Article

Purpose: Establishing treatment effectiveness is a priority for both clinicians and clinical researchers in the rehabilitation field. Although there is a need for high-quality research in rehabilitation, there is also a need for clinicians to learn and practise the skills of critically appraising research literature. In critically appraising the literature, understanding the role of confidence interval is essential. The purpose of this review is to draw attention to the value of confidence intervals in interpreting the clinical significance of trial findings.
Summary of key points: Confidence intervals can be calculated for many different statistical procedures and are advocated for assessing both statistical and clinical significance. Confidence intervals are particularly useful in helping avoid possibly erroneous conclusions that two groups have similar results when non-significant findings are reported.
Conclusions: Probability values and confidence intervals are complementary and closely related mathematically. The function of the confidence interval is fundamentally an inferential one. Confidence intervals are, thus, of particular relevance when interpreting research findings. Confidence intervals should be reported in the results of clinical trials or, if not included, should be calculated wherever possible.

Langue : ANGLAIS