Statistical analysis of patient-reported outcome data in randomised controlled trials of locally advanced and metastatic breast cancer: a systematic review

Research output: Contribution to journalArticle

Authors

  • Setting International Standards in Analyzing Patient-Reported Outcomes and Quality of Life Endpoints Data Consortium (SISAQOL)
  • Madeline Pe
  • Melanie Calvert

Colleges, School and Institutes

Abstract

Although patient reported outcomes (PROs) such as health-related quality of life (HRQOL) are important endpoints in randomized controlled trials (RCTs), there is little consensus about analysis, interpretation and reporting of these data.
A systematic review was conducted to assess variability, quality, and standards of PRO data analyses in advanced breast cancer RCTs. We searched through PubMed for English language articles published in peer-reviewed journals between January 2001 and October 2017. Eligible articles reported PRO results from RCTs involving adult advanced breast cancer patients receiving anti-cancer treatments with reported sample sizes of at least 50 patients.

Sixty-six RCTs met the selection criteria. A small number of RCTs reported a specific PRO research hypothesis (8/66, 12%). There was heterogeneity in the statistical methods used to assess PRO data, with a mixture of longitudinal and cross-sectional techniques. Not all articles addressed the problem of inflated type I error resulting from multiple testing. Fewer than half of RCTs reported the clinical significance of their findings (28/66, 42%). The majority of trials did not report how missing data was handled (48/66, 73%).

Our review demonstrates a need to improve standards in analysis, interpretation and reporting of PRO data in cancer RCTs. Lack of standardization makes it difficult to draw robust conclusions and compare findings across trials. The Setting International Standards in the Analyzing Patient-Reported Outcomes and Quality of Life Data (SISAQOL) Consortium was set up to address this need and develop recommendations on the analysis of PRO data in RCTs.

Details

Original languageEnglish
Pages (from-to)e459-e469
Number of pages11
JournalThe Lancet Oncology
Volume19
Issue number9
Early online date30 Aug 2018
Publication statusPublished - Sep 2018

Keywords

  • health-related quality of life, advanced breast cancer, systematic review, randomized controlled trials, statistical methodology, patient reported outcomes