Estimation of Relative and Absolute Risk in a Competing-Risk Setting Using a Nested Case-Control Study Design: Example From the ProMort Study