Total Diet Study

In Total Diet Studies (TDS), substance occurrence data is obtained from measuring food products as consumed. TDS offers a more direct measure of substance concentrations compared to traditional monitoring and surveillance programs that are concerned with contamination of raw agricultural commodities. In a TDS, food selection is based on national consumption data in such a way that 90 to 95% of the usual diet is represented by the samples. Selected foods are collected, prepared as consumed and related foods are pooled prior to analysis. The compositions these TDS food samples are described by the TDS food sample compositions data module.

In MCRA, TDS concentration data can also be used in dietary exposure assessments, using it as an alternative type of concentration data where the foods-as-measured are not the raw primary commodities (RACs), but these are TDS food compositions. To link the concentration data to the consumed foods, the TDS food sample composition information is used in the food conversion algorithm in a manner analogous to the use of food recipes describing the composition of a composite food. The main difference is that the translation proportion is always 100% (default). Take, as an example, a TDS food FruitMix that is composed of apple, orange and pear, then a consumed food (food-as-eaten) apple-pie is converted to apple, wheat and butter (in some specific proportions) and subsequently, apple is converted to food-as-measured FruitMix (100%). Not necessarily all foods as consumed are represented in a TDS food sample. In addition to the TDS food sample compositions, there may be additional foods that are not officially part of a TDS food, but which can be extrapolated to a TDS food sample. Through the use of food extrapolations (read across translations), these foods may be directly linked to a TDS food sample, e.g., by specifying that pineapple is translated to FruitMix, pineapple or foods containing pineapple will also be matched to a FruitMix concentration.

Because TDS samples only contain one single, average measurement, TDS occurrence data can currently only be used for only applicable for chronic exposures assessments. However, when variability information is available for the raw primary foods in the TDS food samples (e.g., from monitoring), this information may be used to approximate the variance of TDS samples.

For more information about Total Diet Studies, visit the TDS-Exposure website http://www.tds-exposure.eu.

Deriving the variance of TDS samples from monitoring

Variability of TDS food sample concentrations can be derived using concentration distributions for the sub-foods of the TDS food samples. For each sub-food, e.g. apple (sub-food of TDS food FruitMix), a coefficient of variation (CV) is specified that is derived using the available monitoring samples. Note that monitoring samples may be composite samples. For apple, composite food samples are measured and each sample contains, for instance, 12 apples with unit weight 200 g. So monitoring concentrations, \(c_{mi}\), are based on composite samples with a total weight \(w_{mi}\) = 2400 g each.

A TDS food sample is composed of \(w_{i}\) g of food \({i}\) with \({i}\) = 1…k, wi represents the PooledAmount in TDS food sample compositions table. Then, the concentration of a TDS food sample may be represented as:

\[c_{TDS}= \sum_{i=1}^k (w_{i} \cdot c_{i}) /\sum_{i=1}^k w_{i}\]

with variance:

\[\mathit{var(c_{TDS})} = \sum_{i=1}^k (w_{i} \cdot \mathit{var(c_{i} )} )/\sum_{i=1}^k w_{i}\]

and \(\mathit{var(c_{i})}\) is the variance of concentrations \(c_{i}\) of food \(i\) with portion sample size \(w_{i}\).

It is expected that increasing the number of units in a composite sample will have a reverse effect on the variation between concentrations. Suppose TDS food FruitMix is composed of 2 x 200 = 400 g apple. The expected variation between portion sizes of 400 g will be larger than between portion sizes of 2400 g:

\[\mathit{var(c_{i})} = \mathit{var(c_{mi})} \cdot w_{mi} /w_{i}\]

The variance of the monitoring samples are corrected as follows, calculate:

  1. \(\mathit{var(c_{mi})} = \log (CV_{mi}^2 + 1)\)

  2. \(\mathit{var(c_{i})} = \mathit{var(c_{mi})} \cdot w_{mi} /w_{i}\)

  3. \(CV_{i} = \sqrt{\exp (\mathit{var}(c + i) - 1 )}\)

Scenario analysis

The outcome of a MCRA risk assessment may be that some foods dominate the right upper tail of the exposure distribution. A scenario analysis answers the question to what extent the risk of foods with a high exposure would have been diminished by an intervention or by taking any precautions. To be able to do so, some information is needed about the variability of the concentration distribution of the raw agricultural commodities that make up the TDS food sample. These distributions may be characterised by a mean and a dispersion factor, the standard deviation or, preferably, a percentile point e.g. p95. Monitoring samples may be used for this purpose. In addition, for each subsample food an upper concentration limit is needed. This value is interpreted as the concentration that is considered a high risk. The decision to intervene or not is based on the comparison between this upper limit and p95.

  • For p95 ≤ limit, most concentration values are below the value that is considered as a potential risk, so there is no urgency to take any precautions.

  • When the opposite is true, i.c. p95 > limit, there may be an argument to intervene for this specific food.

In MCRA, limits and p95’s are supplied in the concentration distributions table. In the MCRA interface, a scenario analysis is checked (optionally) and in the scroll down menu only foods are shown with p95 > limit. Selected foods enter the risk assessment with a reduced concentration value:

\[c_{TDS} / \mathit{reductionfactor},\]

where \(c_{TDS}\) is the concentration value of the TDS food with reductionfactor = p95 / limit.