Sliding dichotomy compared with fixed dichotomization of ordinal outcome scales in subarachnoid hemorrhage trials

Abstract

Object In randomized clinical trials of subarachnoid hemorrhage (SAH) in which the primary clinical outcomes are ordinal, it has been common practice to dichotomize the ordinal outcome scale into favorable versus unfavorable outcome. Using this strategy may increase sample sizes by reducing statistical power. Authors of the present study used SAH clinical trial data to determine if a sliding dichotomy would improve statistical power. Methods Available individual patient data from tirilazad (3552 patients), clazosentan (the Clazosentan to Overcome Neurological Ischemia and Infarction Occurring After Subarachnoid Hemorrhage trial [CONSCIOUS-1], 413 patients), and subarachnoid aneurysm trials (the International Subarachnoid Aneurysm Trial [ISAT], 2089 patients) were analyzed. Treatment effect sizes were examined using conventional fixed dichotomy, sliding dichotomy (logical or median split methods), or proportional odds modeling. Whether sliding dichotomy affected the difference in outcomes between the several age and neurological grade groups was also evaluated. Results In the tirilazad data, there was no significant effect of treatment on outcome (fixed dichotomy: OR = 0.92, 95% CI 0.80–1.07; and sliding dichotomy: OR = 1.02, 95% CI 0.87–1.19). Sliding dichotomy reversed and increased the difference in outcome in favor of the placebo over clazosentan (fixed dichotomy: OR = 1.06, 95% CI 0.65–1.74; and sliding dichotomy: OR = 0.85, 95% CI 0.52–1.39). In the ISAT data, sliding dichotomy produced identical odds ratios compared with fixed dichotomy (fixed dichotomy vs sliding dichotomy, respectively: OR = 0.67, 95% CI 0.55–0.82 vs OR = 0.67, 95% CI 0.53–0.85). When considering the tirilazad and CONSCIOUS-1 groups based on age or World Federation of Neurosurgical Societies grade, no consistent effects of sliding dichotomy compared with fixed dichotomy were observed. Conclusions There were differences among fixed dichotomy, sliding dichotomy, and proportional odds models in the magnitude and precision of odds ratios, but these differences were not as substantial as those seen when these methods were used in other conditions such as head injury. This finding suggests the need for different outcome scales for SAH.

Therapeutic hypothermia to reduce intracranial pressure after traumatic brain injury: the Eurotherm3235 RCT

“…Using an ordinal analysis of the GOSE scores, together with covariate adjustment (primary efficacy analysis), we were able to increase the statistical efficiency of the analysis 101,102 so that a trial involving 600 participants would have power equivalent to that of a trial involving 1000 participants that assessed a binary outcome. We calculated that, with such an analysis, the study would have the equivalent of 80% power to detect a rate of unfavourable outcome (GOSE score of 1-4) that was 9 percentage points lower with hypothermia than with standard care (51% vs. 60%) at the 5% significance level (two-sided).…”

Lessons from the CONSCIOUS-1 Study

“…Saver described other methods of analysis including global statistic, responder analysis, shift analysis (analysis of distributions, rank analysis, analysis over levels). We compared sliding dichotomy, a type of responder analysis, versus fixed dichotomy of outcome scales in the tirilazad trials, CONSCIOUS-1, and the ISAT cohorts [ 90 ]. A sliding dichotomy analysis did not provide improved power and potentially reduce study sample sizes in these patient cohorts.…”

Haptoglobin genotype and aneurysmal subarachnoid hemorrhage

“…However, this approach poses disadvantages: it discards valuable information of the full ordinal nature of outcome measures and ignores initial prognostic risk of patients. [18][19][20][21] Hence, 2 alternative approaches were used as part of a sensitivity analysis for the primary outcome. First, in the sliding dichotomy method, the cutoff point for binarization of GOS and mRS scores is differentiated by the predicted baseline prognosis risk.…”

“…First, in the sliding dichotomy method, the cutoff point for binarization of GOS and mRS scores is differentiated by the predicted baseline prognosis risk. 18,21 Instead of defining good or bad outcome for all patients using a single dichotomization point, the sliding dichotomy approach customizes the definition of good outcome according to the baseline prognosis risk of each patient (additional Methods available from Eprints, eprints. soton.ac.uk/426525/).…”