Exact sequential analysis for multiple weighted binomial end points.

Sequential analysis is used in clinical trials and postmarket drug safety surveillance to prospectively monitor efficacy and safety to quickly detect benefits and problems, while taking the multiple testing of repeated analyses into account. When there are multiple outcomes, each one may be given a weight corresponding to its severity. This paper introduces an exact sequential analysis procedure for multiple weighted binomial end points; the analysis incorporates a drug's combined benefit and safety profile. It works with a variety of alpha spending functions for continuous, group, or mixed group-continuous sequential analysis. The binomial probabilities may vary over time and do not need to be known a priori. The new method was implemented in the free R Sequential package for both one- and two-tailed sequential analysis. An example is given examining myocardial infarction and major bleeding events in patients who initiated non-steroidal antiinflammatory drugs.

Comparative Safety of Sulfonylureas and the Risk of Sudden Cardiac Arrest and Ventricular Arrhythmia.

OBJECTIVE: To examine the association between individual antidiabetic sulfonylureas and outpatient-originating sudden cardiac arrest and ventricular arrhythmia (SCA/VA). RESEARCH DESIGN AND METHODS: We conducted a retrospective cohort study using 1999-2010 U.S. Medicaid claims from five large states. Exposures were determined by incident use of glyburide, glimepiride, or glipizide. Glipizide served as the reference exposure, as its effects are believed to be highly pancreas specific. Outcomes were ascertained by a validated ICD-9-based algorithm indicative of SCA/VA (positive predictive value ∼85%). Potential confounding was addressed by adjustment for multinomial high-dimensional propensity scores included as continuous variables in a Cox proportional hazards model. RESULTS: Of sulfonylurea users under study (N = 519,272), 60.3% were female and 34.9% non-Hispanic Caucasian, and the median age was 58.0 years. In 176,889 person-years of sulfonylurea exposure, we identified 632 SCA/VA events (50.5% were immediately fatal) for a crude incidence rate of 3.6 per 1,000 person-years. Compared with glipizide, propensity score-adjusted hazard ratios for SCA/VA were 0.82 (95% CI 0.69-0.98) for glyburide and 1.10 (0.89-1.36) for glimepiride. Numerous secondary analyses showed a very similar effect estimate for glyburide; yet, not all CIs excluded the null. CONCLUSIONS: Glyburide may be associated with a lower risk of SCA/VA than glipizide, consistent with a very small clinical trial suggesting that glyburide may reduce ventricular tachycardia and isolated ventricular premature complexes. This potential benefit must be contextualized by considering putative effects of different sulfonylureas on other cardiovascular end points, cerebrovascular end points, all-cause death, and hypoglycemia.