Methods for Analyzing Patient-Reported Outcomes

• Activity Number: 258
• Type: Roundtables
• Date/Time: Tuesday, July 31, 2012 : 7:00 AM to 8:15 AM
• Sponsor: Health Policy Statistics Section
• Author(s): Kelly Zou*+
• Companies: Pfizer Inc.
• Keywords: Health care policy ; Patient-reported outcome ; Patient-centered decision-making ; Responder analysis ; Cumulative distribution function plot ; Minimum important difference

Abstract

A patient-reported outcome (PRO) is any report on the status of a patient's health condition that comes directly from the patient, without interpretation of the patient's response by a clinician or anyone else. The purpose of this round table is to provide an overview of PRO measures to health services and policy statistical researchers. Concepts and methods for analyzing PROs are highlighted and illustrated to inform and educate readers about the measurement and analysis. There are two main focuses during this round-table discussion session. First, responder analysis (e.g., cumulative distribution function plot) and estimation of a minimum important difference based on distribution and anchor based methods are discussed. Second, methods for handling missing data (e.g., entire domain, entire questionnaire, and part of multi-item domain) are discussed. Through appropriate analyses and careful interpretations of PRO data, clinicians, patients, policymakers, and payers can make patient-centered and informed decisions. The target audience of this session includes statisticians engaging in health policy and outcomes research.