Dietary assessment and physical activity measurements toolkit

Dietary assessment - Quantification

The estimation of portion size has been recognised as a source of error in studies measuring dietary intake for more than fifty years (Young, 1953).  Coefficients of variation of the differences between estimates and weights of food portions have been shown to be around 50% for foods and 20% for nutrients; this is a consistent finding (Bingham & Nelson, 1991). The errors associated with quantifying the portion of food consumed are probably the largest measurement error in most dietary assessment methods (Gibson, 2005).

Traditionally the weighing of food in cooking is more common in the UK compared to the US and Australia, where the use of cups and spoons in cooking is more common.  The use of household measures to estimate intake may be particularly difficult for individuals who do not use household measures in food preparation.  Wide variation has been shown in individuals’ ability to estimate food portions. The discrepancies in individual ability to undertake this task has been shown to be independent on age, body weight, social status and gender but do vary with the type and size of food (Young & Nestle, 1995)

Tools are available to assist the individual with this task including food photographs (2 dimensional), food replicas (3 dimensional) and food models (3 dimensional) (Hackett et al 1984; Cameron and Van Staveren, 1988; Nelson et al 1997). Results from studies evaluating food portion aids have shown mixed results and many studies have been carried out in highly controlled and non-representative conditions (Robson & Livingstone, 2000).
As an alternative to weighing all foods, average portion sizes for a given population may be used. For UK adults, average portion sizes for some foods are available (Crawley, 1992), this data requires updating; only recently has equivalent data existed for children in the UK based on data from 1995 and 2000 (Wrieden et al, 2008).  Average portion sizes are used in ascribing portions in food frequency questionnaires (FFQ).  Respondents completing semi-quantitative FFQ often have great difficulty in relating their consumption to pre-defined reference portion sizes (Friedenreich et al, 1992).

Factors which influence portion size estimation
(Rutishauser and Black, 2002)
Food characteristics
    - Foods of indeterminate shape tend to be associated with larger errors and liquids with smaller errors, otherwise there are no consistent observations
    - Larger portion sizes appear more difficult to measure than small portion sizes
Visual aids
    - Household measures (e.g. spoon sizes) may be associated with considerable error
    - Food models may be more reliable than household measures but often only one size is available so introducing bias as individual's report portions which agree with the models available
    - The range of portion sizes available may influence the estimates
    - Graduated food models and two-dimensional pictures may be as helpful as three-dimensional models
    - Multiple photographs seem to result in more accurate estimates
    - Providing scale in the picture is helpful
    - Portion size estimation is difficult for children and adults
    - Women have on occasions been shown to provide better estimates than men, this is likely due to their greater role in food preparation

There is considerable debate as to whether the addition of portion sizes is helpful in a food frequency questionnaire; it may be that respondents take no heed and/or are unable to make the necessary mental calculations if they routinely consume a particular food in a different amount.

Food photographs are the most commonly used tool to assist the estimation of portion sizes.  Studies have shown that both adults and children find the estimation of portion sizes using photographs a difficult task, with a tendency to over report in both populations but particularly in children (Frobisher & Maxwell 2003). Some foods have been reported to cause greater difficulty than others; cheese had the largest error range, muesli was consistently over estimated, and juice incurred the lowest error range in one study (Robson & Livingstone, 2000). There are three main skills involved with the estimation of portion sizes using photographs; perception, the individual’s ability to relate an amount of food, which is present in reality to an amount depicted in a photograph; conceptualisation, the ability of an individual to make a mental construction of an amount which is not present in reality and to relate that to a photograph; and memory, which affects the precision of the conceptualisation process (Nelson et al, 1994).

Photographs that depict a portion (amount consumed), or a serving size (amount served), are typically presented as a series of graduated photographs for each food item, bound together in an atlas (Gibson, 2005). Developing a food atlas is a time consuming process (Nelson and Haraldsdottir, 1998) so where possible an existing tool should be used providing it has been validated in a similar population.  Factors such as the size of the image, number and range of portion sizes depicted should be carefully considered; guidance is given in the Nelson and Haraldsottir paper (1998).  The angle at which the photograph is taken is important, as is whether colour or black and white photographs are used, the use of other objects to give a sense of scale, and how the food is presented. 

It is important that the portion size photographs are based on contemporary data in a similar population.  One study has shown that children overestimated a food’s weight by 46% on average, using adult food photographs but this was reduced to an over-estimation of 7% when age-appropriate photographs were provided. Adults overestimated portions by 18% using the adult food photographs (Foster et al. 2006).

Foster et al (2008) provide a useful description of how food photograph portions were developed using data from previous National Diet and Nutrition Surveys in the UK in children and young people.  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 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. In this study food photographs were compared to a computer tool-the interactive portion size assessment (IPAS) and found to be comparable, with good accuracy but poor precision making it suitable for group estimates not individual estimates.  Estimation of food portions by food models was found to be poor (Foster et al, 2008).

Graduated food models
Canada was one of the first countries to use a collection of three-dimensional graduated food models in its National Nutrition Survey (Gibson, 2005).  Other countries have also included such an aid e.g. United States and New Zealand; NHANES (National Health and Nutrition Examination Survey) utilises 3-dimensional measurement aids in the dietary component of the survey.  The few studies which have compared the use of 2- and 3-dimensional food models suggest that two-dimensional are as effective as three (Pietinen et al, 1998; Posner et al, 1992).  Portion-size estimates are particularly difficult because of the irregular shape of cuts of meats and wide variation in thickness (Gibson, 2005).

In practice
If using food portion estimation aids respondents must be trained on their use to try and reduce the error incurred in this aspect of dietary assessment.

The estimation of portion size is difficult for adults and children and is potentially a large source of error in dietary assessment.  The importance of age appropriate photographs has been demonstrated. Food models appear to be less accurate for estimating portion size.  The advent of technology is likely to improve food portion estimation (see future developments).

Web design by Studio 24