Interval Censoring

interval-censoringsurvival-analysisnonparametric-maximum-likelihoodproportional-hazardsaccelerated-failure-timesemiparametriccensoring

Definition

Interval censoring arises in survival analysis when an event time TT is not observed exactly but is only known to lie in an interval (L,R](L,R] — typically because the subject is inspected at discrete times and the event is known only to have occurred between two inspections. It generalizes right censoring (the interval is (L,)(L,\infty)) and left censoring (the interval is (0,R](0,R]), and it degrades both estimation efficiency and the tidy structure of right-censored inference (Anderson-Bergman 2017).

Key Ideas

How It Works

  1. Encode each observation as an interval (Li,Ri](L_i,R_i] (with Ri=R_i=\infty for right-, Li=0L_i=0 for left-censored).
  2. Write the likelihood in terms of the survival function: each subject contributes S(Li)S(Ri)S(L_i)-S(R_i), the probability mass the event falls in its interval.
  3. Fit the chosen model — NPMLE (Turnbull) for an assumption-free survival curve, a semiparametric PH/PO model for covariate effects with a flexible baseline, or a parametric AFT/PH/PO model for efficiency.
  4. Assess fit by comparing parametric baselines to the nonparametric estimate; obtain uncertainty by bootstrap for the semiparametric fits.

Why It Matters

Open Questions

Related