National Hospital Discharge Survey Sample Design

The National Hospital Discharge Survey (NHDS) is a continuing nationwide sample survey of short-stay hospitals in the United States. The scope of NHDS encompasses patients discharged from noninstitutional hospitals, exclusive of military and Department of Veterans Affairs hospitals, located in the 50 States and the District of Columbia. Only hospitals having six or more beds for in-patient use are included in the survey and before 1988 those in which the average length of stay for all patients was less than 30 days. In 1988 the scope was altered slightly to include all general and children’s general hospitals regardless of the length of stay.

The original sample was selected in 1964 from a frame of short-stay hospitals listed in the National Master Facility Inventory. A two-stage stratified sample design was used, and hospitals were stratified according to bed size and geographic region. Sample hospitals were selected with probabilities ranging from certainty for the largest hospitals to 1 in 40 for the smallest hospitals. Within each sample hospital, a systematic random sample of discharges was selected from the daily listing sheet. Initially, the within-hospital sampling rates for selecting discharges varied inversely with the probability of hospital selection so that the overall probability of selecting a discharge was approximately the same across the sample. Those rates were adjusted for individual hospitals in subsequent years to control the reporting burden of those hospitals.

In 1988 NHDS was redesigned. The sample was selected from a frame of short-stay hospitals listed in the 1987 SMG Hospital Market Data Base. The hospitals with the most beds and/or discharges annually were selected with certainty, but the remaining sample was selected using a three-stage stratified design. The first stage was a sample of Primary Sampling Units (PSU’s) used by the National Health Interview Survey. Within PSU’s, hospitals were stratified or arrayed by abstracting status (whether subscribing to a commercial abstracting service) and within abstracting status arrayed by type of service and bed size. Within these strata and arrays, a systematic sampling scheme with probability proportional to the annual number of discharges was used to select hospitals. The rates of systematic sampling of discharges within hospitals vary inversely with probability of hospital selection within PSU. Discharge records from hospitals submitting data via commercial abstracting services and State data systems (approximately 45 percent of sample hospitals in 2006) were arrayed by primary diagnoses, patient sex and age group, and date of discharge before sampling. Otherwise, the procedures for sampling discharges within hospitals is the same as that used in the prior design.

The hospital sample was updated in 1991, 1994, 1997, 2000, 2003, and 2006 to allow the inclusion or exclusion of hospitals that opened or changed their eligibility status since the previous sample update. The additional hospitals were added at the end of the list for the strata where they belonged, and the systematic sampling was continued as if the additional hospitals had been present during the initial sample selection. Hospitals that were no longer NHDS-eligible were deleted. A detailed description of the design and development of the NHDS is included in “Design and Operation of the National Hospital Discharge Survey: 1988 Redesign” [PDF – 1.3 MB], Vital and Health Statistics Series 1, Number 39.