I /think/ that the way likelihood of HIV/AIDS is calculated needs to take population into account in a better way. Fun comparison, though!