|
Abstract:
|
Percutaneous endoscopic gastrostomy (PEG) is a procedure to insert a tube into the stomach to assist oral feeding. For patients diagnosed with amyotrophic lateral sclerosis (ALS), it is believed that PEG is beneficial and, generally, that earlier insertion is preferable. However, gathering clinical evidence to support these beliefs is challenging because controlled clinical trials are not feasible and clinical endpoints are confounded with PEG in observational data. Moreover, the confounders are time-varying and time to PEG insertion may be only partially observed. We show how one can view this problem as an adaptive treatment length regimen, that is, a special type of dynamic treatment regime with unique features. Then, we compare mean endpoints across different treatment length regimens through a new sequential estimation procedure that models a series of conditional distributions. We show that our estimator is consistent and asymptotically normal for the causal estimand and explore its finite sample properties in simulation studies. In real data, we find that earlier PEG does not reduce 4-year mortality and, in fact, our analysis suggests the opposite trend may be more likely.
|
