Abstract
Nationwide nursing home private-pay prices at the facility-level have not been available for researchers interested in studying this unique health care market. This study presents a new data source, Caregiverlist, for private-pay prices for private and semi-private rooms for 12,000 nursing homes nationwide collected between 2008 and 2010. We link these data to publicly available national nursing-home level data sets to examine the relationship between price and nursing home characteristics. We also compare private-pay prices to average private-pay revenues per day for California nursing homes obtained from facilities’ financial filings. On average, private-pay prices were $224 per day for private rooms compared to $197 per day for semi-private rooms. We find that nursing homes that are non-profit, urban, hospital-based, have a special care unit, chain-owned, and have higher quality ratings have higher prices. We find average revenues per day in California to be moderately correlated with prices reported by Caregiverlist.
Keywords: nursing home, long-term care, price, private pay
INTRODUCTION
Over half of older Americans are predicted to stay at least one night in a nursing home during their lifetime (Hurd, Michaud and Rohwedder 2017). The primary payer for nursing home care is Medicaid, yet in order to qualify for this coverage many older adults have to spend down their savings paying for skilled nursing home care (Kaiser Family Foundation 2017). Out-of-pocket annual costs for nursing home care are substantial and vary widely by state, ranging from $63,510 in Oklahoma to $330,873 in Alaska (Genworth Financial 2018). Since Medicare only pays for post-acute care stays in the nursing home and only 11% of Americans have long-term care insurance, the out-of-pocket cost of nursing home care poses a major financial risk to many individuals (Brown and Finkelstein 2007).
Historically, researchers interested in private-pay prices at the nursing home level have either used state-collected survey data (Nyman [1985, 1988a, 1988b, 1989a, 1989b, 1994]; Mukamel and Spector 2002; Clement, Bazzoli and Zhao 2012; Bowblis 2013; Clement and Khushalani 2015; Loomer 2018) or state-administrative cost reports to derive the average revenue per self-pay resident (Grabowski 2004; Huang and Hirth 2016; Gandhi 2019; Huang, Banaszak-Holl, Yuan, & Hirth 2019). These two sources of private-pay nursing home price data have strengths and limitations for use in research. Studies based on state-level survey data are difficult to generalize to other states or to the national level due to variations in policy and supply of alternative long-term services and supports across states. On the other hand, state-level survey data often has detail that allows the research to differentiate between private and semi-private room prices. Studies based on cost report data from numerous states can improve generalizability, but it is not possible to distinguish between private and semi-private prices using cost report data. Also, depending on how the dollars are attributed within the cost report, average private-pay revenue per day can potentially be biased upward if payments by patients on Medicare or Medicaid for non-covered goods and services are categorized as private-pay revenues.1 State survey data are also often limited to only a few years, whereas cost report data are typically available for a longer time horizon, making panel data analysis possible.
As a resource to facilitate nursing home price comparisons for consumers, Caregiverlist used a “secret shopper” phone survey to collect private-pay prices for private and semi-private beds for 12,000 nursing homes nationally, about 75% of all nursing homes, and provided them online for consumer use. In this study, we explore the use of these online private-pay price data as a new source of information for researchers.
NEW CONTRIBUTION
This study provides an introduction to a new data source for examining nursing home level private-pay price data across the United States. This is the first U.S. study that we are aware of to utilize national nursing home price data collected via an audit (i.e., “secret shopper”) approach. Surveyors were instructed to call nursing facilities posing as a potential customer seeking a nursing home for an elderly relative and solicit specific pricing information on private and semi-private rooms.2 We assess the validity of these private price data by examining how they relate to nursing home characteristics such as quality ratings and ownership status. We also assess the validity of the Caregiverlist price data by comparing them to a measure of average revenue per day from the California nursing home cost reports.
DATA
Caregiverlist (www.caregiverlist.com) is a national senior care referral and career website that provides consumer-friendly interface for comparing nursing homes and other long-term care services and supports nationwide. Caregiverlist is an online source for nursing home private-pay per diems covering all fifty states at the facility-level. Since nursing home prices are typically only available to consumers via phone, Caregiverlist used an audit (or “secret shopper”) approach to call each nursing home and request private-pay quotes for private (single) and semi-private (double) bedrooms. Callers were instructed to pretend to be a potential customer helping an elderly family member to choose a nursing home in order to increase the likelihood that the quotes are accurate. More specifically, the call script asked “Hello, I am calling for the daily cost of a private or semi-private room in your nursing home for a senior relative. Do you have this information or could you connect me with the person who could provide this information?” Then the callers recorded details for the private and/or semi-private rate or notes stating that there was a monthly rate or multiple prices.
These data were collected from 2008 to 2010. The data used for this analysis was the first sample collected between 2008 and 2010, because the updates to prices over time were not well documented in newer version of the Caregiverlist data we accessed for this study. Some of the nursing homes provided ranges for prices instead of a single price. In this case, we took the median price of the range.
One of the many strengths of this novel data set is that it can be easily merged to a number of different publicly available data sets at the facility-level by using the provider number. We merged the price data to sources publicly available data often used in nursing home research: Long-term Care: Facts on Care in the US (LTCFocus), Nursing Home Compare, and the Area Health Resource File. The LTCFocus data are a product of the Shaping Long-Term Care in America Project being conducted at the Brown University Center for Gerontology and Healthcare Research and supported, in part, by the National Institute on Aging (1P01AG027296). LTCFocus includes a number of nursing-home level measures compiled from various data sources including the Online Survey, Certification and Reporting system data, Minimum Data Set and the Area Health Resource File. Nursing Home Compare is an online comparison tool created by the Centers for Medicare and Medicaid Service (CMS) to assist consumers in comparing facilities along several dimensions. We use the 5-star overall ratings obtained from a January 2010 snapshot of Nursing Home Compare. We categorized regions using Census definitions. An indicator for Rural/Urban at the county-level was obtained from the Area Health Resource File.
Lastly, in order to compare Caregiverlist’s private prices to those measured through other means, we merged the Caregiverlist data to private price data for California nursing homes computed using the Long-Term Care Facility Financial Data from the California Office of Statewide Health Planning and Development (OSHPD). Unlike the other datasets we use, these California data do not include CMS provider identification numbers. We therefore use a linkage file created by Huang and Hirth (2016) and augmented by Gandhi (2019) using both automated and manual records-linkage methods applied to facility characteristics such as names and addresses.
ANALYSIS
First, we provide descriptive statistics of the nursing homes that are present in the Caregiverlist data, separated by the availability of data on private and semi-private room rates in the Caregiverlist data. We provide summary statistics of the private and semi-private room rates nationally and for different subgroups of nursing homes (e.g. for-profit vs. non-profit; rural vs. urban). We then aggregate the price data up to the state-level to examine geographic differences across the country.
We use multivariate regression analysis to examine how nursing home characteristics are associated with price. We estimate the regressions separately for private and semi-private room prices. We model the relationship between price and nursing home location in two ways: 1) state fixed effects including an indicator for urban/rural status; 2) county fixed effects excluding the urban/rural indicator and the counties with only one nursing home. Standard errors are clustered at the county-level in both specifications. Our regression also includes the following facility characteristics: ownership type (For-Profit, Non-Profit, Government), urban, hospital-based, with any special care unit (e.g. dementia special care unit), chain status, county Herfindahl-Hirschman index (HHI) measured from 0 (less concentrated) to 1 (more concentrated), CMS Nursing Home Compare 5-star overall rating, proportion of beds occupied (occupancy %), number of beds (1–60, 61–120, 121–180, and 181+), proportion of payers whose primary source is Medicare, and proportion of payer whose primary source is Medicaid, and the OSCAR-based acuity index (Arling and Daneman 2002). We exclude Minnesota and North Dakota from the regression analyses due to their rate equalization laws that prohibit semi-private private-pay room rates to be above Medicaid rates (Minn. Stat. §256B.48, Subd. 1(a); North Dakota Department of Human Services, Rate Setting Manual for Nursing Facilities).3 Additionally, a number of nursing homes charge substantially more than the average price so we excluded outliers at the 1st and 99th percentile from the regression. Most of these 99th outlier facilities were hospital-based facilities (63%) or provided specialty care. For example, two nursing homes in New York City, “St. Mary’s Center Inc.” and “Rivington House the Nicholas A Rango HCF”, provide services only to adults living with human immunodeficiency virus/acquired immunodeficiency syndrome and are listed in the data as charging $650 and $750 per day for a private room, respectively. We do not explicitly exclude specialty or hospital-based nursing homes in the main analysis, but some of them are excluded due to outlier status. We exclude Alaska and Hawaii due to their substantially different nursing home markets. We conduct three sensitivity analyses reported in the Appendix: 1) including MN and ND 2) excluding Government NHs, and 3) excluding Hospital-based NHs.
Finally, we approximate private rates in California using average revenues per day calculated from an average of the 2008, 2009 and 2010 cross-section of facility financials published by California’s OSHPD. Specifically, we sum the self-pay routine and ancillary revenues and divide them by the number of self-pay days in the facility for each year (2008–2010). Then we create an average of the three years. We assess the agreement of these computed private rates with the Caregiverlist data by computing the Pearson correlation coefficient and comparing the average prices from Caregiverlist to the prices computed using OSHPD data.
RESULTS
The sample includes 10,685 nursing homes with private room prices, 11,638 with semi-private room prices, 10,275 with both types of prices and 3,743 with no price data (Table 1). The characteristics of the sub-sample with both prices are similar to the rest of the sample, but those without pricing data are more likely to be government, non-profit and smaller. A little less than three-fourths of the sample with prices are for-profit nursing homes (73%), and more than half are chain owned (57%). The majority of the study sample is in the Midwest (36%) and South (34%) Census regions.
Table 1:
Descriptive statistics of nursing homes with and without Caregiverlist prices available
| Private room | Semi-private room | Both Private and Semi-Private room | No Price data available | |
|---|---|---|---|---|
| N | 10,685 | 11,638 | 10,275 | 3,743 |
| Ownership type, N (%) | ||||
| For-Profit | 7,732 (72.4%) | 8,621 (74.1%) | 7,536 (73.3%) | 1,980 (53.7%) |
| Non-Profit | 2,500 (23.4%) | 2,510 (21.6%) | 2,318 (22.6%) | 1,343 (36.4%) |
| Government | 453 (4.2%) | 507 (4.4%) | 421 (4.1%) | 366 (9.9%) |
| Urban, N (%) | 7,218 (67.6%) | 7,984 (68.6%) | 6,935 (67.5%) | 2,556 (69.3%) |
| Hospital-Based, N (%) | 445 (4.2%) | 432 (3.7%) | 375 (3.6%) | 551 (14.9%) |
| Any Special Care Unit, N (%) | 2,094 (19.6%) | 2,246 (19.3%) | 2,028 (19.7%) | 627 (17.0%) |
| Chain, N (%) | 6,064 (56.8%) | 6,703 (57.6%) | 5,888 (57.3%) | 1,729 (46.8%) |
| County HHI, Mean (SD) | 0.2 (0.2) | 0.2 (0.2) | 0.2 (0.2) | 0.2 (0.3) |
| CMS Overall Rating, N (%) | ||||
| 1 Star | 2,222 (20.8%) | 2,443 (21.0%) | 2,173 (21.2%) | 523 (15.3%) |
| 2 Star | 2,252 (21.1%) | 2,512 (21.6%) | 2,184 (21.3%) | 585 (17.1%) |
| 3 Star | 2,326 (21.8%) | 2,541 (21.8%) | 2,243 (21.8%) | 668 (19.5%) |
| 4 Star | 2,655 (24.9%) | 2,839 (24.4%) | 2,541 (24.7%) | 974 (28.5%) |
| 5 Star | 1,224 (11.5%) | 1,296 (11.1%) | 1,128 (11.0%) | 667 (19.5%) |
| Occupancy %, Mean (SD) | 83.1 (14.2) | 83.1 (14.1) | 83.1 (14.1) | 81.7 (19.3) |
| Total # beds, N (%) | ||||
| 1–60 | 2,098 (19.6%) | 2,222 (19.9%) | 1,932 (18.8%) | 1,419 (38.4%) |
| 61–120 | 5,268 (49.3%) | 5,806 (49.9%) | 5,118 (49.8%) | 1,414 (38.3%) |
| 121–180 | 2,371 (22.2%) | 2,581 (22.2%) | 2,306 (22.4%) | 594 (16.1%) |
| 181+ | 948 (8.9%) | 1,029 (8.8%) | 919 (8.9%) | 265 (7.2%) |
| % Medicaid Share, Mean (SD) | 61.1 (20.7) | 62.2 (20.0) | 61.6 (19.9) | 53.8 (30.8) |
| % Medicare Share, Mean (SD) | 15.6 (14.4) | 15.2 (13.4) | 15.3 (13.4) | 16.7 (22.3) |
| Acuity Index | 11.0 (1.2) | 11.0 (1.2) | 11.0 (1.2) | 11.0 (1.9) |
| Region | ||||
| New England | 1,936 (18.1%) | 2,049 (17.6%) | 1,878 (18.3%) | 588 (15.9%) |
| Midwest | 3,827 (35.8%) | 3,980 (34.2%) | 3,651 (35.5%) | 1,046 (28.3%) |
| South | 3,653 (34.2%) | 4,004 (34.4%) | 3,535 (34.4%) | 1,277 (34.6%) |
| West | 1,269 (11.9%) | 1,605 (13.8%) | 1,211 (11.8%) | 780 (21.1%) |
Notes: Any Special Care Unit (e.g. dementia). HHI: Herfindahl Hirschman Index measured from 0 to 1. Occupancy %, % Medicaid share, % Medicare share measured from 0–100. Centers for Medicaid and Medicare Services (CMS) Star Rating is a 2010 snapshot of overall rating.
On average, the private-pay price per day for a private room is $224 compared to $197 for a semi-private room (Table 2). The national distribution is right-skewed with an interquartile range of $158 to $246 private room price and $140 to $211 for semi-private room prices (Appendix Figure 1). Rural nursing homes charge about $59 dollars less than urban nursing homes. Hospital-based nursing homes ($392) charge, on average, more than freestanding facilities ($217). Regionally, New England has the highest average prices at $307 compared to $209 in the Midwest, $190 in the South, and $240 in the West region. For the subsample of nursing homes with both private and semi-private room rates, the premium for a private room was on average $25/day. When examining state differences in private-pay prices, the three states with the highest annual median prices for a semi-private room are Alaska ($260,610), Connecticut ($118,625), and Hawaii ($116,800), and the three states with the lowest prices are Texas ($41,428), Missouri ($45,260), and Oklahoma (45,625) (Figure 1).
Table 2:
Caregiverlist Private-Pay Rates for Private and Semi-Private Room Per Day ($)
| Sample with either Private or Semi-Private room prices | Sample of NH with both Private and Semi-Private room prices | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Private room rate | Semi-Private room rate | Private room rate | Semi-Private room rate | Difference (Private-Semi-Private) | ||||||
| N | 10,685 | 11,638 | 10,275 | 10,275 | 10,275 | |||||
| Mean ($) | Median ($) | Mean ($) | Median ($) | Mean ($) | Median ($) | Mean ($) | Median ($) | Mean ($) | Median ($) | |
| 224 | 195 | 197 | 173 | 221 | 195 | 196 | 173 | 25 | 15 | |
| Ownership type | ||||||||||
| For-Profit | 216 | 190 | 190 | 170 | 215 | 190 | 189 | 170 | 26 | 15 |
| Non-Profit | 250 | 211 | 222 | 188 | 244 | 210 | 220 | 188 | 23 | 15 |
| Government | 222 | 189 | 205 | 173 | 213 | 185 | 197 | 170 | 16 | 10 |
| Urban | 243 | 210 | 212 | 185 | 241 | 210 | 213 | 185 | 28 | 19 |
| Rural | 184 | 165 | 165 | 149 | 181 | 165 | 162 | 148 | 18 | 10 |
| Hospital-Based | 392 | 249 | 352 | 218 | 365 | 230 | 339 | 219 | 26 | 10 |
| Free-standing | 217 | 194 | 191 | 172 | 216 | 194 | 191 | 172 | 25 | 15 |
| Any Special Care | 232 | 200 | 210 | 178 | 230 | 200 | 206 | 178 | 24 | 15 |
| Unit | ||||||||||
| No Special Care | 222 | 193 | 194 | 171 | 219 | 193 | 194 | 171 | 25 | 15 |
| Unit | ||||||||||
| Chain | 220 | 190 | 194 | 170 | 218 | 190 | 192 | 170 | 26 | 15 |
| Independent | 229 | 200 | 202 | 178 | 226 | 200 | 203 | 179 | 23 | 15 |
| CMS Overall Rating | ||||||||||
| 1 Star | 214 | 186 | 190 | 165 | 214 | 187 | 188 | 165 | 25 | 15 |
| 2 Star | 216 | 195 | 191 | 172 | 214 | 195 | 189 | 172 | 25 | 15 |
| 3 Star | 225 | 195 | 197 | 173 | 222 | 195 | 197 | 173 | 25 | 15 |
| 4 Star | 231 | 200 | 204 | 178 | 228 | 200 | 204 | 178 | 24 | 15 |
| 5 Star | 239 | 200 | 210 | 180 | 233 | 200 | 210 | 180 | 24 | 15 |
| Total # Beds | ||||||||||
| 1–60 | 235 | 187 | 203 | 168 | 225 | 185 | 202 | 168 | 23 | 13 |
| 61–120 | 208 | 187 | 182 | 167 | 187 | 206 | 182 | 166 | 24 | 15 |
| 121–180 | 231 | 205 | 203 | 180 | 231 | 205 | 204 | 180 | 28 | 18 |
| 181+ | 270 | 245 | 254 | 210 | 271 | 245 | 245 | 211 | 26 | 17 |
| Region | ||||||||||
| New England | 307 | 293 | 284 | 270 | 306 | 292 | 283 | 274 | 23 | 18 |
| Midwest | 209 | 180 | 186 | 159 | 206 | 180 | 183 | 157 | 23 | 15 |
| South | 190 | 175 | 163 | 152 | 188 | 175 | 164 | 155 | 25 | 14 |
| West | 240 | 210 | 199 | 185 | 231 | 210 | 198 | 185 | 34 | 20 |
Notes: Any Special Care Unit (e.g. dementia). Centers for Medicaid and Medicare Services (CMS) Star Rating is a 2010 snapshot of overall rating.
Figure 1.
Caregiverlist Median Private-Pay Semi-Private Room Annual Rates
Notes: These data were collected between 2008–2010. The darker green is the fourth quartile and the lighter green is the first quartile. Sample size for each state reported in Appendix Table 4.
We next examined how different characteristics were associated with private-pay prices in a regression framework (Table 3). Models 1 and 3 include state fixed effects. Our preferred specifications are Models 2 and 4, which include county fixed effects. In our preferred specifications, we find that non-profit nursing homes compared to for-profit nursing homes charge $7and $8 higher for private and semi-private room prices, respectively. Hospital-based nursing homes compared to freestanding nursing homes charge about $60 and $34 more in private and semi-private room prices, respectively. Chain-owned nursing homes charge $5 and $4 more for private and semi-private rooms, respectively, compared to independent nursing homes. A positive relationship exists between CMS Nursing Home Compare 5-star overall ratings and the semi-private room price. We find that facilities rated 5-star charge $5 higher for private room prices and charge $3 higher for semi-private room prices than 1-star rated nursing homes. Similar relationships were found in the specifications using state fixed effects.
Table 3:
Regression results
| Model | (1) | (2) | (3) | (4) | (5) | (6) | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Outcome | Private Room Price | Private Room Price | Semi-Private Room Price | Semi-Private Room Price | Private-semi private difference | Private-semi private difference | ||||||
| Coef. | SE | Coef. | SE | Coef. | SE | Coef. | SE | Coef. | SE | Coef. | SE | |
| Ownership type | ||||||||||||
| For-Profit | Ref. | Ref. | Ref. | Ref. | Ref. | Ref. | ||||||
| Non-Profit | 5.83*** | (1.67) | 6.76*** | (1.61) | 6.12*** | (1.18) | 7.70*** | (1.13) | −1.02* | (0.59) | −0.57 | (0.67) |
| Government | 4.81 | (3.78) | 6.67 | (4.30) | 5.03** | (2.45) | 8.05*** | (2.99) | −2.88** | (1.13) | −2.20 | (1.56) |
| Urban | 13.86*** | (1.57) | 10.56*** | (1.08) | 3.79*** | (0.64) | ||||||
| Hospital-Based | 56.50*** | (7.39) | 60.07*** | (8.68) | 30.71*** | (4.86) | 34.28*** | (5.88) | 1.40 | (1.60) | 2.17 | (2.32) |
| Any Special Care | 2.42* | (1.29) | 3.15** | (1.33) | 1.20 | (0.93) | 2.26** | (1.00) | 0.49 | (0.66) | 1.15 | (0.72) |
| Unit | ||||||||||||
| Chain | 3.62*** | (1.17) | 4.97*** | (1.30) | 3.23*** | (0.75) | 4.28*** | (0.85) | −0.12 | (0.52) | 0.23 | (0.59) |
| County HHI | −18.11*** | (3.06) | 10.69 | (11.75) | −11.82*** | (2.02) | 11.57 | (9.17) | −5.45*** | (1.12) | −0.99 | (3.46) |
| CMS Overall Rating | ||||||||||||
| 1 Star | Ref. | Ref. | Ref. | Ref. | Ref. | Ref. | ||||||
| 2 Star | 3.02** | (1.47) | 1.20 | (1.62) | 1.33 | (0.98) | 0.70 | (1.01) | −0.50 | (0.73) | −0.73 | (0.83) |
| 3 Star | 1.35 | (1.70) | −0.86 | (1.92) | 1.26 | (1.10) | −0.09 | (1.12) | −0.90 | (0.70) | −0.82 | (0.79) |
| 4 Star | 4.08** | (1.60) | 2.61 | (1.65) | 3.62*** | (1.09) | 2.25** | (1.09) | −1.27* | (0.70) | −1.04 | (0.76) |
| 5 Star | 8.53*** | (2.28) | 4.93** | (2.42) | 5.57*** | (1.55) | 3.30** | (1.52) | −0.66 | (0.93) | −0.82 | (1.04) |
| Occupancy % | 0.09* | (0.05) | 0.07 | (0.05) | 0.10*** | (0.03) | 0.08*** | (0.03) | 0.01 | (0.02) | −0.04 | (0.03) |
| Total # Beds | ||||||||||||
| 1–60 | Ref. | Ref. | Ref. | Ref. | Ref. | Ref. | ||||||
| 61–120 | 0.39 | (1.49) | 0.78 | (1.74) | 1.63* | (0.94) | 2.49** | (1.08) | −0.29 | (0.64) | −0.87 | (0.74) |
| 120–180 | 9.01*** | (1.73) | 6.80*** | (1.90) | 8.17*** | (1.19) | 7.14*** | (1.35) | 1.98** | (0.85) | 0.91 | (0.98) |
| 181+ | 21.50*** | (2.40) | 10.56*** | (2.61) | 19.64*** | (1.94) | 10.68*** | (1.73) | 4.78*** | (1.35) | 2.26 | (1.43) |
| % Medicaid Share | −0.34*** | (0.05) | −0.40*** | (0.10) | −0.20*** | (0.03) | −0.24*** | (0.04) | −0.18*** | (0.02) | −0.19*** | (0.02) |
| % Medicare Share | 0.83*** | (0.09) | 0.64*** | (0.10) | 0.58*** | (0.07) | 0.49*** | (0.07) | 0.08*** | (0.03) | 0.05* | (0.03) |
| Acuity Index | 2.60*** | (0.73) | 2.60*** | (0.83) | 2.41*** | (0.56) | 2.64*** | (0.65) | 0.36 | (0.26) | 0.09 | (0.30) |
| Constant | 164.18*** | (9.92) | 181.47*** | (10.91) | 137.43*** | (7.48) | 146.09*** | (8.30) | 28.18*** | (4.38) | 38.89*** | (4.69) |
| State FE | Y | N | Y | N | Y | N | ||||||
| County FE | N | Y | N | Y | N | Y | ||||||
| N | 10,248 | 9,454 | 11,161 | 10,357 | 9,206 | 8,391 | ||||||
| R2 | 0.57 | 0.69 | 0.70 | 0.79 | 0.21 | 0.37 | ||||||
Notes:
p < 0.10,
p < 0.05,
p < 0.01.
Excluding Outliers below 1% and above 99% and facilities from Alaska, Hawaii, Minnesota and North Dakota. Standard errors (SE) in parentheses were clustered at the county-level. HHI (Herfindahl-Hirschman Index) is measured at the county-level and ranges from 0 to 1. In Models 2 and 4, singleton counties are omitted. Occupancy %, % Medicaid, % Medicare measured from 0–100. Centers for Medicaid and Medicare Services (CMS) Star Rating is a 2010 snapshot of overall rating.
The characteristics of nursing homes in the California data set are shown in Table 4. Nursing homes with both OSHPD and Caregiverlist data were mostly for-profit and there were no hospital-based nursing homes. Caregiverlist private-pay prices and average revenues per day for California nursing homes is shown in Table 5. The NHs with both private room prices and OSHPD rates charged on average $246 for private rooms in Caregiverlist data and $224 using OSHPD prices. The NHs with both semi-private rates and OSHPD rates charged on average $193 for a semi-private room and $220 using OSHPD prices. That OSHPD rates fall between Caregiverlist’s private and semi-private rates is expected, because the OSHPD rate is computed using revenues from private pay patients occupying both private and semi-private rooms. The Pearson correlation4 between the average revenues per day and private room price is 0.45 and semi-private room is 0.40, indicating a moderate relationship.
Table 4.
Descriptive statistics of nursing homes in California with and without Cargiverlist price data
| OSHPD and Private room price | OSHPD and Semi-private room | No Caregiverlist Price data available | |
|---|---|---|---|
| N | 383 | 614 | 374 |
| Ownership type, N (%) | |||
| For-Profit | 336 (87.7%) | 559 (91.0%) | 312 (83.6%) |
| Non-Profit | 46 (12.0%) | 54 (8.8%) | 60 (16.1%) |
| Government | 1 (0.3%) | 1 (0.2%) | 1 (0.3%) |
| Urban, N (%) | 371 (96.9%) | 596 (97.1%) | 363 (97.3%) |
| Free-standing, N (%) | 383 (100%) | 614 (100%) | 374 (100%) |
| Any Special Care Unit, N (%) | 24 (6.3%) | 39 (6.4%) | 48 (12.9%) |
| Chain, N (%) | 196 (51.2%) | 321 (52.3%) | 200 (53.6%) |
| County HHI, Mean (SD) | 0.1 (0.1) | 0.1 (0.1) | 0.1 (0.1) |
| CMS Overall Rating, N (%) | |||
| 1 Star | 78 (20.4%) | 122 (19.9%) | 79 (21.6%) |
| 2 Star | 97 (25.3%) | 157 (25.6%) | 55 (15.0%) |
| 3 Star | 86 (22.5%) | 136 (22.1%) | 78 (21.3%) |
| 4 Star | 78 (20.4%) | 130 (21.2%) | 94 (25.7%) |
| 5 Star | 44 (11.5%) | 69 (11.2%) | 60 (16.4%) |
| Occupancy %, Mean (SD) | 86.3 (10.4) | 86.6 (10.0) | 86.4 (10.9) |
| Total # beds, N (%) | |||
| 1–60 | 92 (24.0%) | 141 (23.0%) | 110 (29.5%) |
| 61–120 | 196 (51.2%) | 334 (54.4%) | 165 (44.2%) |
| 121–180 | 65 (17.0%) | 102 (16.6%) | 69 (18.5%) |
| 181+ | 30 (7.8%) | 37 (6.0%) | 29 (7.8%) |
| % Medicaid Share, Mean (SD) | 62.4 (24.4) | 64.7 (22.6) | 61.7 (27.5) |
| % Medicare Share, Mean (SD) | 16.5 (14.8) | 15.7 (13.3) | 14.2 (12.9) |
| Acuity Index | 11.8 (1.4) | 11.9 (1.4 | 11.3 (2.1) |
Notes: California Office of Statewide Health Planning and Development (OSHPD). Centers for Medicaid and Medicare Services (CMS) Star Rating is a 2010 snapshot of overall rating. HHI (Herfindahl Hirschman Index) is measured from 0 to 1. Occupancy %, % Medicaid, % Medicare are measured from 0–100.
Table 5:
Comparison between Caregiverlist prices and state financial data for California nursing homes
| Caregiverlist Prices for NH that also have OSHPD average private-pay price | OSHPD average private-pay price | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Private-room rates | Semi-private rates | NHs with Caregiverlist private room rates | NHs with Caregiverlist semi-private room rates | NHs with no Caregiverlist prices | ||||||
| N | 383 | 614 | 383 | 614 | 374 | |||||
| Mean ($) | Median ($) | Mean ($) | Median ($) | Mean ($) | Median ($) | Mean ($) | Median ($) | Mean ($) | Median ($) | |
| 246 | 220 | 193 | 185 | 224 | 208 | 220 | 205 | 238 | 207 | |
| Ownership type | ||||||||||
| For-Profit | 241 | 220 | 192 | 185 | 219 | 206 | 217 | 202 | 221 | 203 |
| Non-Profit | 276 | 261 | 202 | 196 | 266 | 250 | 250 | 230 | 308 | 229 |
| Urban | 246 | 220 | 193 | 185 | 225 | 208 | 221 | 206 | 236 | 208 |
| Rural | 230 | 218 | 181 | 185 | 207 | 207 | 198 | 198 | 202 | 203 |
| Any Special Care | 220 | 202 | 181 | 177 | 211 | 200 | 217 | 197 | 213 | 196 |
| Unit | ||||||||||
| No Special Care | 247 | 221 | 194 | 185 | 225 | 208 | 220 | 206 | 238 | 210 |
| Unit | ||||||||||
| Chain | 250 | 225 | 198 | 190 | 231 | 216 | 224 | 211 | 241 | 215 |
| Independent | 241 | 210 | 187 | 180 | 217 | 203 | 216 | 199 | 228 | 200 |
| CMS Overall Rating | ||||||||||
| 1 Star | 226 | 206 | 190 | 181 | 213 | 208 | 210 | 205 | 218 | 202 |
| 2 Star | 262 | 227 | 196 | 185 | 218 | 210 | 216 | 204 | 215 | 200 |
| 3 Star | 245 | 215 | 192 | 185 | 238 | 206 | 231 | 206 | 238 | 216 |
| 4 Star | 247 | 225 | 192 | 185 | 232 | 208 | 225 | 205 | 230 | 205 |
| 5 Star | 242 | 222 | 193 | 185 | 217 | 208 | 217 | 201 | 240 | 224 |
| Total # beds | ||||||||||
| 1–60 | 252 | 225 | 193 | 185 | 235 | 215 | 221 | 206 | 241 | 211 |
| 61–120 | 241 | 220 | 190 | 185 | 225 | 208 | 222 | 205 | 228 | 204 |
| 121–180 | 254 | 220 | 200 | 185 | 209 | 203 | 211 | 203 | 248 | 220 |
| 181+ | 239 | 224 | 191 | 185 | 221 | 203 | 225 | 204 | 222 | 204 |
Notes: California Office of Statewide Health Planning and Development (OSHPD). OSHPD rates were averaged from 2008 to 2010 and calculated using the sum of self-pay revenue divided by the number of self-pay days. Government nursing homes excluded from this table due to small sample size. The sample contains no nursing homes that are hospital-based. Centers for Medicaid and Medicare Services (CMS) Star Rating is a 2010 snapshot of overall rating.
DISCUSSION
This study introduces a new data source for nursing home level private-pay prices that can be easily merged with publicly available data sources to examine national policy and trends in the nursing home industry. As an internal data check, our analysis suggests that non-profit nursing homes charge approximately $7–8 higher rates than their for-profit counterparts and 5-star nursing homes charge about $3–5 more than 1-star rated nursing homes. These results are consistent with economic theory, which suggests that consumers should be willing to pay a premium for facilities able signal high quality services, such as through non-profit status or high star ratings (Arrow 1963; Scanlon 1980).
As an external check, we compare these Caregiverlist data with another source of private-pay information. In our comparison of California nursing homes, we found a moderate correlation in California across Caregiverlist prices and average revenue per day from the cost reports. As another comparison, Huang and Hirth (2016) used state cost reports to construct private prices for California, Florida, New York, Ohio and Texas. The imputed median private prices from their study were lower compared to the Cargiverlist prices from our study but the magnitude varied by state. For example, the median price in New York ($310 versus $305) was relatively similar, while California ($205 versus $220) and Texas ($124 versus $150) had somewhat larger differences (Appendix Table 4). Researchers using average revenues per day calculated using state financial reporting data should be aware that the average revenues per day likely represents a weighted average in between the private and semi-private room prices and the actual price paid by individuals is likely higher or lower depending on patient mix and the amount of extra services that are not included in the room price.
Another potential comparison is with survey data on private nursing home prices from the Genworth Financial annual report on long-term care costs, which surveys 25% of nursing homes annually. When comparing the two sources at the state-level, the annual median nursing home semi-private room prices from the Caregiverlist data were lower for 44 states and higher for 4 states. At the state-level, the Caregiverlist median annual prices for semi-private rooms were on average $2,838 lower than Genworth Financial estimates but there were outliers like Rhode Island ($10,037 lower than Genworth) and Alaska ($42,157 higher than Genworth).
National private prices have been an important “omitted variable” in many nursing home studies (Grabowski 2008). While this study takes an important step towards filling that gap, it is limited in several ways. First, the Caregiverlist data used in this study were collected for roughly 75% of nursing homes nationwide, so this sample may not be perfectly representative of the entire market. Our analysis suggests that generally nursing homes without Caregiverlist price information are government-owned, non-profit and smaller facilities. Second, the Caregiverlist data are almost a decade old. Our analysis is therefore restricted to linkage with other data from that same period, and we are unable to analyze important price trends since then. For example, annual median nursing home private room prices increased from $75,190 in 2010 to $100,375 in 2018 (Genworth). Third, we acknowledge the potential for error with the Caregiverlist data given these data were not originally collected for research purposes. Fourth, the prices reported by Caregiverlist are likely an underestimate of actual out-of-pocket costs for nursing home care due to add-on services not included in the price.
Given these limitations, we encourage future surveys of nursing home private rates that cover all facilities, re-survey at yearly intervals, and precisely standardize and document each call.5 These data would not just have value to researchers. Indeed, the Caregiverlist data were originally collected as a resource for consumers shopping for nursing homes. Yet, the Federal Nursing Home Compare website does not contain any private price information. This could be a valuable source of data for consumers to compare out-of-pocket price of different nursing homes.
This study provided a unique window into private-pay nursing home prices. These data exhibited strong validity in a series of internal and external checks. Both consumers and researchers would benefit from collection of these data on ongoing basis.
Appendix
Appendix Figure 1:
Distribution of Caregiverlist private and semi-private room per day prices
Appendix Table 1.
Sensitivity analysis including Minnesota and North Dakota
| Model | (7) | (8) | (9) | (10) | (11) | (12) | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Outcome | Private Room Price | Private Room Price | Semi-Private Room Price | Semi-Private Room Price | Private-semi private difference | Private-semi private difference | ||||||
| Coef. | SE | Coef. | SE | Coef. | SE | Coef. | SE | Coef. | SE | Coef. | SE | |
| Ownership type | ||||||||||||
| For-Profit | Ref. | Ref. | Ref. | Ref. | Ref. | Ref. | ||||||
| Non-Profit | 6.14*** | (1.65) | 6.85*** | (1.60) | 6.40*** | (1.17) | 7.84*** | (1.13) | −1.02* | (0.58) | −0.63 | (0.67) |
| Government | 4.95 | (3.74) | 6.43 | (4.27) | 4.41* | (2.40) | 7.12** | (3.00) | −2.52** | (1.12) | −2.15 | (1.53) |
| Urban | 13.77*** | (1.55) | 10.69*** | (1.08) | 3.71*** | (0.63) | ||||||
| Hospital-Based | 54.49*** | (7.08) | 59.10*** | (8.48) | 30.33*** | (4.66) | 34.10*** | (5.76) | 0.90 | (1.53) | 1.98 | (2.26) |
| Special Care Unit | 2.54** | (1.28) | 3.22** | (1.32) | 1.41 | (0.93) | 2.55** | (1.00) | 0.44 | (0.65) | 1.13 | (0.72) |
| Chain | 3.67*** | (1.16) | 5.03*** | (1.30) | 3.33*** | (0.75) | 4.37*** | (0.85) | −0.14 | (0.51) | 0.22 | (0.59) |
| County HHI | −17.88*** | (3.02) | 11.18 | (11.73) | −11.37*** | (2.01) | 11.86 | (9.17) | −5.39*** | (1.11) | −0.94 | (3.45) |
| CMS Overall Rating | ||||||||||||
| 1 Star | Ref. | Ref. | Ref. | Ref. | Ref. | Ref. | ||||||
| 2 Star | 2.77* | (1.46) | 1.00 | (1.62) | 1.39 | (0.98) | 0.69 | (1.01) | −0.63 | (0.72) | −0.81 | (0.83) |
| 3 Star | 1.41 | (1.69) | −0.86 | (1.91) | 1.34 | (1.10) | −0.04 | (1.11) | −0.94 | (0.70) | −0.87 | (0.78) |
| 4 Star | 4.08** | (1.59) | 2.50 | (1.64) | 3.52*** | (1.09) | 2.19** | (1.08) | −1.34* | (0.69) | −1.08 | (0.76) |
| 5 Star | 8.49*** | (2.27) | 5.13** | (2.41) | 5.45*** | (1.54) | 3.19** | (1.52) | −0.75 | (0.92) | −0.83 | (1.03) |
| Occupancy % | 0.09* | (0.05) | 0.06 | (0.05) | 0.11*** | (0.03) | 0.09*** | (0.03) | 0.01 | (0.02) | −0.04 | (0.03) |
| Number of Beds | ||||||||||||
| 0–60 | Ref. | Ref. | Ref. | Ref. | Ref. | Ref. | ||||||
| 61–120 | 0.17 | (1.47) | 0.60 | (1.73) | 1.48 | (0.95) | 2.40** | (1.08) | −0.40 | (0.63) | −0.88 | (0.73) |
| 120–180 | 8.81*** | (1.72) | 6.77*** | (1.89) | 8.06*** | (1.20) | 7.13*** | (1.35) | 1.84** | (0.83) | 0.88 | (0.97) |
| 181+ | 21.23*** | (2.40) | 10.51*** | (2.60) | 19.51*** | (1.94) | 10.70*** | (1.72) | 4.60*** | (1.33) | 2.24 | (1.42) |
| % Medicaid Share | −0.33*** | (0.05) | −0.40*** | (0.05) | −0.20*** | (0.03) | −0.24*** | (0.04) | −0.18*** | (0.02) | −0.19*** | (0.02) |
| % Medicare Share | 0.84*** | (0.09) | 0.64*** | (0.09) | 0.57*** | (0.07) | 0.48*** | (0.07) | 0.09*** | (0.03) | 0.06* | (0.03) |
| Acuity Index | 2.63*** | (0.72) | 2.58*** | (0.83) | 2.36*** | (0.55) | 2.64*** | (0.65) | 0.40 | (0.26) | 0.10 | (0.30) |
| Constant | 164.13*** | (9.86) | 181.96*** | (10.87) | 137.58*** | (7.44) | 145.85*** | (8.29) | 27.72*** | (4.35) | 38.66*** | (4.68) |
| State FE | Y | N | Y | N | Y | N | ||||||
| County FE | N | Y | N | Y | N | Y | ||||||
| N | 10402 | 9554 | 11314 | 10454 | 9350 | 8482 | ||||||
| R2 | 0.56 | 0.69 | 0.70 | 0.79 | 0.21 | 0.37 | ||||||
Notes:
p < 0.10,
p < 0.05,
p < 0.01.
Excluding Outliers below 1st and above 99% and facilities from Alaska and Hawaii. Standard errors (SE) in parentheses were clustered at the county-level. HHI (Herfindahl-Hirschman Index) is measured at the county-level and ranges from 0 to 1. In Models 2 and 4, singleton counties are omitted. Occupancy %, % Medicaid, % Medicare measured from 0–100. Centers for Medicaid and Medicare Services (CMS) Star Rating is a 2010 snapshot of overall rating.
Appendix Table 2:
Sensitivity analysis excluding government owned-nursing homes
| Model | (13) | (14) | (15) | (16) | (17) | (18) | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Outcome | Private Room Price | Private Room Price | Semi-Private Room Price | Semi-Private Room Price | Private-semi private difference | Private-semi private difference | ||||||
| Coef. | SE | Coef. | SE | Coef. | SE | Coef. | SE | Coef. | SE | Coef. | SE | |
| Ownership type | ||||||||||||
| For-Profit | Ref. | Ref. | Ref. | Ref. | Ref. | Ref. | ||||||
| Non-Profit | 6.09*** | (1.66) | 7.12*** | (1.62) | 6.26*** | (1.17) | 7.83*** | (1.12) | −1.05* | (0.60) | −0.58 | (0.68) |
| Urban | 14.49*** | (1.56) | 10.69*** | (1.08) | 3.97*** | (0.65) | ||||||
| Hospital-Based | 50.88*** | (8.48) | 53.36*** | (9.91) | 27.60*** | (5.68) | 31.50*** | (6.63) | 1.12 | (1.91) | 1.80 | (2.68) |
| Special Care Unit | 1.90 | (1.30) | 2.87** | (1.35) | 1.29 | (0.89) | 2.41** | (0.95) | 0.31 | (0.68) | 1.10 | (0.74) |
| Chain | 3.85*** | (1.17) | 5.28*** | (1.31) | 3.55*** | (0.76) | 4.54*** | (0.86) | −0.20 | (0.52) | 0.22 | (0.60) |
| County HHI | −16.96*** | (3.10) | 15.60 | (12.47) | −11.98*** | (2.03) | 15.62 | (9.64) | −5.16*** | (1.17) | −1.13 | (3.67) |
| CMS Overall Rating | ||||||||||||
| 1 Star | Ref. | Ref. | Ref. | Ref. | Ref. | Ref. | ||||||
| 2 Star | 2.92** | (1.46) | 1.56 | (1.63) | 1.60 | (0.98) | 0.61 | (1.02) | −0.46 | (0.74) | −0.74 | (0.85) |
| 3 Star | 0.58 | (1.74) | −1.23 | (1.97) | 1.00 | (1.12) | −0.47 | (1.14) | −0.99 | (0.72) | −0.90 | (0.80) |
| 4 Star | 3.62** | (1.62) | 1.93 | (1.68) | 3.19*** | (1.10) | 1.34 | (1.10) | −1.17 | (0.72) | −0.95 | (0.78) |
| 5 Star | 7.08*** | (2.26) | 4.26* | (2.44) | 4.88*** | (1.52) | 2.77* | (1.50) | −0.58 | (0.96) | −0.72 | (1.06) |
| Occupancy % | 0.11** | (0.05) | 0.10* | (0.05) | 0.11*** | (0.03) | 0.09*** | (0.03) | 0.01 | (0.03) | −0.04 | (0.03) |
| Number of Beds | ||||||||||||
| 0–60 | Ref. | Ref. | Ref. | Ref. | Ref. | Ref. | ||||||
| 61–120 | 0.80 | (1.50) | 1.56 | (1.75) | 1.79* | (0.94) | 2.75** | (1.08) | −0.18 | (0.65) | −0.72 | (0.75) |
| 120–180 | 9.13*** | (1.76) | 7.19*** | (1.94) | 8.23*** | (1.19) | 7.34*** | (1.36) | 2.06** | (0.87) | 1.08 | (0.99) |
| 181+ | 23.20*** | (2.44) | 13.03*** | (2.61) | 19.96*** | (1.98) | 11.58*** | (1.74) | 5.23*** | (1.40) | 2.79* | (1.47) |
| % Medicaid Share | −0.39*** | (0.05) | −0.44*** | (0.05) | −0.22*** | (0.03) | −0.26*** | (0.04) | −0.19*** | (0.02) | −0.20*** | (0.02) |
| % Medicare Share | 0.72*** | (0.09) | 0.56*** | (0.09) | 0.54*** | (0.06) | 0.45*** | (0.07) | 0.08*** | (0.03) | 0.06 | (0.03) |
| Acuity Index | 2.60*** | (0.74) | 2.69*** | (0.85) | 2.57*** | (0.56) | 2.84*** | (0.66) | 0.28 | (0.27) | 0.00 | (0.31) |
| Constant | 167.24*** | (10.01) | 181.09*** | (11.10) | 137.32*** | (7.35) | 144.61*** | (8.24) | 29.34*** | (4.52) | 39.95*** | (4.88) |
| State FE | Y | N | Y | N | Y | N | ||||||
| County FE | N | Y | N | Y | N | Y | ||||||
| N | 9,831 | 9,044 | 10,695 | 9,906 | 8,913 | 8,113 | ||||||
| R2 | 0.58 | 0.70 | 0.72 | 0.80 | 0.21 | 0.36 | ||||||
Notes:
p < 0.10,
p < 0.05,
p < 0.01.
Excluding Outliers below 1st and above 99%, facilities from Alaska, Hawaii, Minnesota and North Dakota and government-owned. Standard errors (SE) in parentheses were clustered at the county-level. HHI (Herfindahl-Hirschman Index) is measured at the county-level and ranges from 0 to 1. In Models 2 and 4, singleton counties are omitted. Occupancy %, % Medicaid, % Medicare measured from 0–100. Centers for Medicaid and Medicare Services (CMS) Star Rating is a 2010 snapshot of overall rating.
Appendix Table 3:
Sensitivity analysis excluding hospital-based nursing homes
| Model | (19) | (20) | (21) | (22) | (23) | (24) | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Outcome | Private Room Price | Private Room Price | Semi-Private Room Price | Semi-Private Room Price | Private-semi private difference | Private-semi private difference | ||||||
| Coef. | SE | Coef. | SE | Coef. | SE | Coef. | SE | Coef. | SE | Coef. | SE | |
| Ownership type | ||||||||||||
| For-Profit | Ref. | Ref. | Ref. | Ref. | Ref. | Ref. | ||||||
| Non-Profit | 6.87*** | (1.53) | 7.93*** | (1.46) | 6.93*** | (1.14) | 8.29*** | (1.12) | −1.09* | (0.58) | −0.73 | (0.67) |
| Government | −2.32 | (2.96) | −0.88 | (3.44) | 1.99 | (2.14) | 3.95 | (2.67) | −3.13** | (1.26) | −2.52 | (1.74) |
| Urban | 13.80*** | (1.39) | 10.46*** | (1.00) | 3.84*** | (0.65) | ||||||
| Special Care Unit | 2.35* | (1.26) | 3.31*** | (1.28) | 1.27 | (0.92) | 2.45** | (0.99) | 0.56 | (0.67) | 1.24* | (0.73) |
| Chain | 3.23*** | (1.09) | 4.47*** | (1.25) | 3.38*** | (0.74) | 4.11*** | (0.85) | −0.16 | (0.52) | 0.21 | (0.59) |
| County HHI | −16.93*** | (2.73) | −5.39 | (5.59) | −10.86*** | (1.88) | −2.39 | (4.61) | −5.50*** | (1.14) | −1.14 | (3.03) |
| CMS Overall Rating | ||||||||||||
| 1 Star | Ref. | Ref. | Ref. | Ref. | Ref. | Ref. | ||||||
| 2 Star | 4.16*** | (1.40) | 2.77* | (1.51) | 2.08** | (0.97) | 0.97 | (1.00) | −0.44 | (0.73) | −0.70 | (0.83) |
| 3 Star | 1.04 | (1.64) | −0.18 | (1.83) | 1.35 | (1.09) | 0.25 | (1.10) | −1.00 | (0.71) | −0.99 | (0.79) |
| 4 Star | 3.98*** | (1.54) | 3.66** | (1.55) | 3.45*** | (1.06) | 2.41** | (1.07) | −1.22* | (0.72) | −1.02 | (0.79) |
| 5 Star | 8.30*** | (2.09) | 5.11** | (2.16) | 5.88*** | (1.47) | 3.81*** | (1.46) | −0.52 | (0.96) | −0.85 | (1.06) |
| Occupancy % Number of Beds | 0.15*** | (0.04) | 0.09* | (0.05) | 0.15*** | (0.02) | 0.13*** | (0.03) | 0.01 | (0.02) | −0.04 | (0.03) |
| 0–60 | Ref. | Ref. | Ref. | Ref. | Ref. | Ref. | ||||||
| 61–120 | 2.97** | (1.28) | 3.60** | (1.53) | 3.66*** | (0.87) | 4.45*** | (1.02) | −0.05 | (0.65) | −0.62 | (0.76) |
| 120–180 | 11.83*** | (1.55) | 9.99*** | (1.71) | 10.33*** | (1.13) | 9.40*** | (1.28) | 2.31*** | (0.85) | 1.25 | (0.99) |
| 181+ | 24.28*** | (2.23) | 13.76*** | (2.44) | 21.40*** | (1.79) | 12.85*** | (1.60) | 5.06*** | (1.31) | 2.60* | (1.38) |
| % Medicaid Share | −0.38*** | (0.05) | −0.42*** | (0.05) | −0.21*** | (0.03) | −0.25*** | (0.04) | −0.18*** | (0.02) | −0.20*** | (0.02) |
| % Medicare Share | 0.62*** | (0.07) | 0.51*** | (0.08) | 0.48*** | (0.05) | 0.39*** | (0.06) | 0.08*** | (0.03) | 0.05 | (0.04) |
| Acuity Index | 2.53*** | (0.71) | 2.44*** | (0.85) | 2.36*** | (0.44) | 2.67*** | (0.53) | 0.35 | (0.26) | 0.06 | (0.29) |
| Constant | 162.89*** | (9.65) | 183.93*** | (10.72) | 134.40*** | (6.40) | 144.73*** | (7.08) | 28.35*** | (4.51) | 39.45*** | (4.90) |
| State FE | Y | N | Y | N | Y | N | ||||||
| County FE | N | Y | N | Y | N | Y | ||||||
| N | 9,905 | 9,121 | 10,838 | 10,052 | 8,951 | 8,158 | ||||||
| R2 | 0.61 | 0.72 | 0.74 | 0.81 | 0.21 | 0.37 | ||||||
Notes:
p < 0.10,
p < 0.05,
p < 0.01.
Excluding Outliers below 1st and above 99%, facilities from Alaska, Hawaii, Minnesota and North Dakota and hospital-base. Standard errors (SE) in parentheses were clustered at the county-level. HHI (Herfindahl-Hirschman Index) is measured at the county-level and ranges from 0 to 1. In Models 2 and 4, singleton counties are omitted. Occupancy %, % Medicaid, % Medicare measured from 0–100. Centers for Medicaid and Medicare Services (CMS) Star Rating is a 2010 snapshot of overall rating.
Appendix Table 4:
Caregiverlist private and semi-private room prices per day at the state-level
| Private Room Price | Semi-Private Room Price | Private Room Price | Semi-Private Room Price | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| State | N | Mean | Median | N | Mean | Median | State | N | Mean | Median | N | Mean | Median |
| AK | 1 | 1250 | 1250 | 5 | 765 | 714 | MT | 76 | 217 | 174 | 81 | 202 | 163 |
| AL | 156 | 173 | 160 | 154 | 152 | 150 | NC | 306 | 197 | 180 | 324 | 173 | 165 |
| AR | 156 | 163 | 145 | 173 | 140 | 130 | ND | 67 | 334 | 217 | 65 | 294 | 199 |
| AZ | 109 | 240 | 210 | 119 | 193 | 171 | NE | 150 | 185 | 162 | 158 | 193 | 150 |
| CA | 446 | 258 | 220 | 701 | 207 | 185 | NH | 35 | 296 | 260 | 49 | 271 | 249 |
| CO | 155 | 224 | 209 | 166 | 183 | 184 | NJ | 259 | 311 | 295 | 259 | 281 | 270 |
| CT | 206 | 350 | 355 | 220 | 321 | 325 | NM | 35 | 200 | 177 | 43 | 173 | 164 |
| DC | 13 | 389 | 300 | 12 | 319 | 250 | NV | 31 | 223 | 208 | 42 | 203 | 185 |
| DE | 21 | 230 | 235 | 23 | 212 | 217 | NY | 528 | 323 | 305 | 529 | 312 | 291 |
| FL | 598 | 235 | 211 | 641 | 207 | 194 | OH | 733 | 210 | 195 | 792 | 188 | 175 |
| GA | 307 | 168 | 159 | 315 | 152 | 146 | OK | 230 | 157 | 137 | 244 | 129 | 125 |
| HI | 12 | 544 | 383 | 13 | 370 | 320 | OR | 96 | 229 | 217 | 105 | 198 | 195 |
| IA | 322 | 163 | 144 | 319 | 146 | 132 | PA | 354 | 265 | 238 | 410 | 240 | 223 |
| ID | 53 | 276 | 199 | 61 | 226 | 180 | RI | 69 | 260 | 260 | 76 | 237 | 237 |
| IL | 652 | 254 | 167 | 705 | 229 | 143 | SC | 138 | 181 | 170 | 137 | 168 | 158 |
| IN | 437 | 202 | 187 | 467 | 164 | 155 | SD | 85 | 179 | 164 | 82 | 157 | 153 |
| KS | 249 | 182 | 146 | 241 | 162 | 135 | TN | 196 | 186 | 172 | 239 | 162 | 159 |
| KY | 63 | 213 | 182 | 61 | 179 | 169 | TX | 779 | 166 | 150 | 926 | 118 | 114 |
| LA | 146 | 150 | 132 | 165 | 133 | 125 | UT | 56 | 194 | 175 | 60 | 150 | 148 |
| MA | 375 | 315 | 309 | 386 | 295 | 289 | VA | 195 | 200 | 185 | 210 | 183 | 170 |
| MD | 146 | 305 | 226 | 166 | 270 | 211 | VT | 36 | 246 | 245 | 35 | 225 | 225 |
| ME | 77 | 301 | 260 | 89 | 258 | 230 | WA | 179 | 240 | 228 | 195 | 211 | 206 |
| MI | 338 | 228 | 210 | 363 | 205 | 195 | WI | 306 | 223 | 209 | 304 | 202 | 189 |
| MN | 96 | 319 | 210 | 93 | 282 | 191 | WV | 102 | 233 | 206 | 104 | 218 | 198 |
| MO | 435 | 145 | 134 | 442 | 129 | 124 | WY | 30 | 200 | 175 | 31 | 181 | 165 |
| MS | 141 | 191 | 180 | 147 | 185 | 175 | |||||||
Notes: Data collected from 2008–2010.
Footnotes
Examples include private rooms, televisions, personal clothing, social events beyond the activity program, and special care services not included in the facility’s Medicaid payment.
Caregiverlist contracted the initial survey to India-based callers with experience calling the United States. After three failed call attempts, Caregiverlists’ Chicago-based team would follow up with the facility to request the pricing information. These follow-up callers did not always pose as potential customers when requesting pricing information. Caregiverlist did not maintain records of which facilities required these additional follow-ups. Caregiverlist estimates that approximately 20% of facilities were called by the Chicago-based team and notes that these facilities reluctant to providing pricing information typically serve predominantly Medicaid-eligible patients.
Rate equalization laws in both states prohibit nursing homes from setting prices for a semi-private room above what Medicaid pays, however, they can charge above the Medicaid rate for private rooms
Excluding outliers of price and average revenues per day at the 99 percentile the Pearson correlation between average revenues per day and private room price is 0.33 and semi-private room is 0.42. The Spearman correlation between average revenues per day and private room prices (including outliers) was 0.49 and 0.68 for semi-private room prices.
For example, implementing a standardized patient vignette with specific diagnoses could help improve the comparability of price quotes across facilities if quoted private rates vary with a patient’s care requirements. Also, documenting specific call dates may help researchers control for seasonal price variation or variation due to temporarily high or low censuses.
References
- Arling G, and Daneman B. 2002. “Nursing Home Case-Mix Reimbursement in Mississippi and South Dakota. Health Services Research 37 (2): 377–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Arrow KJ (1978). Uncertainty and the welfare economics of medical care. In Uncertainty in Economics (pp. 345–375). Academic Press. [Google Scholar]
- Brown JR, & Finkelstein A (2011). Insuring long-term care in the United States. Journal of Economic Perspectives, 25(4), 119–42. [Google Scholar]
- Bowblis JR (2014). Nursing home prices and market structure: the effect of assisted living industry expansion. Health Economics, Policy and Law, 9(1), 95–112. [DOI] [PubMed] [Google Scholar]
- Clement JP, Bazzoli GJ, & Zhao M (2012). Nursing home price and quality responses to publicly reported quality information. Health Services Research, 47(1pt1), 86–105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Clement JP, & Khushalani J (2015). Are Publicly Reported Quality Measures Aligned with Nursing Home Prices?. Journal of Health Care Finance, 42(1). [Google Scholar]
- Gandhi Ashvin D. (2019) “Picking Your Patients: Selective Admissions in the Nursing Home Industry.” Working Paper. [Google Scholar]
- Genworth Financial (2010). Cost of Care Survey 2010. Available from: https://www.genworth.com/dam/Americas/US/PDFs/Consumer/corporate/coc_10.pdf
- Genworth Financial (2018). Cost of Care Survey 2018. Available from: https://pro.genworth.com/riiproweb/productinfo/pdf/282102.pdf
- Grabowski DC (2008). The market for long-term care services. INQUIRY: The Journal of Health Care Organization, Provision, and Financing, 45(1), 58–74. [DOI] [PubMed] [Google Scholar]
- Grabowski DC (2004). A longitudinal study of Medicaid payment, private-pay price and nursing home quality. International journal of health care finance and economics, 4(1), 5–26. [DOI] [PubMed] [Google Scholar]
- Huang SS, & Hirth RA (2016). Quality rating and private-prices: Evidence from the nursing home industry. Journal of health economics, 50, 59–70. [DOI] [PubMed] [Google Scholar]
- Huang SS, Banaszak-Holl J, Yuan S, & Hirth RA (2019). The Determinants and Variation of Nursing Home Private-Pay Prices: Organizational and Market Structure. Medical Care Research and Review, 1077558719857335. [DOI] [PubMed] [Google Scholar]
- Hurd MD, Michaud PC, & Rohwedder S (2017). Distribution of lifetime nursing home use and of out-of-pocket spending. Proceedings of the National Academy of Sciences, 114(37), 9838–9842. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kaiser Family Foundation (2017). Medicaid’s Role in Nursing Home Care. Available from: https://www.kff.org/infographic/medicaids-role-in-nursing-home-care
- Loomer L (2018) “Ownership, Price Mark-ups and Demand Elasticities in the Nursing Home Market” Working Paper. [Google Scholar]
- Nyman JA (1985). Prospective and ‘cost-plus’ Medicaid reimbursement, excess Medicaid demand, and the quality of nursing home care. Journal of Health Economics, 4(3), 237–259. [DOI] [PubMed] [Google Scholar]
- Nyman JA (1988a). Excess demand, the percentage of Medicaid patients, and the quality of nursing home care. Journal of Human Resources, 76–92. [Google Scholar]
- Nyman JA (1988b). The effect of competition on nursing home expenditures under prospective reimbursement. Health Services Research, 23(4), 555. [PMC free article] [PubMed] [Google Scholar]
- Nyman JA (1989a). The private demand for nursing home care. Journal of Health Economics, 8(2), 209–231. [DOI] [PubMed] [Google Scholar]
- Nyman JA (1989b). Analysis of nursing home use and bed supply: Wisconsin, 1983. Health Services Research, 24(4), 511. [PMC free article] [PubMed] [Google Scholar]
- Nyman JA (1994). The effects of market concentration and excess demand on the price of nursing home care. The Journal of Industrial Economics, 193–204. [Google Scholar]
- Mukamel DB, & Spector WD (2002). The competitive nature of the nursing home industry: price mark ups and demand elasticities. Applied Economics, 34(4), 413–420. [Google Scholar]
- Scanlon WJ (1980). A theory of the nursing home market. Inquiry, 17(1), 25–41. [PubMed] [Google Scholar]