Confidence intervals give the reader critical information about the precision of an effect estimate reported in a trial

The Journal of Orthopaedic and Sports Physical Therapy publish a regular feature called “evidence in practice”. The latest evidence in practice article provides a great overview of confidence intervals.

Confidence intervals span a range of values above and below an effect estimate. The confidence interval is the range of effects that will most likely contain the true mean effect of treatment, compared to the control. The reader can be confident about the size of the mean effect of treatment when the confidence interval is narrow, but unsure when the confidence interval is wide.

The concept of minimal clinically important difference is also relevant when interpreting confidence intervals. If the confidence interval includes the minimal clinically important difference, it is possible that the treatment does not have a worthwhile benefit, regardless of the size of the effect estimate. When a confidence interval crosses the line of no effect (ie, contains the value of zero when assessing between-group differences), it is possible that the treatment is not more effective than the control. The article includes a graphic for interpretting confidence intervals based on their location with respect to the minimal clinically important difference and line of no effect.

Kamper SJ. Confidence intervals: linking evidence to practice. J Orthop Sports Phys Ther 2019;49(10):763-4

Sign up to the PEDro Newsletter to receive the latest news