diff --git a/book/P2C8_rank.qmd b/book/P2C8_rank.qmd
index a6f0229..c2c3eb4 100644
--- a/book/P2C8_rank.qmd
+++ b/book/P2C8_rank.qmd
@@ -283,6 +283,7 @@ $$
 
 In the presence of interval censoring, estimation depends on more complex estimators such as the nonparametric maximum likelihood estimator.
 As a more simplistic (yet sometimes realistic on average) alternative, one could impute the outcome time as the midpoint between the interval margins, $t_i = (l_i + r_i)/2$ [@Gadda2016].
+
 As left-censoring is a specialised case of interval-censoring (with $l_i = 0$) the above formulae can be used to handle left-censored data as well.
 As a simpl