Statistical Significance and Clinical Significance

Statistical Significance and Clinical Significance

Most people misinterpret “statistical significance” as “clinical significance” because they equate “significance” with its literal meaning of “importance”. While statistical significance is the claim that findings obtained a medical study are likely to be attributed to a specific cause, the clinical significance is the practical relevance of a treatment effect. The following essay differentiates statistical significance from clinical significance. It also offers an insight into how both terms are the same

While statistical significance leans on hypothesis testing, clinical significance depends on determining the success of a trial. With statistical significance, the p-value is used to decide whether to accept or reject the null hypothesis. The value 0.05 or 5% acts as the conventional cut off for testing the strengths of the findings against the null hypothesis. Therefore, a P < 0.05 implies that the probability of the results of a study being due to chance is <5%. With clinical significance, the decision to implement or not to implement the trial to practice is based on the reduction of symptoms. Thus, treatment is clinically significant if it helps a patient to return to normal from a dysfunctional state.

Statistical significance fails to assure that the findings of a study are relevant while clinical relevance does. The null hypothesis never finds the applicable importance of research because it often offers results that can mislead clinical practice. The reason is that when a large sample is used, even small treatment effects can appear statistically significant (Schober, Bossers & Schwarte, 2018). However, clinical significance confirms if the results of a trial are likely to cause a positive impact on medical practice.

There is one similarity between statistical significance and clinical significance. Both are used as effect measures. Statistical significance measures the differences in medical outcomes between the control group and treatment and clinical significance measures differences in the impacts of treatment. Thus, both of them quantify treatment associations between variables.

Reference

Schober, P., Bossers, S. M., & Schwarte, L. A. (2018). Statistical significance versus clinical importance of observed effect sizes. Anesthesia & Analgesia, 126(3), 1068–1072. doi: 10.1213/ane.0000000000002798