Evidence-based guidelines for interpreting change scores for the European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire Core 30

https://doi.org/10.1016/j.ejca.2012.02.059Get rights and content

Abstract

Aim

To use published literature and experts’ opinion to investigate the clinical meaning and magnitude of changes in the Quality of Life (QOL) of groups of patients measured with the European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30).

Methods

An innovative method combining systematic review of published studies, expert opinions and meta-analysis was used to estimate large, medium, and small mean changes over time for QLQ-C30 scores.

Results

Nine hundred and eleven papers were identified, leading to 118 relevant papers. One thousand two hundred and thirty two mean changes in QOL over time were combined in the meta-analysis, with timescales ranging from four days to five years. Guidelines were produced for trivial, small, and medium size classes, for each subscale and for improving and declining scores separately. Estimates for improvements were smaller than respective estimates for declines.

Conclusions

These guidelines can be used to aid sample size calculations and interpretation of mean changes over time from groups of patients. Observed mean changes in the QLQ-C30 scores are generally small in most clinical situations, possibly due to response shift. Careful consideration is needed when planning studies where QOL changes over time are of primary interest; the timing of follow up, sample attrition, direction of QOL changes, and subscales of primary interest are key considerations.

Introduction

The European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) is one of the most widely used measures of Health-Related Quality of Life (HRQOL) in cancer patients. Population-based reference data1, 2, 3, 4, 5 and some specific guidance on the interpretation of QLQ-C30 scores6, 7, 8, 9, 10, 11 have been published. However QOL results rarely influence clinical practice, perhaps because Randomised Controlled Trials (RCTs) using the QLQ-C30 (even those reported to a high standard) usually lack a clinical interpretation of the QOL differences.12 The Evidence-Based Interpretation Guidelines (EBIG) project aimed to improve the current guidelines for clinical interpretation of the QLQ-C30. Our innovative methods utilise published QLQ-C30 data, and use meta-analytic techniques to combine these with blinded expert opinions to estimate clinically relevant change.6, 13 This method has previously been used to derive guidelines for the Functional Assessment of Cancer Therapy-General scale (FACT-G)13, 14 and for between-group differences in QLQ-C30 scores.6 Here we provide guidelines for interpreting mean changes over time in QLQ-C30 scores.

Section snippets

Methods

Fuller details of the methods employed in the EBIG project are provided elsewhere.6

Expert size classes

For each reviewer, on each contrast, a weighted average of the assigned scale values (–3, 3) was calculated using their percentage certainty as weights. The mean (over reviewers) of these weighted averages was then calculated for each contrast (the ‘overall opinion’). These were rounded into trivial/small/medium/large categories; referred to as the ‘expert size class’.

Quality assessment and analysis dataset

Since the methodology was novel, it was unknown at the start of the study which contrasts would lead to the best quality

Sample

Nine hundred and eleven papers were identified in the literature review. Two hundred and eighty seven papers were subsequently evaluated by the experts. One hundred and eighteen of which yielded the 1232 changes over time which were relevant for this analysis (‘analysis dataset’, see Fig. 1 for full details). The number of contrasts taken from an individual paper ranged from one to eight per subscale with a median of two per paper (per subscale).

The characteristics of the studies/patients have

Discussion

This innovative methodology has previously been used to derive Evidence-Based Interpretation Guidelines for the FACT-G13, 14 and EORTC QLQ-C306 questionnaires. We extended the current literature on interpretation of the QLQ-C306, 7, 8, 9, 10, 11 by using this innovative methodology to produce guidelines for the interpretation of changes over time for this questionnaire (Table 4). Our estimates utilise 1232 changes over time from published QOL scores and incorporate reviews from 34 cancer/QOL

Role of the funding source

Cancer Research UK funded the study [Grant No. C7852/A5653] and a UICC ICRETT fellowship provided a travel and subsistence grant to K.C. at the start of the study. The funders had no role in the design and conduct of the study; the collection, management, analysis and interpretation of data; the writing of the manuscript; or the decision to submit the manuscript for publication. The corresponding author had full access to all the data in the study and had final responsibility for the decision

Conflict of interest statement

K.C., M.K., P.F., G.C., J.B. and M.M.S.J. have no conflicts of interest to declare. G.V. was the Chair of the EORTC Quality of Life Group from 2008 to 2011 (uncompensated).

Acknowledgements

The authors wish to thank Jane Blazeby, Penny Wright and Jane Maher for their advice on design aspects. The EBIG collaborative group consists of the 35 expert panelists involved over the lifetime of the project, to whom we are extremely grateful: Peter Aquilina, Sydney, Australia; Amit Bahl, University Hospitals Bristol, Bristol, UK; Mike Bennett, Lancaster University, UK; Alison Birtle, Rosemere Cancer Centre, Royal Preston Hospital, Lancashire, UK; Jane Blazeby, University of Bristol,

References (21)

There are more references available in the full text version of this article.

Cited by (343)

View all citing articles on Scopus
View full text