Studying at Cambridge

DAPA Measurement Toolkit

Estimated food diaries


Estimated food diaries are detailed assessment methods which provide good estimates of energy intake and most nutrients, foods and food groups. A comparison of methods in the UK arm of EPIC showed that a 7-day diary (estimated record) was in closest agreement to 16 days of weighed intake and had the next highest correlation with biomarkers compared to a food frequency questionnaire and a 24 hour recall [1].

Individuals can record the time, location and whether they consumed meals alone or with others for each eating occasion, providing information on eating patterns and the eating environment. The dietary outcomes which can be assessed by estimated food diaries are summarised in Table D.2.1. Recommendations are based upon multiple days of food diary entries rather than a single day in order to account for variation in diet over time.

Table D.2.1 Dietary outcomes assessed by estimated food diaries over multiple days.

Dietary outcome Possible to assess?
Energy and nutrient intake of total diet Yes
Intake of specific nutrients or food Yes
Infrequently consumed foods Maybe
Dietary pattern Yes
Habitual diet
Within-individual comparison Yes
Between-individual comparison Yes
Meal composition Yes
Frequency of eating/meal occasions Yes
Eating environment Yes
Adult report of diet at younger age

Estimated food diaries or records are a prospective dietary assessment method; an example is displayed in Figure D.2.2.


  • Individuals record details of foods and beverages consumed at the time of consumption
  • Brand names, cooking and preparation methods should be provided by respondents
  • Portion sizes should be estimated as accurately as possible by the respondents, using common unit sizes (e.g. cups of drinks, slices of bread), for which visual guides and in-advance training are helpful
  • Foods can also be weighed and recorded if an individual chooses, but the main purpose of using this method is to avoid the burden of weighing
  • Traditionally the method is pen and paper based but the food diary may also be completed either by digital recording or electronically

  • A short questionnaire can be included to aid interpretation of the record and provide details of core foods regularly eaten e.g. type of milk and to inquire about non-food items such as dietary supplements
  • Parents/carers can complete diaries for young children or provide additional diaries for older children

Time frame

  • Several days of recording are necessary because of daily variations in what people eat
  • The more number of days, the less bias toward certain days of the week (i.e. weekday/weekend variation)
  • The more days of recording, the greater the likelihood of capturing information on foods sporadically consumed. This characteristic is in common with 24-h recall
  • The more number of days, the less accurate towards the end of the period because of study fatigue
  • Researchers may also request a respondent to do the assessment repeatedly to account for seasonal differences in consumption: e.g. 2-day food diaries (weekend and weekday) four times over a year
Figure D.2.2 Example of estimated food diary used in child population.
Source: MRC Epidemiology Unit.

Estimated food diaries have been used successfully in many large-scale studies because of the quality and detail of the data obtained. They are suitable for studies where detailed food and nutrient intakes are required at an individual level. Due to its open ended nature, this method is useful to capture the diversity and complexity of diets. Estimated diaries have been used in:

  • The MRC National Survey of Health and Development [23]
  • The EPIC Norfolk Cohort [3]
  • The UK Women’s Cohort Study which was undertaken in over 35,000 women [5]
  • The method has also been used in children in the Avon Longitudinal Study of Parents and Children [8]

However the data entry requires substantial amounts of time, expertise in dietetics, as well as knowledge on local food market, food culture, and possible dietary preferences. Without sufficient human resources, collection of food diaries would be risky to implement.

As a food diary has an open-ended format, its outputs can be diverse, including consumption of specific food groups, cooking at home, eating meals with family members or alone, use of table salt or dietary supplements, whether breakfast, takeaway foods or snacks between meals, was consumed. See characteristics of the outputs in the page of weighed food diaries.

Estimates of nutrient intakes follow a general procedure common with 24-hour recalls and food-frequency questionnaires. The following steps to estimate nutrient intakes are undertaken:

  • De-compose records of mixed meals quantitatively to individual foods and ingredients (if a food composition table has a mixed meal which is comparable to that in diary’s output, de-composition is not necessary)
  • Convert portion size to gram amounts or amounts in a standard unit (ml for liquids) of each dietary consumption
  • Match all foods and ingredients with items in a food composition table
  • Obtain nutrient intakes from each food, by calculating the product of the amount of each item consumed and nutrients contents per amount of each item
  • Obtain nutrient intakes per day per individual by calculating a sum of intakes of the nutrient per day per individual. This calculation can be specific to intakes in breakfast or different settings in a day.

Nutrition experts often ask the question, “How much of vitamin A intake come from vegetable consumption in the population?” This can be readily calculated by the ratio of the two estimates: a total vitamin A intake from vegetables of all individuals to a total vitamin A intake from all the foods of all individuals. This calculation is important to identify major food sources for a particular nutrient in a population.

Key characteristics are summarised in Table D.2.2.


  • Recorded at point of consumption so there is no reliance on an individual’s memory
  • Portion size often well described so estimates are usually good and this reduces error associated with quantification
  • Detailed description of foods consumed are provided which enhances the accuracy of diet coding
  • Eating occasions are recorded in real time
  • Food consumption away from home (e.g. at a restaurant) can be recorded relatively easily and the method is flexible and suits people with erratic lifestyle habits
  • The estimated diet diary can be incorporated into protocols for large studies where interview time is limited


  • It is time consuming and costly to turn the diaries into dietary data, for example a 4-day diary takes 45 minutes - 1 hour to code
  • The individual must be literate and be motivated as the method imposes a large burden, although much less than the weighed food diary method
  • There is the risk it may be completed from memory, after the food is consumed instead of at the time of intake
  • The individual may alter his/her diet to make it easier to record, or to conceal poor eating habits
  • Subjects may forget to record food items or even meals consumed
  • The portion sizes of some foods may be difficult to estimate if the description given by the individual is inadequate
  • The assessment of foods eaten less than once or twice a week may not be accurate

Table D.2.2 Characteristics of estimated food diaries.

Characteristic Comment
Number of participants Up to ~5000
Cost of development Low
Cost of use Medium
Participant burden High
Researcher burden of data collection Medium
Researcher burden of coding and data analysis High
Risk of reactivity bias Yes
Risk of recall bias Minimised if diary completed at time of consumption
Risk of social desirability bias Yes
Risk of observer bias No
Participant literacy required Yes
Suitable for use in free living Yes
Requires individual portion size estimation Yes

Considerations relating to the use of estimated food diaries for assessing diet in specific populations are described in Table D.2.3.

Table D.2.3 Diet assessment by estimated food diaries in different populations.

Population Comment
Pregnancy Suitable.
Infancy and lactation Requires proxy.
Toddlers and young children May require proxy or adult assistance. A study in toddlers indicated no difference in energy intakes when an estimated diary was compared to a weighed diary [15].
Adolescents Suitable.
Adults Suitable.
Older Adults May require proxy depending on cognitive function. Larger size diaries can be created for those who do not see well.
Ethnic groups Requires language/cultural specificity.
  • Instructions must be provided on food recording, ideally face-to-face or alternatively via phone call; out of hours home visits and phone calls may be required
  • Particular attention should be paid to the estimation of portion sizes
  • The quality of the data has been shown to improve if the record is reviewed by a nutritionist at the end of the data collection period [6]
  • Phone calls or mobile phone messages can be sent to respondents during the period of food recoding
  • Individuals should be instructed to keep food labels of any unusual foods and ready meals consumed to help coders identify or clarify items
  • Diaries should be formatted to allow sufficient space to record the entire dietary intake
  • It is useful to supply a separate instruction book so this can be referred to when the diary is being completed
  • A pocket notebook is also useful to record consumption away from home
  • Collaborative links with centres that routinely carry out diet coding may reduce costs, particularly if the centre has food composition data on specific foods such as ethnic or infant foods or has data on portion size if these are relevant
  • Estimated food diary in record sheet or pocket booklet form for when away from home
  • Additional questionnaire to aid interpretation
  • Trained research assistants to instruct individuals on undertaking a estimated food diary
  • Clear written instructions on completion and return of the tool
  • Trained diet coders
  • Nutrient database and analysis program

A method specific instrument library is being developed for this section. In the meantime, please refer to the overall instrument library page by clicking here to open in a new page.

In recording, taking photos of meals and voice-recording are useful. Photos taken can be digitalized into a volume and then to a weight with information on a density (gram/volume). Estimation of portion sizes can be aided by providing electronic photos that are readily searchable. The advantages include display of portion sizes and reduction of participants’ burdens of documenting foods. Technology assistance is appealing, but apparently computer literacy among participants should be sufficiently high and homogenous in a study population.

The application of such technology assistance is under investigation. Ideally, the advanced approach should be compared to objective measures of dietary exposure.

As well as weighed food diaries, estimated food diaries are resource-intensive, requiring time and human resources to complete data entry and data standardization. Thus, food diaries (estimated or weighed) are realistic and conceivable to implement in a subset of a study population, although this still requires expert knowledge on data entry and further processing. Food diary data in a subset are then to be combined with dietary data of an entire study population deriving from another dietary assessment (e.g. multiple 24-h recalls, food frequency questionnaires). Unique statistical approaches are required to validly merge multiple dietary datasets from multiple methods [18].


  1. Bingham SA, Gill C, Welch A, Cassidy A, Runswick SA, Oakes S, et al. 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. 1997;26 Suppl 1:S137-51.
  2. Bingham SA, Luben R, Welch A, Wareham N, Khaw KT, Day N. Are imprecise methods obscuring a relation between fat and breast cancer? Lancet. 2003;362(9379):212-4
  3. Bingham SA, Welch AA, McTaggart A, Mulligan AA, Runswick SA, Luben R, et al. Nutritional methods in the European Prospective Investigation of Cancer in Norfolk. Public Health Nutr. 2001;4(3):847-58.
  4. Buzzard M. 24-hour dietary recall and food record methods. In: Willett WC, editor. Nutritional Epidemiology 2nd edition. New York, NY: Oxford University Press; 1998. p. 50-73.
  5. Cade JE, Burley VJ, Greenwood DC. The UK Women's Cohort Study: comparison of vegetarians, fish-eaters and meat-eaters. Public Health Nutr. 2004;7(7):871-8.
  6. Cantwell MM, Millen AE, Carroll R, Mittl BL, Hermansen S, Brinton LA, et al. A debriefing session with a nutritionist can improve dietary assessment using food diaries. J Nutr. 2006;136(2):440-5.
  7. Dwyer J. Dietary assessment. In: Shils M, Olson J, Shike M, Ross A, editors. Modern Nutrition in Health and Disease 9th edition. Baltimore, MA: Williams & Wilkins; 1999. p. 937-59.
  8. Emmett PM, Jones LR, Northstone K. Dietary patterns in the Avon Longitudinal Study of Parents and Children. Nutr Rev. 2015;73 Suppl 3:207-30.
  9. Freedman LS, Potischman N, Kipnis V, Midthune D, Schatzkin A, Thompson FE, et al. A comparison of two dietary instruments for evaluating the fat-breast cancer relationship. Int J Epidemiol. 2006;35(4):1011-21.
  10. Gillman MW, Hood MY, Moore LL, Singer MR. Feasibility and acceptance of food records among inner-city fifth-grade students. J Am Diet Assoc. 1994;94(11):1311-3.
  11. Harrington KE, Robson PJ, Kiely M, Livingstone MB, Lambe J, Gibney MJ. The North/South Ireland Food Consumption Survey: survey design and methodology. Public Health Nutr. 2001;4(5a):1037-42.
  12. Kim EH, Willett WC, Colditz GA, Hankinson SE, Stampfer MJ, Hunter DJ, et al. Dietary fat and risk of postmenopausal breast cancer in a 20-year follow-up. Am J Epidemiol. 2006;164(10):990-7.
  13. Knekt P, Albanes D, Seppanen R, Aromaa A, Jarvinen R, Hyvonen L, et al. Dietary fat and risk of breast cancer. Am J Clin Nutr. 1990;52(5):903-8.
  14. Kolar AS, Patterson RE, White E, Neuhouser ML, Frank LL, Standley J, et al. A practical method for collecting 3-day food records in a large cohort. Epidemiology. 2005;16(4):579-83.
  15. Lanigan JA, Wells JC, Lawson MS, Lucas A. Validation of food diary method for assessment of dietary energy and macronutrient intake in infants and children aged 6-24 months. Eur J Clin Nutr. 2001;55(2):124-9.
  16. Prentice RL, Caan B, Chlebowski RT, Patterson R, Kuller LH, Ockene JK, et al. Low-fat dietary pattern and risk of invasive breast cancer: the Women's Health Initiative Randomized Controlled Dietary Modification Trial. Jama. 2006;295(6):629-42.
  17. Sieri S, Krogh V, Ferrari P, Berrino F, Pala V, Thiebaut AC, et al. Dietary fat and breast cancer risk in the European Prospective Investigation into Cancer and Nutrition. Am J Clin Nutr. 2008;88(5):1304-12.
  18. Souverein OW, Dekkers AL, Geelen A, Haubrock J, de Vries JH, Ocke MC, et al. Comparing four methods to estimate usual intake distributions. Eur J Clin Nutr. 2011;65 Suppl 1:S92-101.
  19. Stephen AM. The case for diet diaries in longitudinal studies. Int J Social Res Meth. 2007;10:365-77.
  20. Striegel-Moore RH, Thompson D, Affenito SG, Franko DL, Obarzanek E, Barton BA, et al. Correlates of beverage intake in adolescent girls: the National Heart, Lung, and Blood Institute Growth and Health Study. J Pediatr. 2006;148(2):183-7.
  21. Thiebaut AC, Kipnis V, Schatzkin A, Freedman LS. The role of dietary measurement error in investigating the hypothesized link between dietary fat intake and breast cancer--a story with twists and turns. Cancer Invest. 2008;26(1):68-73.
  22. Velie E, Kulldorff M, Schairer C, Block G, Albanes D, Schatzkin A. Dietary fat, fat subtypes, and breast cancer in postmenopausal women: a prospective cohort study. J Natl Cancer Inst. 2000;92(10):833-9.
  23. Wadsworth M, Kuh D, Richards M, Hardy R. Cohort Profile: The 1946 National Birth Cohort (MRC National Survey of Health and Development). Int J Epidemiol. 2006;35(1):49-54.
  24. Willett W. Commentary: Dietary diaries versus food frequency questionnaires-a case of undigestible data. Int J Epidemiol. 2001;30(2):317-9.