[The influencing factors of selection of collective upon the assessment in epidemiological studies in the example of blood lead levels in children (author's transl)].

In a study of blood lead levels in 1719 children from 2 industrial areas, both with high but differing lead pollution, it is shown how the selection of the collective can influence the results of an epidemiological study. A representative statement on the lead exposure for all children living in a circumscribed area, is not possible if the children are selected according to previous diseases or, for that matter, according to medical criteria of any kind. Also, a selection only upon voluntary basis stemming from parental decision, does not bring correct results. The time chosen for the investigation must also be taken under consideration. The authors suggest a random choice, for example according to date of birth.