National Hospital Care Survey Demonstration Projects: Stroke Inpatient Hospitalizations.

Objective-This report shows the analytical potential of the National Hospital Care Survey (NHCS) through a demonstration of the use of its data to examine inpatient (IP) discharges and ambulatory visits for stroke. Unweighted data of IP and ambulatory encounters from the 2014 NHCS are linked to records from the 2014 and 2015 National Death Index (NDI). Methods-For the 2014 NHCS, 94 hospitals provided IP administrative claims data and 88 provided ambulatory (emergency and outpatient department) claims data. Although these data are not nationally representative, the survey provides unique opportunities to study health conditions such as stroke, because all IP discharges and ambulatory encounters from participating hospitals are collected for a 12-month period. The collection of patient identifiers (e.g., patient name, Social Security number, and date of birth) allows for linkage to outside data sources such as NDI, providing information on patient mortality after hospital discharge. Analyses examined stroke encounters across various hospital settings. Results-Approximately two-thirds of stroke IP discharges originated from hospital emergency departments. IP visits for stroke tended to originate from the IP department among younger patients, and from the emergency department for older patients. The likelihood of postdischarge mortality was higher for stroke patients aged 85 and over. Thirteen percent of stroke patients aged 85 and over died within 30 days of an IP discharge compared with almost 6% of stroke patients aged 75-84. This study highlights the unique analytical capabilities of NHCS.