CBS 2019
CBSMD教育中心
中 文

Other Relevant Articles

Abstract

Recommended Article

Randomized Comparison of Ridaforolimus-Eluting and Zotarolimus-Eluting Coronary Stents 2-Year Clinical Outcomes: From the BIONICS and NIREUS Trials Coronary Angiography after Cardiac Arrest without ST-Segment Elevation A sirolimus-eluting bioabsorbable polymer-coated stent (MiStent) versus an everolimus-eluting durable polymer stent (Xience) after percutaneous coronary intervention (DESSOLVE III): a randomised, single-blind, multicentre, non-inferiority, phase 3 trial The performance of non-invasive tests to rule-in and rule-out significant coronary artery stenosis in patients with stable angina: a meta-analysis focused on post-test disease probability Safety and feasibility of robotic percutaneous coronary intervention: PRECISE (Percutaneous Robotically-Enhanced Coronary Intervention) Study 2017 ACC/AHA Blood Pressure Treatment Guideline Recommendations and Cardiovascular Risk Effectiveness-Based Guidelines for the Prevention of Cardiovascular Disease in Women—2011 Update: A Guideline From the American Heart Association Qualitative and Mixed Methods Provide Unique Contributions to Outcomes Research

Volume 74, Issue 16, October 2019

JOURNAL:J Am Coll Cardiol. Article Link

Nonproportional Hazards for Time-to-Event Outcomes in Clinical Trials: JACC Review Topic of the Week

J Gregson, L Sharples, GW Stone et al. Keywords: clinical trials; Cox proportional hazards; nonproportional hazards; statistics; time-to-event outcomes; trial design

ABSTRACT


Most major clinical trials in cardiology report time-to-event outcomes using the Cox proportional hazards model so that a treatment effect is estimated as the hazard ratio between groups, accompanied by its 95% confidence interval and a log-rank p value. But nonproportionality of hazards (non-PH) over time occurs quite often, making alternative analysis strategies appropriate. This review presents real examples of cardiology trials with different types of non-PH: an early treatment effect, a late treatment effect, and a diminishing treatment effect. In such scenarios, the relative merits of a Cox model, an accelerated failure time model, a milestone analysis, and restricted mean survival time are examined. Some post hoc analyses for exploring any specific pattern of non-PH are also presented. Recommendations are made, particularly regarding how to handle non-PH in pre-defined Statistical Analysis Plans, trial publications, and regulatory submissions.