Risk-Based Monitoring and Fraud Detection in Clinical Trials Using JMP and SAS

5.2 Initials and Birthdate Matching

Perhaps the most straightforward means of identifying subjects who have enrolled multiple times in a clinical trial is to match patients with similar birthdates or initials. At least, this is a very useful first step. The Birthday Problem, known to any student who has taken a course in probability, states that among 50 people there is a 97 percent chance that at least two individuals share the same birthday [1]. In fact, there is a 50-50 chance of a match with only 23 subjects in the clinical trial. This result is based on the assumption that each and every birthday is equally likely, which is not true in practice. The Birthday Problem does illustrate, however, that birthday matches are not as rare as one might ...