U.S. flag

An official website of the United States government

PMC Full-Text Search Results

Items: 5

1.
FIGURE 1.

FIGURE 1. From: Collider Bias Is Only a Partial Explanation for the Obesity Paradox.

Illustration of collider stratification bias.

Matthew Sperrin, et al. Epidemiology. 2016 Jul;27(4):525-530.
2.
FIGURE 2.

FIGURE 2. From: Collider Bias Is Only a Partial Explanation for the Obesity Paradox.

Association (OR) between A and Y in the null case for varying values of αA,αU, βU, and αAU. In each panel, along the x axis, one of these variables is varied from −3 to 3 (left panel: αA, mid-left panel: αU, mid-right panel: βU, right panel: αAU), and the other parameters are set to default values.

Matthew Sperrin, et al. Epidemiology. 2016 Jul;27(4):525-530.
3.
FIGURE 3.

FIGURE 3. From: Collider Bias Is Only a Partial Explanation for the Obesity Paradox.

Association (OR) between A and Y in the null case for varying values of αA,αU, and βU, without interaction (αAU = 0, solid line) and with interaction (αAU = 1, dotted line). Each column (row) in the lattice corresponds to the given value of βU (αA). Within each subgraph, along the x axis, αU is varied from −3 to 3.

Matthew Sperrin, et al. Epidemiology. 2016 Jul;27(4):525-530.
4.
FIGURE 5.

FIGURE 5. From: Collider Bias Is Only a Partial Explanation for the Obesity Paradox.

Association (OR) between A and Y versus causal effect (log odds) of A on Y for varying values of αA,αU, and βU, with βB = 1. Each column (row) in the lattice corresponds to the given of βU (αA). Within each subgraph, along the x axis, αU is varied from −3 to 3. Association without interaction (αAU = 0): solid line; association with interaction (αAU = 1): dotted line; causal effect: dashed line.

Matthew Sperrin, et al. Epidemiology. 2016 Jul;27(4):525-530.
5.
FIGURE 4.

FIGURE 4. From: Collider Bias Is Only a Partial Explanation for the Obesity Paradox.

Association (OR) between A and Y (solid line) versus causal effect (log odds) of A on Y (dashed line) for a range of values of αA,αU, βU, and αAU. In each panel, along the x axis, one of these variables is varied from −3 to 3 (left panel: αA, mid-left panel: αU, mid-right panel: βU, right panel: αAU), and the other parameters are set to default values.

Matthew Sperrin, et al. Epidemiology. 2016 Jul;27(4):525-530.

Supplemental Content

Recent activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...
Support Center