Toward vital status sweeps: a case history in sequential monitoring

Experience collecting interim data on mortality: an example from the RALES study

INTRODUCTION: The Randomized Aldactone Evaluation Study (RALES) randomized 822 patients to receive 25 mg spironolactone daily and 841 to receive placebo. The primary endpoint was death from all causes. Randomization began on March 24, 1995; recruitment was completed on December 31, 1996; follow-up was scheduled to continue through December 31, 1999. Evidence of a sizeable benefit on mortality emerged early in the RALES. The RALES data safety monitoring board (DSMB), which met semiannually throughout the trial, used a prespecified statistical guideline to recommend stopping for efficacy. At the DSMB's request, its meetings were preceded by an 'endpoint sweep', that is, a census of all participants to confirm their vital status. METHODS: We used computer simulation to evaluate the effect of the sweeps. RESULTS: The sweeps led to an estimated 5 to 8% increase in the number of reported deaths at the fourth and fifth interim analyses. The data crossed the statistical boundary at the fifth interim analysis. If investigators had reported all deaths within the protocol-required 24-h window, the DSMB might have recommended stopping after the fourth interim analysis. DISCUSSION: Although endpoint sweeps can cause practical problems at the clinical centers, sweeps are very useful if the intervals between patient visits or contact are long or if endpoints require adjudication by committee, reading center, or central laboratory. CONCLUSION: We recommend that trials with interim analyses institute active reporting of the primary endpoints and endpoint sweeps.

An extension of stochastic curtailment for incompletely reported and classified recurrent events: the Multicenter Study of Hydroxyurea in Sickle Cell Anemia (MSH)

The Multicenter Study of Hydroxyurea in Sickle Cell Anemia (MSH), a double-blind randomized clinical trial, compared the frequency of acute vaso-occlusive (painful) crises during 2 yr of follow-up in 299 patients randomly assigned to hydroxyurea or placebo. Most patients had more than one crisis; all crises reported were included in the primary outcome analysis. A total of 7,229 follow-up medical contact reports were classified as crises/not crises by a Crisis Review Committee. Because of the time required to report, document, and classify contacts, interim analyses were prepared with incomplete data. If a stopping boundary were crossed, early termination could be advised only after assessing the potential impact of the incomplete data. In an extension of stochastic curtailment methods, simulation procedures were used to estimate the probability of detecting differences when group crisis rates projected to the end of the study were compared using a rank test. To account for medical contacts not yet reported and the future occurrence of crises, Poisson process models assuming no treatment effect on crisis rates were used for these simulations. The number of unclassified contacts that would be classified as crises was simulated as a binomial random variable. These methods may be useful for interim monitoring in other studies of recurrent events with ongoing event reporting and classification.