When an antipinworm drug was evaluated in human being, the results are still variable according to authors. Among the various factors involved in these variable results, the selection of subjects must be one of them. In the past, in evaluating a drug, the subjects were simply anal swab positive cases in a population. By that kind of selection method, especially in low endemic population, anal swab positives spontaneously turned to egg negatives at random. Moreover, the time, frequency of negative conversion cannot be predicted. By the results, the cure rates obtained from such ill-defined subjects are variable according to the subjected population. Another problem is that evaluation should represent the efficacy in heavily infected cases. To reinforce and overcome such shortcomings, the evaluator shoud select the cases wha are infected with whole developmental stages of worm. And the evaluator should have a method selecting such ideal cases only with anal swabs. To find out such method the results of anal swabs of long-term follow-up without any antipinworm treatment were reevaluated. In 41 children from 6 orphanages of different endemicity, swabs were taken for 8 or 21 times during 36 or 91 days. The positive rate of follow-up swabs was compared with that of already obtained in 4 preliminary swabs. And we collected other data on the combination of 4 anal swab results in different endemicity. The results are as follows: The positive rate of follow-up swabs was significantly different by the consecutiveness of egg positivity in preliminary swabs as well as the endemicity of subjected population. The consecutively positive cases in higher endemicity over 60% showed the highest, and the least variable positive rate in follow-up swabs. The observed frequency of consecutively positive/consecutively negative combination was always higher than the expected one obtained by binomial distribution equation. This indicates that individual hygiene and previous infection are important factors in determining the actual occurrence of subsequent reinfection. Summarizing these results, the subjects for drug evaluation should be those of consecutively positive cases in population over 60% endemicity. These subjects have the lowest chance of spontaneous negative conversion at random period of follow-up swabs. Eligible cases for such criteria would be sufficiently secured in highly endemic population.