Readability of the effect measures on health interventions

Abstract

The measure most used to report treatment effects in survival studies is the Hazard RateRatio (HRR). Patients should be able to make decisions about interventions based on information provided by a health expert. However, the medical literature is replete with erroneous interpretations of the HRR which threatens the decision-making process. When confronted with a treatment which may affect a patient's longevity, the most fundamental question that both patients and doctors must face is: Which option would allow me to live longer? But HRR evaluates differences based on the proportions of survivors. Instead, the ratio of two survival medians (MR: Median Ratio) allows the comparison of times to event. The ultimate goal is to improve the readability of the survival studies. Mainly, we focus on determining the relationship of the HRR with the MR. The specific objective is to empirically quantify the concordance between the inverse of HRR and MR in survival studies published in the New England Journal of Medicine(NEJM).