Basing Intention-to-Treat on Cause and Effect Criteria
- PDF / 12,543,941 Bytes
- 9 Pages / 596.639 x 780.599 pts Page_size
- 35 Downloads / 224 Views
361
Basing Intention-to-Treat on Cause and Effect Criteria
Willion H. Stewort, PLD Biostatistics Department, Ouintiles. . Inc... c Kansas City, Missouri
Key Words Cause and effect; Counterfactual; Finite population; Full analysis set; Imputation; Intention-to-treat; Per-protocol; Randomization Correrpondencr Addrrrs William H . Stewart, PhD. Biostatistics Department, Quintiles, Inc., PO Box 9708,Kansas City, MO MU4-0708 (e-mail: [email protected]).
Statisticians often employ the intention-totreat principle in clinical trials to address any possible bias if subjects are dropped from the analysis data set because of events that might be caused by treatment,for example, protocol violations or missing data. However, there are issues that occur in actual trials that lead to concern about and confusion in identifying a full analysis set for an intention-to-treat analysis. An approach clarifyingthe problems and suggesting solutions is to consider thefinite population induced by randomization. Achikving a balance of observed data from a specific randomization and the unobserved data (counterfactuals) from other possible
INTRODUCTION Statisticians (1-5) have proposed intention-totreat (ITT), sometimes known as intent-to-treat, as a principle for the analysis of randomized clinical trials. An International Conference on Harmonisation guideline (ICH-9) (6) further supports the use of ITT with a "full analysis set" that is "as complete as possible and as close as possible to the intention-to-treat ideal of including all randomized subjects." The guideline defines the "intention-to-treat principle" as, "the principle that asserts that the effect of treatment policy can be best assessed by evaluating on the basis of the intention to treat a subject (ie, the planned treatment regimen) rather than the actual treatment given. It has the consequence that subjects allocated to a treatment group should be followed up, assessed, and analyzed as members of that treatment group irrespective of their compliance with the planned course of treatment." The guideline discusses some special cases in which subjects may be excluded, as well as the use of imputation techniques when data are missing. In general, the guideline presents a standard approach to ITT. Phillips et al. (7)provide a helpful discussion
randomizationsis the main principle. In some cases, subsets from the population of all treatment-subject cases can be identified as reduced populations, retaining the cause and effect structure. Correspondingreduced analysis sets, which are subsets of all randomized subjects, can be selected for inference to reduced populations. With a reduced analysis set, unbiased statistics for the reduced population's parameters may be constructed. A subset of all randomized subjects can be arguably superiorfor the analysis of cause and effect, since it is more consistent with the intended population and its scientific procedures and measurements.
of the ICH-9 guideline. They further discuss the concept of the full analysis set and identif
Data Loading...
