PPubMed4 Jun 2020Cited 47×

Inference about causation from examination of familial confounding (ICE FALCON): a model for assessing causation analogous to Mendelian randomization

Shuai Li, Minh Bui, John L Hopper

Abstract

1. Inference about Causation from Examination of FAmiliaL CONfounding (ICE FALCON) uses observational data for related individuals, in particular twin pairs, to assess causality between measured variables. 2. ICE FALCON has some similarities to Mendelian randomization (MR) but uses in effect all the familial determinants of the exposure, not just those captured by measured genetic variants, and does not make strong assumptions. ICE FALCON and MR empirically are giving similar conclusions with possibly more information per subject from ICE FALCON. 3. ICE FALCON does not require genetic data and

A figure from Inference about causation from examination of familial confounding (ICE FALCON): a model for assessing causation analogous to Mendelian randomization — fig. from the paper

§ The Valyu brief

The paper, in five minutes

Read from the full text via Valyu Contents

ICE FALCON assesses causality using related individuals, yielding similar conclusions to Mendelian randomization but with up to 2.5‑fold more information per subject.

What’s new

i.Developed the ICE FALCON statistical method for causal inference using familial data.
ii.Demonstrated ICE FALCON on BMI and DNA methylation at the ABCG1 locus, confirming a causal effect.
iii.Applied ICE FALCON to longitudinal BMI data, showing evidence of tracking and familial confounding.
iv.Provided simulation studies that support the validity of ICE FALCON.

How they did it

The authors described the ICE FALCON framework and applied it to two empirical examples using twin pairs from the Australian Mammographic Density Twins and Sisters Study. Regression models (Model 1‑3) were fitted and changes in coefficients were tested via non‑parametric bootstrap (1,000 resamples). Simulations were also performed to verify the method’s theoretical expectations.

What they found

BMI showed a causal effect on DNA methylation at the ABCG1 locus (βself = 0.13, 95% CI 0.05‑0.22) with about 2.5 × more information per subject than MR.
In Example 1, the co‑twin BMI association attenuated by 97% after conditioning on own BMI (β'co‑twin = 0.003, P = 0.08).
In Example 2, follow‑up BMI was strongly associated with baseline BMI (βself = 0.81, 95% CI 0.72‑0.90) and attenuated by 79.9% for the co‑twin effect (β'co‑twin = 0.15, P = 1.5 × 10⁻³⁰).
Simulation studies reproduced the expected patterns of regression‑coefficient changes, confirming ICE FALCON’s validity.

Caveats

✕Requires data from related individuals measured in the same way for all variables.
✕Relies on measured exposures that may be subject to measurement error.
✕Statistical power and effect‑size quantification need further investigation.

§ The Valyu brief

Reading the full paper and taking notes. This takes a few seconds…

§ Ask this paper

Ask a question about this paper

Valyu reads the full text and answers from what the paper actually says.

§ Related work in other archives

Inference about causation from examination of familial confounding (ICE FALCON): a model for assessing causation analogous to Mendelian randomization

The paper, in five minutes

Ask a question about this paper

Similar papers in all four archives

The paper, in five minutes

Similar papers in all four archives