Joint Hypothesis Testing and Gatekeeping Procedures for Studies with Multiple Endpoints

Department of Quantitative Health Sciences, Cleveland Clinic, Cleveland, OH 44195, USA.
Anesthesia and analgesia (Impact Factor: 3.47). 05/2012; 114(6):1304-17. DOI: 10.1213/ANE.0b013e3182504435
Source: PubMed


A claim of superiority of one intervention over another often depends naturally on results from several outcomes of interest. For such studies the common practice of making conclusions about individual outcomes in isolation can be problematic. For example, an intervention might be shown to improve one outcome (e.g., pain score) but worsen another (e.g., opioid consumption), making interpretation difficult. We thus advocate joint hypothesis testing, in which the decision rule used to claim success of an intervention over its comparator with regard to the multiple outcomes are specified a priori, and the overall type I error is protected. Success might be claimed only if there is a significant improvement detected in all primary outcomes, or alternatively, in at least one of them. We focus more specifically on demonstrating superiority on at least one outcome and noninferiority (i.e., not worse) on the rest. We also advocate the more general "gatekeeping" procedures (both serial and parallel), in which primary and secondary hypotheses of interest are a priori organized into ordered sets, and testing does not proceed to the next set, i.e., through the "gate," unless the significance criteria for the previous sets are satisfied, thus protecting the overall type I error. We demonstrate methods using data from a randomized controlled trial assessing the effects of transdermal nicotine on pain and opioids after pelvic gynecological surgery. Joint hypothesis testing and gatekeeping procedures are shown to substantially improve the efficiency and interpretation of randomized and nonrandomized studies having multiple outcomes of interest.

1 Follower
22 Reads
  • Source

    Anesthesia & Analgesia 07/2009; 109(1):217-231. DOI:10.1213/ane.0b013e3181a977c0 · 3.47 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Introduction: The GlideScope(®) video laryngoscope has a 60° angled blade and the blade of the Truview PCD™ video laryngoscope has an optical lens that provides a 46° refraction of the viewing angle. Despite successful results using the GlideScope in adults, few studies have been published regarding its use in pediatric patients. We therefore tested our joint primary hypothesis that the GlideScope and the Truview PCD video laryngoscopes provide superior visualization to direct laryngoscopy and are non-inferior regarding time to intubation. Methods: One hundred thirty-four patients (neonate to ten years of age, American Society of Anesthesiologists physical status I-III) scheduled for general surgical procedures were randomized to tracheal intubation using the Truview PCD or GlideScope video laryngoscope or direct laryngoscopy (Macintosh blade). The laryngoscopic view was scored using the Cormack-Lehane scale. Time to intubation (defined as the time from the moment the device entered the patient's mouth until end-tidal CO2 was detected) and the number of attempts were recorded. Results: The Cormack-Lehane views attained using the GlideScope (P > 0.99) and Truview PCD (P = 0.18) were not superior to the views attained with direct laryngoscopy. Furthermore, the view attained using the GlideScope was significantly worse than that attained using direct laryngoscopy (P < 0.001). Fewer patients showed Cormack-Lehane grade I views with the GlideScope than with the Truview PCD (14% vs 82%, respectively; 95% confidence interval [CI] -91% to -46%). The observed median [Q1, Q3] times to intubation were: 39 [31, 59] sec, 44 [28, 62] sec, and 23 [21, 28] sec with the GlideScope, Truview PCD, and direct laryngoscopy, respectively, with median differences of 14 sec (95% CI 7 to 26, GlideScope - direct laryngoscopy) and 17 sec (95% CI 6 to 28, Truview PCD - direct laryngoscopy). Conclusion: The Cormack-Lehane views attained using the GlideScope and the Truview PCD video laryngoscopes were not superior to views attained using direct laryngoscopy. Visualization with the GlideScope was significantly worse than with direct laryngoscopy. Use of the GlideScope and Truview PCD systems should be restricted to patients with specific indications.
    Canadian Anaesthetists? Society Journal 02/2013; 60(5). DOI:10.1007/s12630-013-9906-x · 2.53 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The optimal concentration of ropivacaine for continuous interscalene block after shoulder surgery is currently unknown. Fifty-six patients received a perineural infusion of either ropivacaine 0.1% or 0.2% for 48 hours after shoulder surgery. We assessed pain scores as primary end points and supplemental analgesia, ropivacaine consumption, motor block, side effects, and patient satisfaction as secondary end points. Pain scores were not statistically different during the infusion periods; however, supplemental analgesia consumption was higher in the group receiving ropivacaine 0.1% during the first 24 hours (64% vs 28%, P = 0.022). Other secondary end points were statistically inconclusive. These results suggest that ropivacaine 0.2% provides more effective analgesia than ropivacaine 0.1% during the first 24 hours for continuous interscalene block after shoulder surgery.
    Anesthesia and analgesia 03/2013; 116(3):730-3. DOI:10.1213/ANE.0b013e318280e109 · 3.47 Impact Factor
Show more