Facility data from the 2010 National Survey of Residential Care Facilities (NSRCF), conducted by the Centers for Disease Control and Prevention’s National Center for Health Statistics, Division of Health Care Statistics, were used for these analyses. To be eligible for the study, RCFs must be licensed, registered, listed, certified, or otherwise regulated by the state; have four or more licensed, certified, or registered beds; have at least one resident currently living in the facility; and provide room and board with at least two meals a day, around-the-clock on-site supervision, and help with personal care such as bathing and dressing or health-related services such as medication management. These facilities serve a predominantly adult population. Facilities licensed to serve the mentally ill or developmentally disabled populations exclusively are excluded. Nursing homes are also excluded unless they have a unit or wing meeting the above definition and residents can be separately enumerated.
The 2010 NSRCF used a stratified two-stage probability sample design. The first stage was the selection of RCFs from the sampling frame representing the universe of RCFs. The second stage was the selection of current residents. For the 2010 NSRCF, 3,605 RCFs were sampled with probability proportional to facility size. Interviews were completed with 2,302 facilities, for a first-stage facility unweighted response rate of 79% and a weighted response rate (for differential probabilities of selection) of 81%. Highlights about residents in RCFs will be provided in a forthcoming report. More details about sampling design and data collection are available elsewhere (8).
Two other data sources were used to compare the supply of residential care beds with that of nursing home beds per 1,000 persons aged 85 and over included in the Summary. 2010 Census Summary File 1 was used to estimate the number of persons aged 85 and over in the United States. Centers for Medicare & Medicaid Services Online Survey, Certification and Reporting (CMS OSCAR) data of December 2010 were used for the number of nursing home beds in the United States.
Differences among subgroups were evaluated using chi-square and t tests. If chi-square tests were statistically significant, a post hoc t-test procedure was used to make pairwise comparisons of different facility sizes (Figures 1-3, 5, and 6). Significant results from the post hoc procedure are reported here. Differences between regions were evaluated with t tests for rates per 1,000 persons aged 85 and over (Figure 4). All significance tests were two-sided using p < 0.05 as the level of significance. Terms such as “no significant differences” are used to denote that the estimates being compared are not significantly different statistically. The difference between any two estimates is mentioned in the text only if it is statistically significant. Lack of comment regarding the difference between any two statistics does not necessarily suggest that the difference was tested and found not to be statistically significant. In addition, a weighted least-squares technique was used (9) to test linear trends across four facility sizes. A downward or upward trend is mentioned if it is statistically significant. Data analyses were performed using the SAS-callable SUDAAN version 10.0 statistical package (10). Cases with missing data were excluded from the analyses on a variable-by-variable basis. Variables used in this report had a weighted percentage of cases with missing data between 0.1% and 0.6%. Because estimates were rounded, individual estimates may not sum to totals.