The level of patient-reported outcome reporting in randomised controlled trials of brain tumour patients: A systematic review

Background
To determine the net clinical benefit of a new treatment strategy, information on both survival and patient-reported outcomes (PROs) is required. However, to make an adequately informed decision, PRO evidence should be of sufficiently high quality.

Objective
To investigate the methodological quality of PRO reporting in randomised controlled trials (RCTs) in patients with brain tumours, and to assess the proportion of studies that should impact clinical decision-making.

Methods
We conducted a systematic literature search in several databases covering January 2004 to March 2012. We selected relevant RCTs and retrieved the following data: (1) basic trial demographics and PRO characteristics, (2) quality of PRO reporting and (3) risk of bias. Studies that should impact clinical decision-making based on their methodological robustness were analysed systematically.

Results
We identified 14 RCTs, representing over 3000 glioma patients. Only two RCTs (14%) satisfied sufficiently many key methodological criteria to provide high-quality PRO evidence, and should therefore impact clinical decision-making. Important methodological limitations in other studies were lack of reporting of the extent (43%) and reasons (86%) of missing data and statistical approaches to handle this (71%). PRO results were not interpreted in 79% of the studies and clinical significance was not discussed in 86%. Studies with high-quality PRO evidence generally showed lower risk of bias.

Conclusions
Investigators involved in brain tumour research should pay special attention to methodological challenges identified in current work. The level of PRO reporting should continue to improve in order to facilitate a critical appraisal of study results.