The sample design of NEISS is a stratified, probability
sample of all United States hospitals that had at least six beds and provided 24-hour emergency services. There were four hospital-sized strata (defined as very large, large, medium, and small, based on the number of annual ED visits) and one children's hospital stratum. From 1993 through 1996, there were 91 NEISS hospital EDs in the sample. In
1997, the sampling frame was updated so that in 1997 through 1999 the sample included 101 NEISS hospital EDs. In 2000-2001, one NEISS hospital dropped out of the system, so there were 100 NEISS hospital EDs in the sample. In 2002, another hospital dropped out of the system, so there were 99 NEISS hospital EDs in the sampling frame. In 1997, CPSC collected firearm-related cases using the "old" and "new"
NEISS hospital samples for a nine-month period. This dataset includes data from the "new" sample. The overlapping "old" sample is not included. Comparisons of weighted estimates based on the "old" and "new" samples indicated a difference of about 1 percent in the overall
national estimate using these samples. The characteristics of firearm-related cases from these two overlapping samples were also very similar.
United States hospitals providing emergency services.
nonfatal firearm-related injury incidents
administrative records data
computer-assisted self interview (CASI)