Assessing the dietary intake of young children is challenging. In any 1 day, children may have several carers responsible for providing them with their dietary requirements, and once children reach school age, traditional methods such as weighing all items consumed become impractical. As an alternative to weighed records, food portion size assessment tools are available to assist subjects in estimating the amounts of foods consumed. Existing food photographs designed for use with adults and based on adult portion sizes have been found to be inappropriate for use with children. This article presents a review and summary of a body of work carried out to improve the estimation of portion sizes consumed by children.
Feasibility work was undertaken to determine the accuracy and precision of three portion size assessment tools; food photographs, food models and a computer-based Interactive Portion Size Assessment System (IPSAS). These tools were based on portion sizes served to children during the National Diet and Nutrition Survey. As children often do not consume all of the food served to them, smaller portions were included in each tool for estimation of leftovers. The tools covered 22 foods, which children commonly consume. Children were served known amounts of each food and leftovers were recorded. They were then asked to estimate both the amount of food that they were served and the amount of any food leftover.
Children were found to estimate food portion size with an accuracy approaching that of adults using both the food photographs and IPSAS.
Further development is underway to increase the number of food photographs and to develop IPSAS to cover a much wider range of foods and to validate the use of these tools in a ‘real life’ setting.
Measuring young children's dietary intake is challenging. When children are responsible for reporting their own intake issues of literacy, writing skills, memory constraints and concentration span are of particular concern. From previous studies (Foster, 2003; Anderson et al., 2005; Foster et al., 2008), a personal observation is that there is a great deal of variation in the ability to recall foods consumed within any one age group. This may be dependent on an individual's cognitive development, interest in food, and the attention paid to mealtimes.
Children often have multiple carers in any one day, and although parents may provide accurate accounts of what their children eat at home, many are unable to give details of what they consume at school (Livingstone et al., 1992). The school dinner hall is too busy an environment for staff to pay attention to individual children's intakes.
This article presents a review and summary of a body of work carried out to improve the estimation of portion sizes consumed by children. For further details on methodology refer Foster (2003), Anderson et al. (2005) and Foster et al. (2008).
Portion size assessment
In order for intakes of food to be converted into intakes of nutrients or other food constituents, a measure or estimate of the portion size of each food item consumed is required. Weighing and recording all foods eaten requires a highly motivated and committed subject population, which is unlikely to be representative of the general population. Where children are the subjects of dietary investigations, weighing foods consumed outside the home and away from parents poses additional practical problems.
As an alternative to weighing all food items eaten, a number of methods of measuring dietary intake are available which rely on subjects’ estimates of portion size. Tools are available to assist the subject with this task including food photographs, food replicas and food models (Hackett et al., 1984; Cameron and Van Staveren, 1988; Nelson et al., 1997). However, the application of these tools in improving portion size estimation by children has not been investigated systematically (Livingstone and Robson, 2000).
Another alternative to weighing all foods is the use of average portion sizes for a given population. For adults, average portion sizes for some foods are available (Crawley, 1988), but until recently no equivalent data existed for children. Work to determine average portion sizes for children based on the National Diet and Nutrition Surveys (Gregory et al., 1995; Gregory and Lowe, 2000) has been completed recently (Wrieden et al., 2008).
Use of adult food photographs in children
As part of the evaluation of a primary school-based fruit and vegetable intervention (Anderson et al., 2005) an estimated weight food diary was developed. This included an interview with estimates of portion size being made using the four lowest weight photographs for each food in the adult food atlas (Nelson et al., 1997). In the absence of an alternative, this method has been used quite widely as a visual aid to portion size assessment in children (Curtis et al., 2001; Revill et al., 2001; Anderson et al., 2005).
During this work (Anderson et al., 2005), it was evident that a number of children clearly struggled with the estimates of portion size during the interview. This led to questions regarding the validity of this method for assessing portion size in young children. Small scale validation of the utility of the adult food photograph atlas was undertaken (Foster, 2003). School meals were observed and the portion size of foods served to the children (aged 6–12 years) taking part in the study were weighed along with any food leftover. On the following day, the children were interviewed regarding their school meal using the adult food atlas, and the weights selected by the children were compared against the known weights of foods consumed. In total, 648 estimates of food weight were compared against the actual weight of food consumed.
Even at the group level there were significant differences between the mean of the children's estimates and the actual weights of the foods consumed. There was no clear trend for an improvement with age in the ability to estimate food portion size using the adult food photographs. The mean percentage errors ranged from 15.9% of the actual weight of food consumed for the 11-year-old children, who also had the lowest mean difference between the children's estimates and the actual weight of 2.7 g, to 30.3% for the 12 year olds for whom the mean difference between estimated and actual weight was 10.3 g. Frobisher and Maxwell (2003) also found greater errors in children's estimates of portion size, using both descriptions and adult food photographs, compared with adult's estimates.
Arising from the results of this small scale study, and the need for alternatives to the labour-intensive weighed inventory, a more extensive validation study was conducted (Foster, 2003). The study assessed the ability of children (21 children aged 5–6 years and 35 children aged 10–11 years) to use the adult food photographs to estimate the amount of food on a plate while that plate was in front of them. The foods included in the tests were 16 frequently consumed foods varying in morphology.
Accuracy and precision improved with age; however, the performance of the food photographs was poor with both the younger and older children. Only 43% of children's estimates using the food photographs were within 30% of the actual portion size. The errors in portion size estimation were relatively high, even at the group level. The accuracy of estimates varied with the type of food but no clear pattern could be seen relating to food morphology.
Development of food photographs based on children's portion sizes
To investigate whether the estimates of portion sizes for children would improve if they were provided with age appropriate tools, a feasibility study was conducted. Three portion size assessment tools were developed using three different media. These were food photographs, food models and a novel Interactive Portion Size Assessment System (IPSAS) (Foster et al., 2008).
The foods selected for inclusion in the feasibility study were foods commonly consumed by children in four age groups from 4–16 years. The principal data source was the National Diet and Nutrition Survey: young people aged 4–18 years (Gregory and Lowe, 2000). A total of 22 foods were selected to include a variety of appearances, consistencies and textures. Both single foods (for example, apples) and foods served as part of a meal (for example, sausages and baked beans) were included.
Although adults tend to serve themselves food, and often consume all the foods they have selected, children are usually served by others, and frequently leave a proportion of the food served to them. The amount leftover varies with food type. Consequently, children may never see the amount of food they actually consume. It was therefore decided to develop tools to enable the estimation of the amount of food served and amount leftover.
Four sets of each tool were produced based on portion sizes appropriate to each of four age groups (4–6 year olds, 7–10 year olds, 11–14 year olds and 15–16 year olds). The portion sizes were identified from the weights of foods served to children in each of the four age groups during the National Diet and Nutrition Survey: young people aged 4–18 years (Gregory and Lowe, 2000). For all three tools, seven weights from the 5th to the 95th centile of weight served were calculated and presented for estimation of the amount of food served. These same images/models were used for the estimation of leftovers, where these were a significant proportion of the original food served. In addition, seven separate images/models were presented to obtain a measure of the food left over when the amount left was small. These were seven weights between the 5th centile and the smallest representable portion. The seven weights were equal increments on a log scale.
Children aged 4–16 years (n=201) were supplied with known quantities of foods to eat during a school day; this included breakfast, a snack and a main meal. Food leftovers were weighed. On the following day (24 h after consumption) children estimated the amount of each food they had consumed using each tool. Children were provided with food on 6 days, and separate interviews were conducted for each tool.
Significant differences were found between the accuracy of estimates using the three tools. Children of all ages performed well using the IPSAS and food photographs. The accuracy and precision of estimates made using the food models were poor. For all tools, estimates of the amount of food served were more accurate than estimates of the amount consumed. Issues relating to reporting of foods left over that impact on estimates of the amounts of foods actually consumed require further study.
The accuracy of the children's estimates of portion sizes using age-appropriate food photographs was compared with children's estimates using adult food photographs and adult's estimates using adult food photographs. Accuracy of children's estimates of portion size using age-appropriate photographs was not significantly different from that of adults. Children overestimated a food's weight by 46% on average using adult food photographs but this was reduced to an overestimation of 7% when age-appropriate photographs were provided. Adults overestimated by 18% using the adult food photographs (Foster et al., 2006).
Table 1 and Figures 1, 2 and 3 show the accuracy and precision of estimates in terms of absolute gram difference (estimate of weight in gram minus actual weight in gram). Providing children with food photographs depicting age-appropriate portion sizes greatly increased the accuracy of portion size estimates compared with estimates using photographs designed for use with adults (an underestimate of 8 g compared with an overestimate of 40 g). The limits of agreement for children estimating portion size using age-appropriate photographs are only slightly wider than adults using age-appropriate photographs (−84 to 70 g for children and −78 to 60 g for adults).
The mean overestimate in terms of percentage error and mean underestimate in absolute gram is more likely to be due to the tendency for larger quantities to be underestimated, although smaller quantities are overestimated (Nelson et al., 1994).
Recommendations for future research
Given the importance of monitoring dietary intake of children and the low response rate associated with weighed intakes, it is important that alternative methods of measuring what children eat are explored. It is necessary when assessing the validity of any method of measuring dietary intake in young children that we accept the limitations that the subject's cognitive skills impose.
Collecting an absolutely accurate record of intake is virtually impossible, but by tailoring dietary methods to the specific needs of the population under investigation, the accuracy of dietary reports can be greatly improved.
Further work is required to develop ways of recording dietary intake in children, which move towards that ideal with minimum perturbation of the habitual dietary behaviour of the child.
The age-appropriate food photographs and IPSAS have shown potential for use in assessment of dietary intake in children. To date, the tools have been validated in an artificial situation where the foods were provided and all children were served the same foods and the same portion size. It is essential that the tool is validated against habitual intake in a free living situation. Work is currently underway to develop the food photographs and IPSAS to cover around 85% of all foods consumed by children taking part in the National Diet and Nutrition Survey (Gregory and Lowe, 2000) (both by weight consumed and contribution to energy intake). The photographs and IPSAS will be validated against concurrent 4-day weighed intakes. Interviews including estimation of portion size using the tools will be conducted with both the parent and child. These data will be used to identify the approximate age at which children become more accurate than their parents in the estimation of the portion size of foods consumed outside of the home.
A future research priority should be improving portion size estimation of those foods, which are the main contributors to energy and nutrients of interest and those foods least accurately estimated. The best way in which to present each food in the portion size assessment aid needs to be determined. This should represent the way in which the food is most commonly consumed and should include factors such as angle of the food photograph. For young children, the angle at which food will be viewed on a plate will be smaller compared with adults because of their smaller stature. Improving the estimation of only 20–30 foods could have an important impact on the accuracy of total dietary intake data collected from children.
Anderson AS, Porteous LEG, Foster E, Higgins C, Stead M, Hetherington M et al. (2005). The impact of a school-based nutrition education intervention on dietary intake and cognitive and attitudinal variables relating to fruits and vegetables. Public Health Nutr 8, 650–656.
Cameron ME, Van Staveren WA (1988). Manual on Methodology for Food Consumption Studies. Oxford University Press: Oxford.
Crawley H (1988). Food Portion Sizes, 2nd edn. HMSO: London.
Curtis PJ, Adamson AJ, Mathers JC (2001). Associations between fruit and vegetable, starch and fat consumption? Proc Nutr Soc 60, 183A.
Foster E (2003). Assessing dietary intake in primary school children. PhD thesis, University of Newcastle upon Tyne: Newcastle upon Tyne, UK.
Foster E, Matthews JNS, Nelson M, Harris J, Adamson AJ (2006). Accuracy of estimates of food portion size using food photographs—the importance of providing age-appropriate tools. Public Health Nutr 9, 509–514.
Foster E, Matthews JNS, Lloyd J, Marshall L, Mathers JCM, Nelson M et al. (2008). The development and evaluation of three portion size assessment tools for use with children. Br J Nutr 99, 175–184.
Frobisher C, Maxwell SM (2003). The estimation of food portion sizes: a comparison between using descriptions of portion sizes and a photographic food atlas by children and adults. J Hum Nutr Diet 16, 181–188.
Gregory J, Collins DL, Davies PSW, Hughes JM, Clarke CP (1995). National Diet and Nutrition Survey: Children Aged 1½–4¼ Years. HMSO: London.
Gregory J, Lowe S (2000). National Diet and Nutrition Survey: Young People Aged 4–18 Years. HMSO: London.
Hackett AF, Rugg-Gunn AJ, Appleton DR, Allinson M, Eastoe JE (1984). Sugars-eating habits of 405 11- to 14-year-old English children. Br J Nutr 51, 347–356.
Livingstone MBE, Prentice AM, Coward WA, Strain JJ, Black AE, Davies PSW et al. (1992). Validation of estimates of energy intake by weighed dietary record and diet history in children and adolescents. Am J Clin Nutr 56, 29–35.
Livingstone MBE, Robson PJ (2000). Measurement of dietary intake in children. Proc Nutr Soc 59, 279–293.
Nelson M, Atkinson M, Darbyshire S (1994). Food photographs improve estimates of nutrient intake in dietary surveys. Proc Nutr Soc 56, 205A.
Nelson M, Atkinson M, Meyer J (1997). A Photographic Atlas of Food Portion Sizes. MAFF publications: London.
Revill SA, Adamson AJ, Stacy R, Hooper J, Moynihan P (2001). The effect of an after school ‘food club’ on intake of foods and nutrients by children from deprived social backgrounds. Proc Nutr Soc 60, 189A.
Wrieden WL, Longbottom PJ, Adamson AJ, Ogston SA, Payne A, Haleem MA et al. (2008). Estimation of typical food portion sizes for children of different ages in Great Britain. Br J Nutr 99, 1344–1353.
The work to develop the age-appropriate portion size assessment tools for use in children was funded by the Food Standards Agency (project number N08019). The work on adult estimates of portion size was funded by the Ministry of Agriculture, Fisheries and Food, many thanks to Dr Michael Nelson for allowing us to use this data. Further development of the IPSAS and Food Photographs was funded by the Food Standards Agency (project number N08027). Drs Foster and Adamson are funded by NIHR (National Institute for Health Research) Personal Awards. The views expressed in this publication are those of the authors and not necessarily those of the NHS, NIHR or Department of Health.
AS Anderson has received grant support from Glaxo Smith Kline. The remaining authors have declared no financial interests.
About this article
Cite this article
Foster, E., Adamson, A., Anderson, A. et al. Estimation of portion size in children's dietary assessment: lessons learnt. Eur J Clin Nutr 63, S45–S49 (2009). https://doi.org/10.1038/ejcn.2008.64
- dietary assessment
- portion size
- food photographs
Portion size estimation in dietary assessment: a systematic review of existing tools, their strengths and limitations
Nutrition Reviews (2020)
Simple methods to obtain food listing and portion size distribution estimates for use in semi-quantitative dietary assessment methods
PLOS ONE (2019)
Reliability and validity of food portion size estimation from images using manual flexible digital virtual meshes
Public Health Nutrition (2018)
Accuracy of food photographs for quantifying food servings in a lunch meal setting among Danish children and adults
Journal of Human Nutrition and Dietetics (2018)