Overview of Recent Developments for Interval-Censored Data - Interval-Censored-Time-to-Event Data: Methods and Applications - page 17

models only apply to limited situations and a lot of more work is needed for

the problem.

1.6 Analysis of Multivariate or Clustered Interval-

Censored Data

So far the discussion has focused on independent or univariate interval-
censored data. In other words, it has been assumed that there exists only

one failure time variable of interest and all samples are independent of each

other. Sometimes the failure times of interest are clustered into small groups

or some study subjects are related, such as siblings, families, or communi-

ties. The subjects from the same cluster or group usually share certain unob-

served characteristics and their failure times tend to be correlated as a result.

Siblings, for example, share the same genetic and environmental influences

(Jonker et al. (2009)). In other words, for such data, the failure times within

the same cluster are dependent, while the failure times from different clus-

ters can be regarded as independent. With the existence of interval-censoring,

such failure time data are commonly referred to as clustered interval-censored

data. It is apparent that for the analysis of such data, one needs different

inference procedures than those discussed above, and one key and important

feature of these different procedures is that they need to take into account the

correlation among the failure time variables.
A related and special case of clustered interval-censored data is multi-

variate interval-censored data, which arise if a survival study involves several

related failure time variables of interest and they may suffer interval-censoring.

In this case, the related failure times can be seen a cluster and we have clus-

tered interval-censored data with the same cluster sizes, while for general clus-

tered data, the cluster size could differ from one to another. For the analysis