Communicating the results of randomised clinical trials: do patients understand multi-dimensional patient-reported outcomes?

Angus McNair, Sara Brookes, Chris Davis, Jane Blazeby

Research output: Contribution to journalArticle (Academic Journal)peer-review



Evidence suggests that patient reported outcomes (PROs) from randomised trials may not influence clinical decision-making. Reasons for this are unclear, although little is known about patients’ interpretation of PROs in consultations. This study assessed patients’ understanding of multi-dimensional PROs in a graphical format.


Patients with oesophago-gastric cancer underwent a semi-structured interview. A hypothetical clinical scenario involving two treatments (A and B) was described, in which these treatments had identical chances of survival but different PRO profiles. Four line graphs were then presented. Graph 1 showed changes in a single PRO (dysphagia) over time, and Graph 2 described a different outcome (dyspnoea) with a reversed treatment effect (i.e. treatment B better than treatment A). Graph 3 showed convergent PROs in one dimension (physical function) and Graph 4 increased this complexity showing multiple PRO changes over 18 months. Participants answered questions to assess interpretation of the outcomes. The primary outcome was the proportion of participants who correctly interpreted Graphs 1 and 2. Logistic regression examined associations with clinical and sociodemographic details.


A total of 132 patients (101 male, mean age 67) were interviewed, of whom 115 (87%, 95% CI 81-93) correctly interpreted the first two graphs depicting different PRO dimensions. Simultaneous interpretation of adverse and beneficial treatment effects was achieved by 74 (66%, 95% CI 57-75). The final two graphs of more complex data were correctly interpreted by 97 (73%, 95% CI 66-81) and 108 (81%, 75-88) respectively. Univariable analyses demonstrated associations between patient understanding and age, educational level and cancer site (P≤0.02 for all), however, in a multivariable model these associations were attenuated.


Most patients with oesophago-gastric cancer understand graphical PRO data. It is recommended that graphs be used to compliment verbal information during informed consent for surgery.
Original languageEnglish
JournalJournal of Clinical Oncology
Publication statusPublished - 10 Feb 2010

Structured keywords

  • Centre for Surgical Research


Dive into the research topics of 'Communicating the results of randomised clinical trials: do patients understand multi-dimensional patient-reported outcomes?'. Together they form a unique fingerprint.

Cite this