A randomized, double-blind, placebo-controlled trial investigates a novel oral Janus kinas… — USMLE Step 3

A randomized, double-blind, placebo-controlled trial investigates a novel oral Janus kinase (JAK) inhibitor for the treatment of moderate-to-severe plaque psoriasis. The primary endpoint is achieving a 75% reduction in the Psoriasis Area and Severity Index (PASI 75) at 16 weeks. A total of $500$ patients are randomized $1:1$ to the treatment or placebo arm. Baseline demographics and disease severity are well balanced between the two groups. During the 16-week trial period, 85 patients (34%) in the placebo group drop out, predominantly citing a perceived lack of efficacy and worsening pruritus. In contrast, only 15 patients (6%) in the treatment group drop out. The investigators conduct a 'per-protocol' analysis restricted only to participants who completed the entire 16-week follow-up. They report that the new drug is vastly superior to placebo, with a relative risk of $4.2$ (95% CI 2.8 - 6.1).

Which of the following best describes the primary threat to the validity of this study's conclusion and the most appropriate methodological approach to mitigate it?

A
Differential dropout has caused confounding; a multivariable regression analysis should be performed adjusting for baseline psoriasis severity.
B
The lack of treatment effect in the placebo group acts as an effect modifier; stratum-specific relative risks based on completion status should be reported.
Differential attrition has introduced selection bias; an intention-to-treat analysis should be performed utilizing a conservative imputation method for missing data.Answer
D
The subjective nature of the PASI score led to observer bias; the analysis should be stratified by the evaluating clinician to control for assessment differences.
E
Non-random loss to follow-up skewed the underlying data distribution; a different inferential statistical test should be applied to the remaining data.

Answer

The study suffers from selection bias due to differential attrition, which must be addressed by performing an intention-to-treat (ITT) analysis using missing data imputation.

The correct answer identifies that the study suffers from attrition bias, a form of selection bias, and that an intention-to-treat (ITT) analysis is required to mitigate it. Attrition bias occurs when there is a differential loss to follow-up between study groups. In this scenario, a massive 34% of the placebo group dropped out due to a lack of efficacy, compared to only 6% in the treatment group. By performing a per-protocol analysis, the investigators only analyzed those who completed the study, effectively removing the treatment failures from the placebo group. This destroys the prognostic balance achieved by randomization. To preserve the benefits of randomization and avoid this bias, an ITT analysis must be used. ITT analyzes all patients in the groups to which they were originally randomized, regardless of whether they completed the trial. Missing data from dropouts is typically handled via conservative methods, such as non-responder imputation (assuming dropouts did not achieve the primary endpoint).

Step-by-Step Solution

Identify the primary methodological issue described in the trial scenario.

There is a highly asymmetric dropout rate between the two groups (34% in placebo vs. 6% in treatment), largely due to a lack of efficacy.

Recognizing post-randomization differences in study completion is the critical first step in diagnosing threats to internal validity.

Determine the type of bias caused by this differential loss to follow-up.

This unequal loss to follow-up introduces attrition bias, a specific subtype of selection bias.

When patients drop out for reasons related to the treatment or the outcome, the remaining groups are no longer comparable, destroying the prognostic balance achieved by initial randomization.

Evaluate the investigators' analytical approach (per-protocol analysis).

The per-protocol analysis is flawed because it excludes the dropouts, evaluating only a biased subset of the original randomized population.

A per-protocol analysis in the setting of massive differential attrition severely compromises validity by artificially altering the apparent success rate of the study arms.

Identify the appropriate statistical method to correct or mitigate this bias.

An intention-to-treat (ITT) analysis should be utilized, where all randomized patients are included in the final analysis, often using non-responder imputation for missing data.

ITT analysis preserves the baseline comparability established by randomization and provides a more realistic, conservative estimate of clinical effectiveness.

Key Concept

Attrition bias is a form of selection bias caused by differential loss to follow-up. It is best mitigated by utilizing an Intention-to-Treat (ITT) analysis.

Question

Answer

Step-by-Step Solution

Key Concept

Common Mistakes

Topics

More questions from this topic