Study population and study design:
We analyzed NIS data from January 2015 to December 2017. The study population was selected by using the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) and International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) codes. We selected patients 18 years of age and above for the purpose of our study. Age was further divided into three groups, <65, 65-74 and ≥75. Patients implanted with a WATCHMAN were identified by ICD-9 code of 37.90 and ICD-10 code of 02L73DK. Baseline characteristics and key complications were identified (ICD codes for complications provided in supplement), as previously described (13). Hospital outcomes including in-patient mortality, discharge disposition, length of stay (LOS) and cost of hospitalization (inflation adjusted) were derived. For the computation of hospitalization costs, cost-to-charge ratio files from NIS were utilized.
Complication rate trends were analyzed over the study years. The primary outcome of the study was the prevalence of Watchman implantation over our study period. Additional outcomes analyzed including associated complications, in-hospital mortality and resource utilization (including LOS, cost of stay and discharge disposition to home, short term care, long-term care or home with institutional care or home health).