Mode of Collection:
Manual sample selection and abstraction of inpatient medical records by field personnel or automated data collection through the purchase of electronic files from commercial abstracting sources, States, or hospitals.
Patient information collected includes demographics, length of stay, diagnoses, and procedures. Hospital characteristics collected include region, ownership, and bed size.
The survey design covers the 50 states and the District of Columbia. The survey includes nonfederal hospitals with an average length of stay of less than 30 days for all inpatients, general hospitals, and children hospitals. Hospital units of institutions (e.g., prisons) and hospitals with fewer than six inpatient beds are not included.
Hospitals are selected using a modified three-stage stratified design. Units selected at the first stage consist of either hospitals or geographic areas (primary sampling units (PSUs)). Hospitals within PSUs are selected at the second stage. Strata at this stage are defined by geographic region, PSU size, abstracting service status, and hospital specialty-size groups. Within these strata, hospitals are selected with probabilities proportional to their annual number of discharges. At the third stage, a random sample of discharges is selected. The NHDS hospital sample is updated every 3 years.
Response Rates and Sample Size:
In 2006, 501 hospitals were selected; 478 were within scope, 438 participated (92%), and data were collected from medical records for approximately 376,000 discharges.
Annual modifications to the International Classification of Diseases, ninth revision, Clinical Modification (ICD-9-CM) may affect diagnosis and procedure categories. The NHDS was redesigned in 1988. Caution is required in comparing trend data from before and after the redesign.