PAUL E. LEAVERTON, PH.D.
To the Editor: It seems apparent that Dr. Feinstein's article, "Clinical Epidemiology III: The Clinical Design of Statistics in Therapy" (Ann. Intern. Med. 69: 1287, 1968) was not reviewed by a biostatistician. Being an "average biostatistician" and since much of the discussion dealt with statistics in medical research, I feel compelled to comment on some of the misrepresentations.
1. It is not true that for "statistical evaluation" a group of patients may be regarded as homogeneous only if they are all equally likely, before treatment, to achieve the condition that is the object of therapy. It is true that inattention
LEAVERTON PE. STATISTICS IN MEDICINE. Ann Intern Med. ;70:869–870. doi: 10.7326/0003-4819-70-4-869_2
Download citation file:
Published: Ann Intern Med. 1969;70(4):869-870.
Copyright © 2018 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use