Limitations:
Our study results should be interpreted in the context of following key
limitations. (1) NIS is an administrative, claims based database that
relies on ICD coding system. These codes can be subjected to errors,
however, HCUP quality control measures are routinely instituted that
mitigate such concerns (9). Additionally, the positive predictive value
of ESRD codes is close to 96% (24). (2) There are no specific ICD codes
for in-hospital arrest and previous studies have utilized demonstration
of CPR as evidence of in-hospital arrest and we have replicated the same
methodology in our current analysis. (3) NIS does not inform on detailed
management of SCA and specifically no data is collected on quality of
CPR and other measures that are practiced as part of advanced life
support. (4) NIS censors data gathering at discharge and patients are
not followed longitudinally. ESRD patients who sustained SCA have been
shown to have poor survival on follow-up studies and that cannot be
investigated from NIS dataset (25).