PMC

Not yet available.

Highlights

• This tutorial paper summarizes when and how to split the follow-up time in the analysis of epidemiological or clinical studies with follow-ups.
• Typical situations to split the follow-up time include estimating the outcome incidence rate (ratio) according to specific follow-up time periods.
• The proportional hazards assumption in Cox models can also be assessed by splitting the follow-up time.
• This tutorial also covers the situations to deal with time-varying exposures and confounders, as well as self-controlled case-series analysis.
• By splitting the follow-up time, the “wide-format” type data is transformed into the “long-format” type data.

Selected Result