Detecting Bias Arising from Delayed Recording of Time

<jats:title>Summary</jats:title> <jats:p>Sometimes in studies of the dependence of survival time on explanatory variables the natural time origin for defining entry into study cannot be observed and a delayed time origin is used instead. For example, diagnosis of disease may in some patients be made only at death. The effect of such delays is investigated both theoretically and in the context of the England and Wales National Cancer Register.</jats:p>