Article
Comprehension of confidence intervals through audio-visual patient information materials for people with multiple sclerosis (COCO-MS): a web-based randomised controlled trial (DRKS-ID: DRKS00008561)
Search Medline for
Authors
Published: | February 23, 2016 |
---|
Outline
Text
Background: Increasing immunotherapy options lead to complex decision making processes for people with relapsing-remitting multiple sclerosis (PwMS). Besides information about treatment effects, confidence intervals (CIs) can be helpful to assess the reliability of point estimates of treatment effects derived from clinical studies in order to make informed treatment choices.
The objective is the evaluation of audio-visual patient information on CIs through a web-based randomised controlled trial (RCT) in PwMS.
Methods: PwMS (age ≥ 18 years) with suspected or definite relapsing-remitting MS were recruited via the website of the German and Austrian MS society and randomised to one of three versions of audio-visual patient information or a corresponding written standard information (group 1), using the survey tool UNIPARK. Participants were blinded to group assignment.
In the short version (group 4), CIs are explained without using an example, while in the other two versions illustrative examples of an apple farmer were used. The farmer wants to know the average weight of his apples (group 3) and the effectiveness of a treatment to prevent worms affecting apple trees (group 2).
Primary endpoint is comprehension of CIs, assessed with a previously piloted six-item multiple choice questionnaire (α≤.05, β≥.9). Further, numeracy was measured.
Results: The RCT is still ongoing, 572 PwMS are needed for the RCT. We present results from an interim-analysis with 245 finishers. In total 72 participants in group 1, 53 in group 2, 56 in group 3 and 62 in group 4 finished the study. Groups were comparable at baseline. Participants in group 2 and 3 reached the same comprehension score (mean 4.8, SD=1.2 and 1.4 respectively). Mean comprehension score was 4.5 (SD=1.2) in group 1 and 4.1 (SD=1.1) in group 4. Mean comprehension score was significantly higher in group 2 and 3 compared to group 1 (p = 0.01). There were modest significant correlations (Pearson's r) between numeracy and comprehension on CIs in group 1 (0.431, p=0.01) and 2 (0.447, p=0.01).
Conclusions: Understanding CIs may help PwMS to compare treatment options and make informed decisions. Pilot evaluation of different versions of patient information and interim-analysis of an RCT indicate that communication of CIs is possible and that comprehensive versions are better understood than short versions. However, analysis of the full data set is needed for definitive conclusions.