Examkin

Question

Difficulty: MediumInferential Statistical Tests

A team of neurologists is conducting a randomized, double-blind crossover trial to compare the efficacy of two disease-modifying therapies (Drug X and Drug Y) in patients with relapsing-remitting multiple sclerosis. Fifty patients are enrolled and randomized to receive either Drug X or Drug Y for 12 months. After a 3-month washout period, the patients are switched to the alternative therapy for another 12 months. The primary outcome measure is the total volume of T2-hyperintense white matter lesions, measured in cubic millimeters (mm3mm^3) via brain magnetic resonance imaging (MRI) at the end of each 12-month treatment period. Preliminary data analysis confirms that the lesion volume measurements follow a normal distribution. Which of the following statistical tests is the most appropriate to compare the primary outcome between the two treatment periods?

  1. Paired t-testAnswer
  2. B
    Two-sample t-test
  3. C
    Wilcoxon signed-rank test
  4. D
    Analysis of variance
  5. E
    McNemar test

Answer

Paired t-test
The correct statistical test is the paired t-test. In this crossover trial, the outcome (lesion volume in mm3mm^3) is a continuous numerical variable. The stem states that the measurements follow a normal distribution, which warrants a parametric test. Furthermore, because each patient acts as their own control by receiving both Drug X and Drug Y, the two sets of measurements are dependent (paired). The paired t-test is the appropriate parametric test for comparing the means of two related groups.

Step-by-Step Solution

1
Identify the type of data being analyzed in the primary outcome.
The outcome is the volume of white matter lesions measured in mm3mm^3. This is numerical, continuous data.
The type of variable (categorical, ordinal, continuous) dictates the branch of statistical tests to consider.
2
Determine if the data follows a normal distribution.
The stem explicitly states that the preliminary data analysis confirms a normal distribution.
Normally distributed continuous data requires parametric tests (like t-tests), whereas skewed data requires non-parametric tests.
3
Evaluate the relationship between the study groups.
The study uses a crossover design where the same 50 patients receive Drug X and later Drug Y. Therefore, the two groups of measurements are dependent (paired).
Independent groups use two-sample tests, while related groups (same subjects measured twice) use paired tests.
4
Select the appropriate test based on the defined parameters.
A paired t-test is selected.
The paired t-test is specifically designed to compare the means of a normally distributed, continuous variable between two dependent groups.

Key Concept

Selecting the correct statistical test requires identifying the variable type (e.g., continuous), data distribution (e.g., normal), and study design (e.g., paired/dependent vs. independent).
Estimated Time:1m 0s
Rate this question

Topics

Inferential Statistical Tests

More questions from this topic