Premium
How much data should we collect in a randomized clinical trial?
Author(s) -
Verter Joel
Publication year - 1990
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/sim.4780090117
Subject(s) - comparability , randomized controlled trial , data collection , clinical trial , computer science , baseline (sea) , medicine , medical physics , data science , statistics , surgery , pathology , oceanography , mathematics , combinatorics , geology
Multicentre randomized controlled clinical trials are usually designed to answer a specific question. In accomplishing this they often collect a large quantity of data during screening, baseline and follow‐up visits. These data are not all necessarily related to the main study question. This collection impacts on the overall recruitment process and the participants' time. During the planning phase of the trial we need to consider the reasons for collecting data and the use of that data for the current study as well as its potential use in future investigations. We discuss briefly some of the possible reasons that data may be collected: screening or determining eligibility; conducting a run‐in or dose titration phase; assessing group comparability; aiding patient management; evaluating the natural history of the disease; monitoring the study; testing the study hypotheses; evaluating adherence; estimating side effects and the use of other therapies; analysis of other study questions.