In these cases, results were more mixed. We observed frequent disagreement between methods as to whether there was colocalization, non-colocalization, or insufficient evidence. In the worst-case scenario, colocalization was only agreed by all four methods for 20% of proteins.

The green dashed arrows indicate potential mechanisms that would lead to heterogeneity and hence differences in MR estimates between populations - examples of each are given in Table 1.

4) PAH prelevance in the All of US dataset based on electronic health records is far higher than expected, and many identified "cases" do not have corresponding medication prescription consistent with PAH.

However, the separation between mean exposure levels in centres is far less than between subgroups defined by the residual-based or doubly-ranked method, allowing us to consider non-linearity over a much narrower range.

We can perform MR analyses in each centre, obtaining context-stratified MR estimates that can be analysed using a heterogeneity test or trend test (i.e. meta-regression).

An alternative is to stratify on existing structure in the data, such as recruitment centres. For instance, in UK Biobank, average vitamin D levels differ across centres - higher in the south-west, lower in Scotland.

Plots below represent the association of moderate periodontitis with cardiovascular disease (left) and cognitive function (right) for 393,216 different covariate adjustment options. The presence of positive and negative estimates is sometimes known as a Janus effect.

Mendelian randomization is becoming a widely-used tool in all areas of epidemiology (including perio). However, its ease of use does not mean its use is always appropriate. We divide investigations into 3 categories: plausible, questionable, and implausible.

This suggests the possibility of developing more targeted statins - achieving the CAD lowering effect but not the T2D increasing effect. We tried to use tissue-specific gene expression to resolve the signal to a body compartment, but we were not able.

We show that the signals for LDL-cholesterol and body mass index (BMI) also do not colocalize, and in an MVMR model using variants in the HMGCR gene region only, LDL-cholesterol and BMI are causal risk factors for CAD, and only BMI is a causal risk factor for T2D.

We show that immortal time bias can arise from confounding by survival until exposure allocation or selection bias from selecting on survival until eligibility.