There are currently two separate, yet similar, database products offered. The first is based upon the primary diagnosis upon release from hospital and the second based on procedures undertaken by emergency departments. While there is much overlap in terms of disease, these measure two separate types of health care utilization.
For each of the current and five year projected years, this dataset includes the estimated number of hospital discharges and the total length of the stays to be expected in a year by diagnostic category. These are standard categories as defined by the ICD-10 (International Classification of Diseases), of which there are 822, with summarizations to the MDC (Major Diagnostic Category) , and MDC by type (Surgical, Medical).
For each of these categories, an estimate of the number of expected cases for the year and the total length of hospital stays associated with them. These are available for block group and higher. From these building blocks can be constructed such measures as average hospital stays for individual or groups of categories, estimated total demand in terms of bed-nights, and so forth.
The primary data source is the Nationwide Inpatient Sample (NIS) for 2010, available from H-CUP (Healthcare Cost and Utilization Project) of the US Department of Health and Human Services.
Emergency Department Diagnoses/Procedures
This dataset includes estimates of the total number of procedures by category likely to be performed on residents of any particular area in a year. It is not a count of individual patients, as many incoming patients have multiple procedures.
The data source is the Nationwide Emergency Department Sample (NEDS) for 2010, available from H-CUP (Healthcare Cost and Utilization Project), US Department of Health and Human Services.
The database consists of data from 29 states, representing a total of 961 emergency departments. A total of 28.6 million “events” are recorded, roughly 20% of the estimated 130 million events nationwide per year. The sampling frame is similar, utilizing US Census region, urban/rural location, teaching status, ownership/control. Rather than use the number of beds, the sample utilizes whether the facility is a trauma center.
The results are presented at the block group level and above, again to four digits of significance. There are a total of 259 procedural categories, organized into 17 groups and a total. In addition, a tally of total trauma and multiple trauma cases is created.
Note that it is not possible to add together discharge and emergency department numbers, since many emergency department visits culminate in admission to the hospital. Since the admission rate varies by condition, it is not possible to add numbers together.