A randomized controlled trial is conducted to evaluate the safety and efficacy of monotherapy with a new disease-modifying antirheumatic drug (DMARD) after switching from methotrexate versus continuing methotrexate in patients with inadequate response to methotrexate. The study enrolled adults (age ≥18) who fulfilled the American College of Rheumatology (ACR)-European League Against Rheumatism (EULAR) classification criteria for rheumatoid arthritis. The investigators defined monotherapy with the new DMARD as effective if the treatment resulted in a 20% improvement in the ACR criteria (ACR20) at week 14. In designing the study, the researchers set the probability of wrongly finding that DMARD monotherapy after switching from methotrexate is more effective than continuing methotrexate as 1%, and they set the probability of wrongly finding that the effectiveness of the two therapies is the same as 5%. Which of the following is the most accurate estimate of the statistical power in this study?

Question

A randomized controlled trial is conducted to evaluate the safety and efficacy of monotherapy with a new disease-modifying antirheumatic drug (DMARD) after switching from methotrexate versus continuing methotrexate in patients with inadequate response to methotrexate.  The study enrolled adults (age ≥18) who fulfilled the American College of Rheumatology (ACR)-European League Against Rheumatism (EULAR) classification criteria for rheumatoid arthritis.  The investigators defined monotherapy with the new DMARD as effective if the treatment resulted in a 20% improvement in the ACR criteria (ACR20) at week 14.  In designing the study, the researchers set the probability of wrongly finding that DMARD monotherapy after switching from methotrexate is more effective than continuing methotrexate as 1%, and they set the probability of wrongly finding that the effectiveness of the two therapies is the same as 5%.  Which of the following is the most accurate estimate of the statistical power in this study?

Quizplus · Accepted Answer

Explanation: 11ec2b69_bdfa_9917_8f63_bf5c991bfb36_MD0005_00 Research studies generally compare a null hypothesis (H₀) (eg, no difference in effectiveness between ≥2 treatments) against an alternative hypothesis (H_a) (eg, difference in treatment effectiveness between ≥2 treatments). Hypothesis testing may result in 1 of 4 possible outcomes: _a Two correct decisions:Determining there is no difference in treatment effectiveness when one truly does not exist (correctly failing to reject a true H₀)Determining there is a difference in treatment effectiveness when one truly exists (correctly rejecting a false H₀); this reflects the power of a study Two incorrect decisions:Determining there is a difference in treatment effectiveness when one truly does not exist (incorrectly rejecting a true H₀); this reflects a type I errorDetermining there is no difference in treatment effectiveness when one truly exists (incorrectly failing to reject a false H₀); this reflects a type II error The power of a test (probability of correctly rejecting a false H₀) is the complement of the probability of a type II error (β) (probability of incorrectly failing to reject a false H₀), or power = (1 − β). In this study, H₀ represents no difference in the effectiveness of monotherapy with a new disease-modifying antirheumatic drug (DMARD) after switching from methotrexate compared to continuing methotrexate. The probability of wrongly finding that the effectiveness of the two therapies is the same (incorrectly failing to reject a false H₀) was set at 5%; this reflects the probability of a type II error. Therefore, the probability of its complement, or power (probability of correctly rejecting a false H₀), is 100% − 5% = 95% (Choice B). (Choices A and E) The probability of incorrectly finding that monotherapy with the new DMARD after switching from methotrexate is more effective than continuing methotrexate (probability of a type I error) was set at 1%. The probability of its complement (probability of correctly failing to reject a true H₀) is 100% − 1% = 99%. (Choice C) 20% represents the minimum change in the American College of Rheumatology criteria used to define treatment effectiveness. Educational objective: The power of a test is the probability of rejecting a false null hypothesis H₀ (determining a difference in treatment effectiveness when one truly exists). It is the complement of a type II error: failure to reject a false H₀ (determining no difference in treatment effectiveness when one truly exists).

A Randomized Controlled Trial Is Conducted to Evaluate the Safety