When calibrating NFC item parameters in book, grmCat repeatedly prints a warning NaNs produced. I think its because it's in an R chunk, but I'm not sure. Works fine when written directly in the console/in an R script. So potentially not really an issue with catSurv, but still storing this as an item on our to-do list.