JOHN F. WARNER, M.D.
To the editor: The article by Meyers and associates (1) claimed a significant reduction in the incidence of cytomegalovirus infections in bone marrow transplant recipients who received cytomegalovirus immune globulin (infections in 2 of 17 patients) when compared with controls who received no globulin (infections in 8 of 19 patients) provided these patients received no prophylactic granulocyte transfusions. The authors note that analysis of the data by Fisher's exact test yielded a p value of 0.05. However, this is the value of only one tail of the test (one tail, p = 0.047; two tails, p = 0.094).
WARNER JF. Statistical Tests in a Published Paper. Ann Intern Med. 1983;99:416–417. doi: 10.7326/0003-4819-99-3-416_2
Download citation file:
Published: Ann Intern Med. 1983;99(3):416-417.
Copyright © 2018 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use