Exposure to cadmium from foods, estimated by analysis and calculation--comparison of the methods.

Cadmium intake of 40 middle-aged Finnish men was determined both by calculation based on computer files and by analysis of duplicate portions. A significant difference was observed between intake estimates of the two methods. The average calculated cadmium intake was 15.8 micrograms/day compared to the 8.2 micrograms/day obtained by analysis. Factors causing the difference between the methods and affecting the comparability are: (i) expression of concentrations below the detection limit; (ii) the calculation is based on the analysis of foods being only washed and peeled whereas duplicate portions are normally processed; (iii) when duplicate portions are prepared using non-representative lots or brands of food, their cadmium content differs from average values used in calculation; (iv) the level of cadmium content in duplicate portion samples is very low affecting the accuracy of the results. The analytical level of cadmium during the analysis of duplicate portions was somewhat lower as compared to the analysis on which the food composition file and the calculation is based. It seems that the complicating factors identified in the present study potentially affect the reliability and comparability of studies of trace element intakes. Therefore one has to be careful when comparing intake estimates derived independently or by different methods.