Analysis of clinical trials by treatment actually received: Is it really an option?

Abstract

The primary analysis of a randomized clinical trial should compare patients in their randomly assigned treatment groups (intention to treat analysis). When a substantial number of subjects fail to take a prescribed medication or are switched to a different study medication, it is tempting to consider treatment comparisons using only those subjects with treatment as actually received rather than as prescribed. There are several arguments against this approach: the prognostic balance brought about by randomization is likely to be disturbed; sample size will be reduced; and the validity of the statistical test procedures will be undermined. Further, results of analysis by treatment actually received may suffer from a bias introduced by using compliance, a factor often related to outcome independently of the treatment received, to determine the groups for comparison. The extent and nature of this bias will be related to the definition of compliance in an as treated analysis, a definition which could be unintentionally self-serving. We have investigated the problem of the definition of actual treatment in the context of a recent clinical trial. We used several definitions to classify patients as having received or not received treatment as prescribed. These definitions, when used in as treated analyses, provided results that were at times inconsistent or counter-intuitive, and which neither helped to confirm nor further explain the intention to treat analysis.

Once-Weekly Administration of Dapsone/Pyrimethamine vs. Aerosolized Pentamidine as Combined Prophylaxis for Pneumocystis carinii Pneumonia and Toxoplasmic Encephalitis in Human Immunodeficiency Virus-Infected Patients

“…Because the substantial number of switches from dapsone/ pyrimethamine to aerosolized pentamidine diminished the power of the intent-to-treat analysis to detect a difference between the effects of treatment [26], a subgroup analysis was subsequently performed to evaluate the biological effect ofdapsone/pyrimethamine. One group consisted ofall participants randomized to receive dapsone/pyrimethamine who did not need to withdraw from the treatment because ofintolerance and who had received no previous trial dose (n = 120).…”

“…Frequent symptoms of intolerance of dapsone/pyrimethamine caused 30% of participants to cross over to prophylaxis with aerosolized pentamidine, whereas only 4% crossed over to prophylaxis with dapsone/pyrimethamine. Because of this substantial number of treatment switches, the power of the intent-to-treat analysis to detect a difference between the efficacy of treatments was reduced [26], and the results are difficult to interpret. If the analysis is restricted to those events that occurred during the originally assigned treatment (or within 1 month after its discontinuation), the reduction in the incidence of toxoplasmic encephalitis among recipients of dapsone/pyrimethamine is dramatic (table 2).…”

Treatment Switching: Statistical and Decision-Making Challenges and Approaches

“…However, this approach is highly prone to selection bias, because switchers are likely to be prognostically different from nonswitchers (9)(10)(11)(12). There is a general view that this "naïve" approach should be avoided (2)(3)(4).…”

Causal Inference in Randomized Trials with Noncompliance

“…The PP analysis generally yields biased estimates of treatment effects, because whether patients comply with the assigned treatment is not randomized and several factors may affect it. This problem can be avoided by intention-to-treat (ITT) analysis, in which patients are analyzed according to the assigned treatment regardless of the treatment actually received (Fisher et al, 1990;Lee et al, 1991Lee et al, ): 194/1065Lee et al, -523/2695. The ITT estimate may represent the effect of the treatment intended, but generally does not represent the treatment effect itself (Schwartz & Lellouch, 1967;Sheiner & Rubin, 1995).…”