Statistical significance versus clinical relevance. Part I. The essential role of the power of a statistical test.
H. G. Luus, F. O. Müller, B H Meyer
Abstract
H. G. Luus, F. O. Müller, B H Meyer
Abstract
When comparing two treatment groups, hypothesis testing is widely used. However, clinical trialists should be more interested in statistical methods which elicit the magnitude of the differences between treatment groups, rather than a simple indication of whether or not the differences are statistically significant. Statistical significance does not necessarily imply clinical relevance. If the true difference between two treatment groups is so small that it is clinically irrelevant, a sample size can be found for which this difference is statistically significant. On the other hand, if the difference between treatment groups is statistically non-significant, it may still be clinically important. The limitations of conventional hypothesis testing of equal true means as such are highlighted. The need to control the power of the test--which takes into account the difference in treatment means which is considered important (clinically relevant) by the researcher--is discussed.
OpenAlex reports 12 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
When comparing two treatment groups, hypothesis testing is widely used. However, clinical trialists should be more interested in statistical methods which elicit the magnitude of the differences between treatment groups, rather than a simple indication of whether or not the differences are statistically significant. Statistical significance does not necessarily imply clinical relevance. If the true difference between two treatment groups is so small that it is clinically irrelevant, a sample size can be found for which this difference is statistically significant. On the other hand, if the difference between treatment groups is statistically non-significant, it may still be clinically important. The limitations of conventional hypothesis testing of equal true means as such are highlighted. The need to control the power of the test--which takes into account the difference in treatment means which is considered important (clinically relevant) by the researcher--is discussed.
Key concepts: Medicine, Statistical significance, Clinical significance, Statistical power, Statistical hypothesis testing, Relevance (law), Sample size determination, Significant difference