Joint Hypothesis Testing and Gatekeeping Procedures for Studies with Multiple Endpoints
ABSTRACT 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.
- [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 CO(2) 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 · 2.50 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 · 3.42 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: To determine whether approaching patients for consent on the day of surgery impairs understanding or produces unacceptable anxiety compared with obtaining consent before the day of surgery. We assessed the effect of the timing of obtaining consent for a moderate- to high-risk factorial trial of clonidine and aspirin in patients having noncardiac surgery. Between February 1, 2011, and November 31, 2011, 2 study personnel used the same standardized script to recruit patients before the day of surgery or on the day of surgery. Patients eligible for the trial were preferentially approached to obtain consent before the day of surgery in the preoperative clinic. Patients who did not attend the preoperative clinic or could not be approached that day were approached for consent on the day of surgery. We evaluated anxiety before and after the trial was discussed, protocol knowledge, consent rates, and perceived obligation to participate. All comparisons were adjusted for differences in potentially confounding variables using inverse propensity score weighting. Patients approached on the day of surgery compared with before the day of surgery had noninferior understanding of the comprehension score (adjusted mean difference, -0.19; 90% CI, -0.47 to 0.10; P<.001 for noninferiority) and a noninferior mean increase in the postapproach anxiety score (adjusted mean difference, 0.19; 90% CI, -0.29 to 0.68; P=.003 for noninferiority). Perceived obligation to participate was not greater on the day of surgery (adjusted mean difference, 0.09; 95% CI, -0.21 to 0.40; P=.57 for superiority); however, consent rates were significantly lower (31% vs 59%; odds ratio, 0.49; 90% CI, 0.33 to 0.72; P=.46 for noninferiority). Approaching patients to obtain consent to participate in a perioperative interventional trial on the day of surgery does not compromise essential elements of the consent process. clinicaltrials.gov Identifier: NCT01082874.Mayo Clinic Proceedings 05/2013; 88(5):446-54. DOI:10.1016/j.mayocp.2012.12.014 · 5.81 Impact Factor