Sample Design, 1996

The sample design for the 1996 National Home and Hospice Care Survey (NHHCS) was a stratified two-stage probability design. The first stage consisted of selecting a stratified sample of agencies. Each agency was placed into 1 of 24 strata based on type of agency (home health agency, hospice, or mixed agency), metropolitan statistical area (MSA) status (has an MSA code versus no code), and region (Northeast, Midwest, South, and West).  MSA is defined by the U.S. Office of Management and Budget on the basis of the 1980 Census. Within these sampling strata, agencies were arrayed by four types of ownership (profit, nonprofit, government, and unknown), three types of certification status (certified by Medicare and/or Medicaid, not certified, and unknown), State, MSA code, county, zip code, and size (number of current patients).

The second stage of sample selection, sampling of six current patients and six discharges within each agency, was done using a sample selection table to obtain systematic probability samples of current patients and of discharges. The patients and discharges were selected from lists constructed for each agency at the time of the interview. Current patients were defined as those patients who were on the rolls of the agency as of midnight on the day immediately before the date of the survey. Discharges referred to those patients who were discharged from care by the home health agency or hospice during a designated month between October 1995 and September 1996. Discharges that occurred because of death were included.