scholarly journals Two-step Randomisation: Applying the Results of Small Feasibility Studies of Interventions to Large-scale Mendelian Randomisation Studies to Robustly Infer Causal Effects on Clinical Endpoints

2019 ◽  
Author(s):  
Meda R Sandu ◽  
Rhona Beynon ◽  
Rebecca Richmond ◽  
Diana L. Santos Ferreira ◽  
Lucy Hackshaw-McGeagh ◽  
...  
Keyword(s):  
Standard Deviation ◽  
Green Tea ◽  
Instrumental Variable ◽  
Mendelian Randomization ◽  
Causal Effect ◽  
Two Stage ◽  
Instrumental Variable Analysis ◽  
Metabolic Traits ◽  
Randomization Analysis ◽  
Variable Analysis

Background: Feasibility trials are preliminary trials that assess the viability and acceptability of intervention studies and the effects of the intervention on intermediate endpoints. Due to their short duration, they are unable to establish the effects of the intervention on long-term clinical outcomes. We propose a novel method that could transform the interpretation of feasibility trials using modified two-stage randomisation analyses. Methods In this two-stage process, we explored the effects of a 6-month feasibility factorial randomised controlled trial (RCT) of lycopene and green tea dietary interventions (ProDiet) on 159 serum metabolic traits in 133 men with raised PSA levels but prostate cancer (PCA) free. In the first stage, we conducted an intention-to-treat analysis, using linear regression to examine the effects of the interventions on metabolic traits, compared to the placebo group and instrumental variable analysis to assess the causal effect of the intervention on the outcomes. In the second stage, we used a two-sample Mendelian Randomization (MR) approach to assess the causal effect of metabolic traits altered by the interventions, on PCA risk, using summary statistics data from an international PCA consortium of 44,825 cancer cases and 27,904 controls. ResultsThe systemic effects of lycopene and green tea supplementation on serum metabolic profile were comparable to the effects of the respective dietary advice interventions (R2= 0.65 and 0.76 for lycopene and green tea respectively). Metabolites which were altered in response to lycopene supplementation were acetate (standard deviation difference versus placebo (β)): 0.69; 95% CI= 0.24, 1.15; p=0.003), valine (β: -0.62; -1.03, -0.02; p=0.004), pyruvate (β: -0.56; -0.95, -0.16; p=0.006), and docosahexaenoic acid (β: -0.50; -085, -0.14; p=0.006). The instrumental variable analysis showed there was no evidence that green tea altered the metabolome, but lycopene was associated with an increase in acetate (β=2.13; p=0.006) and decreases in pyruvate (β=-1.90; p=0.009), valine (β=-1.79; p=0.023), diacylglycerol (β=-1.81; p=0.026), alanine (β=-1.55; p=0.015) and DHA (p=0.097), where the regression coefficient represents the standard deviation (SD) difference in metabolite measures per unit change in lycopene (µmol/L) or EGCG (nM).Using MR, a genetically instrumented SD increase in pyruvate increased the odds of PCA by 1.29 (1.03, 1.62; p=0.027). Conclusion Using a two-stage randomization analysis in a feasibility RCT, we found that lycopene lowered levels of pyruvate, which our Mendelian randomization analysis suggests may be causally related to reduced PCA risk.

Using instrumental variables to estimate the attributable fraction

2019 ◽  
Vol 29 (8) ◽  
pp. 2063-2073
Author(s):  
Elisabeth Dahlqwist ◽  
Zoltán Kutalik ◽  
Arvid Sjölander
Keyword(s):  
Coronary Heart Disease ◽  
Risk Factor ◽  
Heart Disease ◽  
Instrumental Variables ◽  
Instrumental Variable ◽  
Mendelian Randomization ◽  
Attributable Fraction ◽  
Two Stage ◽  
Instrumental Variable Analysis ◽  
Variable Analysis

In order to design efficient interventions aimed to improve public health, policy makers need to be provided with reliable information of the health burden of different risk factors. For this purpose, we are interested in the proportion of cases that could be prevented had some harmful exposure been eliminated from the population, i.e. the attributable fraction. The attributable fraction is a causal measure; thus, to estimate the attributable fraction from observational data, we have to make appropriate adjustment for confounding. However, some confounders may be unobserved, or even unknown to the investigator. A possible solution to this problem is to use instrumental variable analysis. In this work, we present how the attributable fraction can be estimated with instrumental variable methods based on the two-stage estimator or the G-estimator. One situation when the problem of unmeasuredconfounding may be particularly severe is when assessing the effect of low educational qualifications on coronary heart disease. By using Mendelian randomization, a special case of instrumental variable analysis, it has been claimed that low educational qualifications is a causal risk factor for coronary heart disease. We use Mendelian randomization to estimate the causal risk ratio and causal odds ratio of low educational qualifications as a risk factor for coronary heart disease with data from the UK Biobank. We compare the two-stage and G-estimator as well as the attributable fraction based on the two estimators. The plausibility of drawing causal conclusion in this analysis is thoroughly discussed and alternative genetic instrumental variables are tested.

scholarly journals Mendelian Randomization using Public Data from Genetic Consortia

2016 ◽  
Vol 12 (2) ◽  
Author(s):  
John R. Thompson ◽  
Cosetta Minelli ◽  
Fabiola Del Greco M
Keyword(s):  
Risk Score ◽  
Instrumental Variable ◽  
Mendelian Randomization ◽  
Causal Effect ◽  
Regression Coefficients ◽  
Risk Scores ◽  
Instrumental Variable Analysis ◽  
Genome Wide ◽  
Public Data ◽  
Variable Analysis

AbstractMendelian randomization (MR) is a technique that seeks to establish causation between an exposure and an outcome using observational data. It is an instrumental variable analysis in which genetic variants are used as the instruments. Many consortia have meta-analysed genome-wide associations between variants and specific traits and made their results publicly available. Using such data, it is possible to derive genetic risk scores for one trait and to deduce the association of that same risk score with a second trait. The properties of this approach are investigated by simulation and by evaluating the potentially causal effect of birth weight on adult glucose level. In such analyses, it is important to decide whether one is interested in the risk score based on a set of estimated regression coefficients or the score based on the true underlying coefficients. MR is primarily concerned with the latter. Methods designed for the former question will under-estimate the variance if used for MR. This variance can be corrected but it needs to be done with care to avoid introducing bias. MR based on public data sources is useful and easy to perform, but care must be taken to avoid false precision or bias.

A National Multi-City Instrumental Variable Analysis of the Causal Effect of Local Particles on Mortality

2018 ◽  
Vol 2017 (1) ◽  
pp. 973
Author(s):  
Joel Schwartz ◽  
Antonella Zanobetti ◽  
Kelvin Fong ◽  
Petros Koutrakis
Keyword(s):  
Instrumental Variable ◽  
Causal Effect ◽  
Instrumental Variable Analysis ◽  
Variable Analysis

scholarly journals Predicting the Direction of Causal Effect Based on an Instrumental Variable Analysis: A Cautionary Tale

2016 ◽  
Vol 4 (1) ◽  
pp. 49-59 ◽  
Author(s):  
Stephen Burgess ◽  
Dylan S. Small
Keyword(s):  
Opposite Direction ◽  
Instrumental Variable ◽  
Null Hypothesis ◽  
Causal Effect ◽  
Cautionary Tale ◽  
Instrumental Variable Analysis ◽  
Variable Analysis ◽  
Additional Assumptions

AbstractAn instrumental variable can be used to test the causal null hypothesis that an exposure has no causal effect on the outcome, by assessing the association between the instrumental variable and the outcome. Under additional assumptions, an instrumental variable can be used to estimate the magnitude of causal effect of the exposure on the outcome. In this paper, we investigate whether these additional assumptions are necessary in order to predict the direction of the causal effect, based on the direction of association between the instrumental variable and the outcome, or equivalently based on the standard (Wald) instrumental variable estimate. We demonstrate by counterexample that if these additional assumptions (such as monotonicity of the instrument–exposure association) are not satisfied, then the instrumental variable–outcome association can be in the opposite direction to the causal effect for all individuals in the population. Although such scenarios are unlikely, in most cases, a definite conclusion about the direction of causal effect requires similar assumptions to those required to estimate a causal effect.

Sign in / Sign up

Export Citation Format

Share Document