Examkin

Question

Difficulty: Very hardBias and Confounding in Research

A multicenter observational study is investigating the impact of a novel targeted biologic therapy, compared to conventional disease-modifying antirheumatic drugs (DMARDs), on the 3-year incidence of severe infectious complications in adults with seropositive rheumatoid arthritis.

Initial unadjusted analysis reveals that patients receiving the biologic therapy have a significantly lower risk of severe infections (Relative Risk [RR] = 0.70, 95% CI 0.62-0.79, p < 0.01).

Concerned about potential demographic influences on immune function, the research team categorizes the cohort based on patient age at enrollment (< 65 years vs. ≥ 65 years). The subsequent stratified analysis yields the following results:

Age StratumRelative Risk (Biologic vs. DMARD)95% Confidence Intervalp-value
< 65 years0.450.38 - 0.53< 0.01
≥ 65 years1.150.98 - 1.350.12

Based on these epidemiological findings, which of the following is the most appropriate approach for the investigators to adopt regarding the age variable in their final analysis?

  1. Present the incidence outcomes independently for the younger and older patient subgroupsAnswer
  2. B
    Calculate a single adjusted summary estimate across all age categories
  3. C
    Conduct a new trial that randomly assigns participants to specific age categories
  4. D
    Remove participants aged 65 years and older from the dataset to prevent susceptibility bias
  5. E
    Design a follow-up investigation that pairs biologic recipients with conventional DMARD recipients of identical chronological age

Answer

Present the incidence outcomes independently for the younger and older patient subgroups.
The correct action is to present the incidence outcomes independently for the younger and older patient subgroups. The data demonstrates effect modification (interaction), which occurs when the magnitude of an exposure's effect on an outcome differs depending on the level of a third variable (in this case, age). Because the biologic therapy is highly protective in younger patients (RR=0.45) but shows no protective effect in older patients (RR=1.15, not statistically significant), pooling these results into a single adjusted estimate would be misleading. Effect modification is a biological reality to be reported, not a statistical error to be fixed.

Step-by-Step Solution

1
Analyze the crude versus stratified relative risks provided in the data table.
The crude RR is 0.70. The stratified RRs are 0.45 (<65 years) and 1.15 (≥65 years).
Comparing crude and stratified estimates is the primary method to distinguish between confounding and effect modification.
2
Determine the epidemiological phenomenon occurring with the age variable.
Because the RR in one stratum (0.45) is substantially different from the RR in the other stratum (1.15), this demonstrates effect modification.
Effect modification occurs when the effect of an exposure on an outcome is altered by a third variable. Confounding would present with both stratified RRs being similar to each other, but different from the crude RR.
3
Identify the standard statistical reporting protocol for this phenomenon.
The results must be reported separately for each stratum.
Unlike confounding, which is a nuisance variable that must be mathematically controlled or adjusted away, effect modification is a real biological phenomenon that provides valuable clinical information and must be highlighted.

Key Concept

Distinguishing effect modification from confounding through stratified analysis, and recognizing that effect modification dictates reporting stratum-specific results rather than pooled adjustments.
Rate this question

Topics

Bias and Confounding in Research

More questions from this topic