A group of investigators is assessing the efficacy of a newly approved oral anticoagulant … — USMLE Step 2 CK

A group of investigators is assessing the efficacy of a newly approved oral anticoagulant compared to warfarin in preventing ischemic stroke among patients with non-valvular atrial fibrillation. They conduct a multi-center retrospective cohort study using electronic medical records of $10,000$ patients over a $2$ -year follow-up period.

The initial unadjusted relative risk (RR) for ischemic stroke with the new anticoagulant compared to warfarin is $0.80$ ( $95\%$ CI: $0.72$ - $0.89$ ).

Suspecting that concurrent chronic kidney disease might impact the outcomes, the investigators stratify the patient population based on the presence or absence of stage 3 or 4 chronic kidney disease. The subsequent analysis yields the following results:

Renal Function Status	Relative Risk ( $95\%$ CI)
Normal renal function	$0.55$ ( $95\%$ CI: $0.48$ - $0.63$ )
Stage 3-4 chronic kidney disease	$1.05$ ( $95\%$ CI: $0.92$ - $1.18$ )

Which of the following epidemiological concepts is best demonstrated by the difference between the unadjusted and stratified relative risks?

A
Attrition bias
B
Confounding
Effect modificationAnswer
D
Observer bias
E
Selection bias

Answer

Effect modification

The difference between the unadjusted and stratified relative risks demonstrates effect modification. Effect modification (or effect measure modification) occurs when the effect of an exposure on an outcome varies depending on the level of a third variable (the effect modifier). In epidemiological studies, this is identified through stratified analysis. If the stratum-specific measures of association (e.g., relative risks of

0.55

and

1.05

) are significantly different from one another, effect modification is present. In this scenario, the new anticoagulant is highly effective in patients with normal renal function (RR =

0.55

) but provides no significant benefit compared to warfarin in patients with advanced chronic kidney disease (RR =

1.05

). Unlike confounding, which is a bias that researchers attempt to eliminate or control for in analysis, effect modification is a natural biological phenomenon that should be described and reported.

Step-by-Step Solution

Analyze the unadjusted (crude) measure of association

The crude RR is

0.80

, indicating a protective effect of the new anticoagulant overall compared to warfarin.

Establishing the baseline association is necessary before evaluating the impact of stratification.

Examine the stratum-specific measures of association

The RR for patients with normal renal function is

0.55

, while the RR for patients with stage 3-4 CKD is

1.05

Comparing stratified results reveals how the third variable influences the primary relationship.

Compare the stratified results to each other and to the crude result

The stratified RRs (

0.55

and

1.05

) are significantly different from each other.

If the stratified RRs are different from each other, the third variable is an effect modifier. If they are similar to each other but different from the crude RR, the variable is a confounder.

Key Concept

Effect modification occurs when the magnitude of an association between an exposure and an outcome differs depending on the level of a third variable. It is identified when stratified measures of association are significantly different from one another.

Question

Answer

Step-by-Step Solution

Key Concept

Common Mistakes

Topics

More questions from this topic