Research and statistics: understanding and identifying bias in research studies.

摘要

Surveillance bias may result from one population being monitored more closely or more frequently than the general population. In the study described, in which an association was found between medroxyprogesterone and hypertension, the young women taking medroxyprogesterone were monitored at a greater frequency, every 3 months, than were the young women who were not taking medroxyprogesterone, who were monitored annually.