Elsevier

The Lancet Oncology

Volume 17, Issue 11, November 2016, Pages e510-e514
The Lancet Oncology

Personal View
Analysing data from patient-reported outcome and quality of life endpoints for cancer clinical trials: a start in setting international standards

https://doi.org/10.1016/S1470-2045(16)30510-1Get rights and content

Summary

Measures of health-related quality of life (HRQOL) and other patient-reported outcomes generate important data in cancer randomised trials to assist in assessing the risks and benefits of cancer therapies and fostering patient-centred cancer care. However, the various ways these measures are analysed and interpreted make it difficult to compare results across trials, and hinders the application of research findings to inform publications, product labelling, clinical guidelines, and health policy. To address these problems, the Setting International Standards in Analyzing Patient-Reported Outcomes and Quality of Life Endpoints Data (SISAQOL) initiative has been established. This consortium, directed by the European Organisation for Research and Treatment of Cancer (EORTC), was convened to provide recommendations on how to standardise the analysis of HRQOL and other patient-reported outcomes data in cancer randomised trials. This Personal View discusses the reasons why this project was initiated, the rationale for the planned work, and the expected benefits to cancer research, patient and provider decision making, care delivery, and policy making.

Introduction

Patient-centredness is increasingly identified as a critical component of quality health care.1 With an enhanced emphasis on patient-centred care, health-related quality of life (HRQOL), and other patient-reported outcomes that quantify how a patient feels or functions, are assuming a more prominent role as important endpoints in cancer clinical trials.2, 3

The terms “patient-reported outcome” and “HRQOL” have, at times, been used interchangeably, leading to confusion in terminology.4 However, patient-reported outcome and HRQOL are two distinct terms that complement each other. Patient-reported outcomes are defined as any clinical outcome that is reported directly by the patient and can be captured either through self-report or interview, as long as the interviewer directly records the patient’s responses.5, 6 HRQOL, which is often assessed as a patient-reported outcome, is a multidimensional concept that refers to the patient’s subjective perception of the effect of their disease and treatment on physical, psychological, and social aspects of daily life.6, 7 Many HRQOL questionnaires also cover symptoms of disease, functional impairments, and adverse effects of treatment. This distinction between patient-reported outcomes and HRQOL implies that, on the one hand, patient-reported outcomes can be used to measure constructs other than HRQOL (eg, adherence and experiences of care), and that, on the other hand, HRQOL can be measured by means other than a patient-reported outcome (eg, observer or proxy reports).

The expanding adoption of patient-reported outcomes has revealed an important challenge: the diverse ways of analysing and interpreting patient-reported outcome endpoints make it difficult to compare results across various cancer clinical trials. Insufficient standardisation risks the suboptimal use of these findings to inform policy and treatment decisions, and results in an inefficient use of increasingly finite research funding.8 Moreover, improved standardisation of endpoint definitions and the analysis and presentation of patient-reported outcome data would strengthen the rationale for the use of these endpoints. This in turn would generate rigorous data needed to power future trials that could statistically test important patient-reported outcome hypotheses, thereby complementing traditional radiological and survival-based endpoints.9

What is promising is the increased awareness in the research community that this issue needs to be addressed. Efforts to standardise specific aspects of patient-reported outcome evaluations in cancer clinical trials are underway. For example, recent and ongoing efforts have focused on standardising the outcomes to be measured;10, 11, 12 the content that should be included in protocols (Standard Protocol Items: Recommendations for Interventional Trials in Patient Reported Outcomes);13, 14 and the reporting of clinical trials findings (International Society of Quality of Life reporting standards; Consolidated Standards of Reporting Trials in Patient Reported Outcomes).15, 16 Although these efforts have emphasised standards for collecting and reporting patient-reported outcome data, guidelines and best practices for the analysis and interpretation of patient-reported endpoints in cancer clinical trials are insufficient.

Section snippets

Main objective

The Setting International Standards in Analyzing Patient-Reported Outcomes and Quality of Life Endpoints Data (SISAQOL) consortium is a collaborative initiative assembled by the European Organisation for Research and Treatment of Cancer (EORTC) to address the gap in guidelines and best practices for standardising the analysis and interpretation of patient-reported outcome endpoints in randomised cancer clinical trials, with the aim of developing recommendations to address this issue.

These

What is the problem?

Imagine that a randomised trial was done to assess the relative efficacy of two cancer treatments, A and B. Patients reported their level of physical functioning (measured using a multi-item scale) at baseline and every 6 weeks thereafter until disease progression or study discontinuation. Several analyses could be done, such as a time to deterioration of patient-reported physical functioning compared to baseline; analysis of between-group differences in overall means; or a cross-sectional

How did this problem emerge?

The problem of inconsistency in the analytical approach to HRQOL endpoints does not stem from the relevance of the data or the quality of the information that can be extracted from the data. Rather, the problem is that many different research questions can be asked about HRQOL and other patient-reported outcome data. Therefore, clear and well-defined research questions must be selected a priori and matched with appropriate study design and analyses. Furthermore, although guidelines7, 20 would

Towards a solution

The SISAQOL consortium has been established to develop a consensus on international standards for the analysis of HRQOL and other patient-reported outcome data in clinical trials. In assembling the consortium, it is crucial that key stakeholders are involved so that the insights gathered from this initiative are technically correct, comprehensive, and balanced. Therefore, the group is comprised of not only leading HRQOL researchers and statisticians, but also key individuals from various

Future steps

The objective of the SISAQOL initiative is to produce a suite of tools, guidance, and international consensus standards for the analysis of HRQOL and other patient-reported outcome data from clinical trials. We also aim to provide template macros to be used in a number of commonplace missing data settings and illustrative macros to address these requirements. We expect that having freely available guidelines and tools to facilitate their implementation will result in more reliable findings and

References (26)

  • LB Mokkink et al.

    The COSMIN checklist for assessing the methodological quality of studies on measurement properties of health status measurement instruments: an international Delphi study

    Qual Life Res

    (2010)
  • ME Porter et al.

    Standardizing patient outcomes measurement

    N Engl J Med

    (2016)
  • E Gargon et al.

    The COMET Initiative database: progress and activities from 2011 to 2013

    Trials

    (2014)
  • Cited by (0)

    View full text