS)

Dietary supplements remain commonly consumed in the UK, particularly cod liver/fish oil preparations. A nutrient calculation program was designed to convert free text into a specific or generic supplement composition. The nutrient quantities of the supplements were converted to units compatible to the units of consumption in diet. The two sources may therefore be added together in order to obtain total nutrient intake (food + supplement).



For further information about the ViMiS programs and validation:


  1. Lentjes MAH, Bhaniani A, Mulligan AA, et al. (2011) Developing a database of vitamin and mineral supplements (ViMiS) for the Norfolk arm of the European Prospective Investigation into Cancer (EPIC-Norfolk). Public Health Nutr. 14, 459–71.

  2. Lentjes MAH, Mulligan AA, Welch AA, Bhaniani A, Luben RN, Khaw KT. (2015) Contribution of cod liver oil-related nutrients (vitamins A, D, E and eicosapentaenoic acid and docosahexaenoic acid) to daily nutrient intake and their associations with plasma concentrations in the EPIC-Norfolk cohort. J Hum Nutr Diet. 28(6):568-582.


DietWebQ

The EPIC-Norfolk study introduced its first online dietary questionnaire in the 5HC, the DietWebQ, a renaming of the Oxford WebQ, which was developed by our colleagues at Oxford University. The aim of the DietWebQ is to obtain information on the amounts of all foods and drinks consumed over the previous day.


Trulli

The format is similar in style to the FFQ, although it is considerably more detailed. Both a brief user guide and a detailed user guide are provided to assist with the completion of the questionnaire. If a participant is unsure about how to answer a question or would like tips on portion sizes, there is a help section to assist, which is accessed by clicking on the ‘Show Help’ link.


Trulli


Participants are asked to complete multiple DietWebQs in order to take into account weekend and weekday consumption and to obtain a better idea of their typical diet. This new instrument can be used to compare the EPIC-Norfolk cohort with other populations, such as the UK Biobank population, but can also be used as a cross comparison and validation of the FFQ. The data from this new questionnaire has yet to be analysed; we are currently working with colleagues at Oxford University to update the nutrient database underlying the WebQ.


Data entry for the DietWebQ is now closed, thank you for your participation.


Validation

  1. S. A. Bingham, J. H. Cummings. (1985) Urine nitrogen as an independent validatory measure of dietary intake: a study of nitrogen balance in individuals consuming their normal diet. Am J Clin Nutr Dec 42 (6):1276-1289.

  2. S. A. Bingham, R. Williams, T. J. Cole, C. P. Price, J. H. Cummings. (1988) Reference values for analytes of 24-h urine collections known to be complete. Ann Clin Biochem Nov 25 ( Pt 6) ():610-619.

  3. S. A. Bingham. (1994) The use of 24-h urine samples and energy expenditure to validate dietary assessments. Am J Clin Nutr Jan 59 (1 Suppl):227S-231S.

  4. S. A. Bingham, C. Gill, A. Welch, K. Day, A. Cassidy, K. T. Khaw, M. J. Sneyd, T. J. Key, L. Roe, N. E. Day. (1994) Comparison of dietary assessment methods in nutritional epidemiology: weighed records v. 24 h recalls, food-frequency questionnaires and estimated-diet records. Br J Nutr Oct 72 (4):619-643.

  5. S. A. Bingham, A. Cassidy, T. J. Cole, A. Welch, S. A. Runswick, A. E. Black, D. Thurnham, C. Bates, K. T. Khaw, T. J. Key. (1995) Validation of weighed records and other methods of dietary assessment using the 24 h urine nitrogen technique and other biological markers. Br J Nutr Apr 73 (4):531-550.

  6. S. A. Bingham, C. Gill, A. Welch, A. Cassidy, S. A. Runswick, S. Oakes, R. Lubin, D. I. Thurnham, T. J. Key, L. Roe, K. T. Khaw, N. E. Day. (1997) Validation of dietary assessment methods in the UK arm of EPIC using weighed records, and 24-hour urinary nitrogen and potassium and serum vitamin C and carotenoids as biomarkers. Int J Epidemiol 26 Suppl 1 ():S137-S151.

  7. S. A. Bingham, N. E. Day. (1997) Using biochemical markers to assess the validity of prospective dietary assessment methods and the effect of energy adjustment. Am J Clin Nutr Apr 65 (4 Suppl):1130S-1137S.

  8. A. E. Black, A. A. Welch, S. A. Bingham. (2000) Validation of dietary intakes measured by diet history against 24 h urinary nitrogen excretion and energy expenditure measured by the doubly-labelled water method in middle-aged women. Br J Nutr Apr 83 (4):341-354.

  9. N. Day, N. McKeown, M. Wong, A. Welch, S. Bingham. (2001) Epidemiological assessment of diet: a comparison of a 7-day diary with a food frequency questionnaire using urinary markers of nitrogen, potassium and sodium. Int J Epidemiol Apr 30 (2):309-317.

  10. N. M. McKeown, N. E. Day, A. A. Welch, S. A. Runswick, R. N. Luben, A. A. Mulligan, A. McTaggart, S. A. Bingham. (2001) Use of biological markers to validate self-reported dietary intake in a random sample of the European Prospective Investigation into Cancer United Kingdom Norfolk cohort. Am J Clin Nutr Aug 74 (2):188-196.


Calibration

  1. S. Voss, U. R. Charrondiere, N. Slimani, A. Kroke, E. Riboli, J. Wahrendorf, H. Boeing. (1998) [EPIC-SOFT a European computer program for 24-hour dietary protocols] Z Ernahrungswiss Sep 37 (3):227-233.

  2. G. Deharveng, U. R. Charrondiere, N. Slimani, D. A. Southgate, E. Riboli. (1999) Comparison of nutrients in the food composition tables available in the nine European countries participating in EPIC. European Prospective Investigation into Cancer and Nutrition. Eur J Clin Nutr Jan 53 (1):60-79.

  3. N. Slimani, G. Deharveng, R. U. Charrondiere, A. L. van Kappel, M. C. Ocke, A. Welch, A. Lagiou, M. van Liere, A. Agudo, V. Pala, B. Brandstetter, C. Andren, C. Stripp, W. A. van Staveren, E. Riboli. (1999) Structure of the standardized computerized 24-h diet recall interview used as reference method in the 22 centers participating in the EPIC project. European Prospective Investigation into Cancer and Nutrition. Comput Methods Programs Biomed Mar 58 (3):251-266.

  4. N. Slimani, P. Ferrari, M. Ocke, A. Welch, H. Boeing, M. Liere, V. Pala, P. Amiano, A. Lagiou, I. Mattisson, C. Stripp, D. Engeset, R. Charrondiere, M. Buzzard, W. Staveren, E. Riboli. (2000) Standardization of the 24-hour diet recall calibration method used in the European prospective investigation into cancer and nutrition (EPIC): general concepts and preliminary results. Eur J Clin Nutr Dec 54 (12):900-917.

  5. S. Crispim, G. Nicolas, C. Casagrande, V. Knaze, A. Illner, I. Huybrechts, N. Slimani. (2014). Quality assurance of the international computerised 24 h dietary recall method (EPIC-Soft). British Journal of Nutrition, 111(3), 506-515.