Gaussian Graphical models (GGM) are widely used to estimate the network structures in many applications ranging from biology finance. In practice, data is often corrupted by latent confounders which biases inference of underlying true graphical structure. this paper, we compare and contrast two strategies for with confounders: variables (LVGGM) PCA-based removal confounding (PCA+GGM). While the...

To the Editor: With some degree of amazement, we read the article by Kashy et al.1 on the influence of 6% hydroxyethyl starch (HES) 670/0.75 (Hextend; Hospira Inc., USA) on perioperative acute kidney injury in patients undergoing noncardiac surgery. The data are derived from a database of more than 120,000 patients treated in Cleveland hospitals, in which 6% HES 670/0.75 was the most commonly u...