MedicalResearch.com Interview with:
Rohan Khera MD
Division of Cardiology
University of Texas Southwestern Medical Center
MedicalResearch.com: What is the background for this study? What are the main findings?
Response: An increasing number of studies have used administrative claims (or billing) data to study in-hospital cardiac arrest with the goal of understanding differences in incidence and outcomes at hospitals that are not part of quality improvement initiatives like the American Heart Association’s Get With The Guidelines-Resuscitation (AHA’s GWTG-Resuscitation). These studies have important implications for health policies and determining targets for interventions for improving the care of patients with this cardiac arrest, where only in 1 in 5 patient survive the hospitalization.
Therefore, in our study, we evaluated the validity of such an approach. We used data from 56,678 patients in AHA’s GWTG-Resuscitation with a confirmed in-hospital cardiac arrest, which were linked to Medicare claims data. We found:
(1) While most prior studies have used a diagnosis or procedure code alone to identify cases of in-hospital cardiac arrest, we found that the majority of confirmed cases in a national registry (AHA’s GWTG-Resuscitation) would not be captured using either administrative data strategy.
(2) Survival rates using administrative data to identify cases from the same reference population varied markedly and were 52% higher (28.4% vs. 18.7%) when using diagnosis codes alone to identify in-hospital cardiac arrest.
(3) There was large hospital variation in documenting diagnosis or procedure codes for patients with in-hospital cardiac arrest, which would have consequences for using administrative data to examine hospital-level variation in cardiac arrest incidence or survival, or conducting single-center studies to validate this administrative approach.
MedicalResearch.com: What should readers take away from your report?
Response: Our study highlights the challenges of using administrative billing data to conduct research on in-hospital cardiac arrest. Therefore, in the future, investigators should limit the use of billing data to study in-hospital cardiac arrest and readers should evaluate prior studies using this approach against other sources of data.
MedicalResearch.com: What recommendations do you have for future research as a result of this study?
Response: Future research should not use administrative billing data to study in-hospital cardiac arrest. The use of dedicated registries may be more appropriate.
Disclosure: I am supported by funding from the National Institutes of Health
MedicalResearch.com: Thank you for your contribution to the MedicalResearch.com community.
Khera R, Spertus JA, Starks MA, Tang Y, Bradley SM, Girotra S, Chan PS. Administrative Codes for Capturing In-Hospital Cardiac Arrest. JAMA Cardiol. Published online September 06, 2017. doi:10.1001/jamacardio.2017.2904
Note: Content is Not intended as medical advice. Please consult your health care provider regarding your specific medical condition and questions.