LDL‐c is observationally positively associated with CAD risk (Di Angelantonio et al., 2009). Evidence from randomized trials of pharmaceutical interventions to lower LDL‐c concentrations (such as statins (Pedersen et al., 1994; Cheung et al., 2004)), and from Mendelian randomization (Linsel‐Nitschke et al., 2008; Do et al., 2013) suggests that this association is reflective of a causal effect of LDL‐c on CAD risk. On the contrary, although HDL‐c is observationally inversely associated with CAD risk, interventions to raise HDL‐c concentrations (Schwartz et al., 2012) and focused Mendelian randomization investigations (Voight et al., 2012) suggest that there is not even a moderate causal effect of HDL‐c on CAD risk. However, a more liberal Mendelian randomization analysis including genetic variants associated with other lipid fractions (in particular, triglycerides) did suggest a causal effect of HDL‐c on CAD risk (Holmes et al., 2015). As a proof of concept example, we use summary data from the GLGC on genetic associations with lipid fractions, and from CARDIoGRAM on associations with CAD risk, and perform the analysis methods discussed in this paper to see whether the causal
