The Authors Reply

摘要

Goldberg and Zins’ comments ( 1) on our study ( 2) compare our results with those from their previous study ( 3). Although Goldberg and Zins emphasize that both articles “provide more specific insights into the potential biases… in epidemiologic surveys” (1, p. 1345), they disagree with our claim that ours was the first study “to obtain information about nonparticipation bias using a longitudinal design” (2, p. 1310).