Skip to main content

Table 1 Overview whether the estimators deal with the possible sources of bias

From: Survival analysis for AdVerse events with VarYing follow-up times (SAVVY)—estimation of adverse event risks

 

Accounts for censoring

Makes no constant hazard assumption

Accounts for CEs

Incidence proportion

No

Yes

Yes

Probability transform incidence density ignoring CEs

Yes

No (AE Hazard)

No

1-Kaplan-Meier

Yes

Yes

No

Probability transform incidence density accounting for CEs

Yes

No (AE and CE Hazard)

Yes

Death only AJE

Yes

Yes

Yes (Death only)

Gold-standard AJE

Yes

Yes

Yes