Abstract
Objectives:
Sustaining systolic blood pressure (SBP) control reduces the risk for cardiovascular events that impair function, but its association with nursing home admission has not been well studied.
Methods:
We conducted an analysis of sustained SBP control and long-term nursing home admissions using data from the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) linked to Medicare claims restricted to participants with fee-for-service coverage, ≥8 study visits with SBP measurements, who were not living in a nursing home during a 48-month baseline BP assessment period (n=6,557). Sustained SBP control was defined as <140 mm Hg at <50%, 50% to <75%, 75% to <100%, and 100% of visits. Nursing home admissions were identified using the Medicare Long Term Care Minimum Data Set.
Results:
The mean age of participants was 73.8 years and 44.3% were men. Over a median follow-up of 9.2 years, 844 participants (12.8%) had a nursing home admission. Rates of nursing home admission per 100 person-years were 16.3 for participants with SBP control at <50%, 14.1 at 50% to <75%, 7.8 at 75% to <100%, and 5.3 at 100% of visits. Compared to those with sustained SBP control at <50% of visits, hazard ratios (95% confidence intervals) for nursing home admission were 0.79 (0.66-0.93), 0.70 (0.58-0.84), and 0.57 (0.44-0.74) among participants with SBP control at 50% to <75%, 75% to <100%, and 100% of visits, respectively.
Conclusions:
Among Medicare beneficiaries in ALLHAT, sustained SBP control was associated with a lower risk of long-term nursing home admission.
INTRODUCTION
Many older adults prefer to remain living in their communities and avoid being admitted to a nursing home.1,2 In addition to the loss of independence associated with long-term nursing home admission, institutional care can lead to high out-of-pocket costs, depletion of life savings, and the need for Medicaid enrollment to cover nursing home expenses.3,4 Furthermore, nursing home outbreaks of coronavirus 2019 infection (COVID-19) have exposed additional risks related to congregated living and the challenges limiting the spread of infections.5,6 Despite these reasons for reducing nursing home admissions, population-level strategies to reduce the need for institutional care are limited.
Improving population health by treating and controlling hypertension may be one approach to reduce nursing home admissions. Among older adults, antihypertensive medication has been shown to lower the risk for cardiovascular disease (CVD) events including stroke, heart failure (HF), and coronary heart disease (CHD) which are known to be associated with functional decline.7,8 Sustaining systolic blood pressure (SBP) control over time is also associated with a lower risk of multimorbidity, a key factor contributing to frailty.9,10 For example, achieving SBP control at a greater percentage of visits over a 48-month time period was shown to be associated with a slower rate of accumulation of chronic conditions in the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT), delaying the onset of six or more concurrent chronic conditions by 8 years.9
As older adults often prioritize independent living, data on the association of sustained SBP control with reduced need for nursing home admission could support patient-centered care to achieve greater SBP control. The purpose of this analysis was to determine the association between sustained SBP control and long-term admission to a nursing home using data from a large hypertension clinical trial linked to Centers for Medicare and Medicaid (CMS) health insurance claims.
METHODS
Study design and population
We conducted a retrospective cohort study using data from ALLHAT linked to CMS health insurance claims. The details of this ALLHAT-CMS linkage have been previously reported.9,10 Briefly, ALLHAT is a large simple trial designed to compare the occurrence of major CVD events among participants randomized to receive antihypertensive medications from four different drug classes.11,12 Adults aged 55 years or older who had hypertension and at least one additional CHD risk factor were recruited from 625 clinical sites between 1994 and 1998. The analytic cohort for the current proposal was selected from 6,591 ALLHAT participants who met the following four criteria: 1) were randomized to receive amlodipine, lisinopril, or chlorthalidone, 2) had SBP measurements at eight or more of the twelve ALLHAT visits during a BP assessment period, 3) did not have a nonfatal myocardial infarction (MI), stroke, or HF event prior to the end of the BP assessment period, and 4) had Medicare fee-for-service Part A and B coverage for the six months prior to the end of the BP assessment period.9 We required participants to have both Medicare Part A (i.e., inpatient, skilled nursing facility [SNF], home health and hospice care coverage) and Medicare Part B (i.e., outpatient coverage) so that we could identify comorbid conditions through the claims data. Participants with Medicare Advantage were not included because claims data for individuals with this coverage are not complete. We excluded 34 participants who had Medicare data indicating they were receiving nursing home care during the BP assessment period, resulting in a final sample of 6,557 participants for the current analysis. This analysis of ALLHAT data was approved by the Duke University Institutional Review Board.
Blood pressure assessment
BP was measured by trained staff at each ALLHAT study visit following a standardized protocol. BP was measured with a standard mercury sphygmomanometer while participants were seated quietly, with their feet flat on the floor, after a 5-minute rest period. BP levels were calculated as the average of two measurements obtained with a 30-second interval separation.11
ALLHAT study visits occurred at regularly scheduled intervals, every three months for twelve months after randomization, and then every four months thereafter through March 2002.12 For the current analysis, we defined the BP assessment period as the twelve visits conducted between 6 and 48 months following randomization (i.e., 6, 9, 12, 16, 20, 24, 28, 32, 36, 40, 44, and 48 months). BP measurements obtained at randomization and 1 and 3 months after randomization were not included in the BP assessment period because a high proportion of participants were having their antihypertensive medication titrated during this period. We required participants to have at least eight visits during the BP assessment period in order to obtain a reliable estimate of SBP control.
Sustained SBP control was defined based on the percentage of study visits at which participants had SBP < 140 mm Hg during the BP assessment period. We defined sustained control based on a SBP < 140 mm Hg because this was the ALLHAT systolic treatment goal. We did not consider diastolic BP control because isolated systolic hypertension as this is the most common form of hypertension among older adults and most commonly targeted to reduce CVD outcomes in this population.13,14 The percentage of visits with sustained SBP control was categorized into four groups: 1) at less than 50%, 2) at 50% to less than 75%, 3) at 75% to less than 100%, and 4) at 100% of visits. These categories were chosen based on prior studies of sustained SBP control and to provide a metric that could that easily calculated and communicated with patients.7
Long-term nursing home admission
Follow-up for first nursing home admission began at the end of the BP assessment period. Nursing home admissions were identified using data from the CMS Long Term Care Minimum Data Set (MDS) 2.0. The MDS includes results from a standardized clinical assessment tool administered to nursing home patients that is used to monitor quality of care.15 MDS assessments are conducted routinely including on nursing home admission and discharge. Completion of MDS assessments and reporting to CMS is required for facilities to become Medicare or Medicaid certified and receive CMS payments for care. As 95% of all US nursing homes are Medicare or Medicaid certified, MDS data identify almost all of US adults receiving nursing home care.16,17
In the US, nursing homes often provide both long-term care (i.e., residential nursing home care) and post-acute care (i.e., short stay rehabilitation). The goal of post-acute care is to facilitate patient recovery so that they can be discharged home.18 However, it is common for patients admitted to a nursing home for post-acute care to switch to long-term care if they do not recover functional independence or if they expend their Medicare post-acute care benefit.19 It is also common for those receiving nursing home care to be intermittently admitted to an acute care hospital.20 This break in nursing home care requires an MDS assessment be completed on discharge from the nursing home and a new MDS completed when the patient returns. Based on these considerations, we calculated the cumulative days for which ALLHAT participants were in a nursing home using dates corresponding to MDS assessments conducted on admission and discharge. Nursing home admission was then defined as > 90 cumulative days without a break in nursing home care of more than 30 days. The average length of a post-acute care stay in a nursing home is approximately 25 days and a 90-day threshold was chosen to avoid misclassification of poste-acute care as long-term nursing home care.21 Additionally, participants who died before meeting the 90-day threshold were considered to be receiving nursing home care if MDS data indicated that they were in a nursing home when they died or if they died within 7 days following discharge form a nursing home. The nursing home admission date was defined as the first MDS admission date for participants who met the above criteria.
Other covariates
Data on covariates were obtained as part of ALLHAT conducted before randomization or obtained through Medicare claims. Age was calculated at the end of the BP assessment period (i.e., beginning of follow-up for this analysis). Information on race/ethnicity, sex, education level, body mass index (BMI), and smoking status was obtained prior to randomization in ALLHAT. Diabetes, hyperlipidemia, dementia, HF, and stroke were defined based on ICD-9 codes indicating a service or treatment for these conditions in Medicare claims before the end of the BP assessment period using recommended algorithms for defining and measuring chronic conditions when using Medicare claims.9,22
Statistical analysis
Characteristics were summarized for participants overall and in each of the four categories of sustained SBP control (<50%, 50% to <75%, 75% to <100%, and 100% of visits) using mean and standard deviation (SD) for continuous variables and count (percent) for categorical variables. We calculated the cumulative incidence for nursing home admission by category of sustained SBP control accounting for the competing risk of death. Cox proportional hazards models were used to obtain hazard ratios (HRs) for the association of sustained SBP control with the first nursing home admission. Follow-up time began on the date corresponding to the end of the BP assessment period. Time-to-event outcomes were censored upon death, loss of Medicare Part A or B coverage, or the end of Medicare follow-up (December 31, 2010). We used age as the time scale for the proportional hazards models to reduce potential bias in risk estimates related to the higher incidence of nursing home admission at older age. Comparing participants at similar age accounts for age as a potential confounder.23,24 Participants who had SBP control at < 50% of visits served as the referent group. We conducted three progressively adjusted regression models: Model 1 was unadjusted, Model 2 was adjusted for sex, race, education, smoking status, and BMI, and Model 3 was adjusted for sex, race, education, smoking status, BMI, and history of diabetes, hyperlipidemia, stroke, HF, dementia, and CHD. As rates of nursing home admission have been shown to differ by gender and race,25 we repeated analysis stratified by gender (men and women) and race (Black and non-Black). We formally tested for effect modification using multiplicative interaction terms for sustained SBP control category and gender (SBP category*gender) and, separately, race (SBP category*race). All analyses were performed with SAS Studio 9.4.
RESULTS
Participant characteristics
Participants had a mean (SD) age of 73.8 (6.6) years, 27.5% were non-Hispanic Black, and 44.3% were men (Table 1). Compared to participants with SBP < 140 mm Hg at 100% of visits, those with SBP control at <50% of visits were older, and more likely to be non-Hispanic Black and less likely to be men. The prevalence of diabetes and stroke was lower among those without SBP control at a greater percentage of visits. Hyperlipidemia was more common among those with SBP control at a greater percentage of visits.
Table 1.
Characteristics of ALLHAT participants linked to CMS health insurance claims included in the current post-hoc analysis, overall and by percentage of visits with systolic blood pressure (SBP) < 140 mm Hg.
| Percentage of visits with SBP < 140 mm Hg | ||||||
|---|---|---|---|---|---|---|
| Overall (n=6557) |
< 50% (n=1990) |
≥50% to <75% (n=1915) |
≥75% to <100% (n=1788) |
100% (n=864) |
p-value*** | |
| Mean age (SD), years* | 73.8 (6.6) | 74.3 (6.6) | 74.1 (6.8) | 73.3 (6.6) | 72.8 (6.6) | <0.0001 |
| Race, n (%) | <0.0001 (Black vs non-Black) | |||||
| Non-Hispanic white Non-Hispanic black Hispanic white Hispanic black Other |
3278 (50.0) 1804 (27.5) 970 (14.8) 166 (2.5) 339 (5.2) |
960 (48.2) 702 (35.3) 154 (7.7) 39 (2.0) 135 (6.8) |
969 (50.6) 524 (27.4) 301 (15.7) 45 (2.4) 76 (4.0) |
918 (51.3) 410 (22.9) 308 (17.2) 61 (3.4) 91 (5.1) |
431 (49.9) 168 (19.4) 207 (24.0) 21 (2.4) 37 (4.3) |
|
| Men, n (%) | 2902 (44.3) | 797 (40.1) | 775 (40.5) | 887 (49.6) | 443 (51.3) | <0.0001 |
| Mean education level (SD), years | 10.8 (4.1) | 10.5 (4.4) | 11.0 (4.0) | 10.9 (4.0) | 10.9 (4.3) | 0.0007 |
| Mean BMI (SD), kg/m2 | 29.2 (5.6) | 29.5 (5.8) | 29.2 (5.6) | 29.0 (5.3) | 28.9 (5.3) | 0.012 |
| Current smoker,** n (%) | 1162 (17.7) | 330 (16.6) | 361 (18.9) | 314 (17.6) | 157 (18.2) | 0.31 |
| Chronic conditions | ||||||
| Diabetes mellitus, n (%) | 2352 (35.9) | 778 (39.1) | 661 (34.5) | 602 (33.7) | 311 (36.0) | 0.003 |
| Hyperlipidemia, n (%) | 2423 (37.0) | 663 (33.2) | 700 (36.6) | 675 (37.8) | 385 (44.6) | <0.0001 |
| Coronary heart disease, n (%) | 1480 (22.6) | 429 (21.6) | 432 (22.6) | 410 (22.9) | 209 (24.2) | 0.45 |
| Dementia, n (%) | 102 (1.6) | 24 (1.2) | 35 (1.8) | 29 (1.6) | 14 (1.6) | 0.46 |
| Heart failure, n (%) | 412 (6.3) | 146 (7.3) | 112 (5.9) | 108 (6.0) | 46 (5.3) | 0.12 |
| Stroke, n (% | 366 (5.6) | 132 (6.6) | 108 (5.6) | 92 (5.2) | 34 (3.9) | 0.026 |
CMS = Centers for Medicare and Medicaid Services, SD = standard deviation, SBP = Systolic blood pressure, BMI = body mass index.
Age at the end of the BP assessment period
Smoking status obtained at time of ALLHAT randomization
ANOVA F-test for continuous variables and chi-square test for categorical variables
Long-term nursing home admission
Over a median follow-up of 9.2 years, 844 (12.8%) of participants had a long-term nursing home admission. The cumulative incidence of nursing home admission was lower among participants with controlled SBP at a higher percentage of visits (Figure 1). Rates of nursing home admission per 100 person-years (95% CI) were 16.3 (14.3, 18.4), 14.1 (12.0, 16.0), 7.8 (5.5, 9.9), and 5.3 (4.0, 7.7) for participants with SBP control at <50%, 50% to <75%, 75% to <100%, and 100% of visits, respectively (Table 2). Compared to those with SBP control at <50% of visits, multivariable adjusted HRs (95% confidence intervals [CI]) of nursing home placement were 0.79 (0.66, 0.93), 0.70 (0.58, 0.84), and 0.57 (0.44, 0.74) among participants with SBP control at 50% to <75%, 75% to <100%, and 100% of visits, respectively.
Figure 1.
Cumulative incidence of nursing home admission by percentage of visits with systolic blood pressure (SBP) < 140 mm Hg.
Table 2.
Incidence rates and hazard ratios (95% CI) for nursing home admission by percentage of visits with SBP < 140 mm Hg.
| Percentage of visits with SBP < 140 mm Hg | ||||
|---|---|---|---|---|
| < 50% (n=1990) |
≥50% to < 75% (n=1915) |
≥75% to <100% (n=1788) |
100% (n=864) |
|
| Nursing home admission | ||||
| N (%) | 332 (16.7) | 244 (12.7) | 195 (10.9) | 73 (8.5) |
| Incidence rate, per 100 PY (95% CI) | 16.3 (14.3, 18.4) | 14.1 (12.0, 16.0) | 7.8 (5.5, 9.9) | 5.3 (4.0, 7.7) |
| HR (95% CI) | ||||
| Model 1 | 1 (ref) | 0.77 (0.65, 0.92) | 0.67 (0.56, 0.81) | 0.53 (0.41, 0.68) |
| Model 2 | 1 (ref) | 0.78 (0.66, 0.92) | 0.69 (0.57, 0.83) | 0.54 (0.42, 0.70) |
| Model 3 | 1 (ref) | 0.79 (0.66, 0.93) | 0.70 (0.58, 0.84) | 0.57 (0.44, 0.74) |
SBP = systolic blood pressure, PY = person-years, HR = hazard ratio, CI = confidence interval
Model 1: Unadjusted
Model 2: Adjusted for sex, race, education, smoking status, body mass index (BMI)
Model 3: Adjusted for sex, race, education, smoking status, BMI, and history of diabetes, hyperlipidemia, stroke, heart failure, dementia, and coronary heart disease
Time scale = left-censored age
Subgroup analysis
Long-term nursing home admission occurred in 498 (13.6%) women versus 346 (11.9%) men and 273 (13.8%) of Black participants versus 571 (12.5%) of non-Black participants. The association between sustained SBP control and nursing home admission in gender and race subgroups are presented in Figure 2 and Supplemental Tables 1 and 2. There was no evidence of effect modification between sustained SBP control and nursing home admissions by gender or race (p-interaction = 0.63 for gender and 0.39 for race).
Figure 2. Multivariable adjusted hazard ratios (95% CI) for nursing home admission by percentage of visits with systolic blood pressure control among (A) gender and (B) race subgroups.
Participants who had SBP control at < 50% of visits served as the referent group within each subgroup. There was no evidence of effect modification between sustained SBP control and nursing home admissions by gender or race (p-interaction = 0.63 for gender and 0.39 for race). Time scale = left-censored age. Multivariable adjustment included gender (for race subgroups only), race (for gender subgroups only), education, smoking status, body mass index, and history of diabetes, hyperlipidemia, stroke, heart failure, dementia, and coronary heart disease. Time scale = left-censored age. N=6,557.
DISCUSSION
In the current study of ALLHAT participants with Medicare health insurance, there was a graded association of lower risk of long-term nursing home admission with a higher percentage of visits with sustained SBP control. A consistent pattern of lower risk with a higher percentage of visits with SBP control was present for women and men and Black and non-Black ALLHAT participants. These findings suggest that sustaining SBP control may be a strategy to prevent nursing home admission. Describing the potential health benefits of sustained SBP control in terms of maintaining independence and remaining in one’s home could be a patient-centered approach to goal setting for older adults with hypertension.
Prior studies have shown sustained SBP control to be associated with lower risk for CVD events as well as age-related adverse health outcomes. For example, Mancia and colleagues reported lower risk for CVD events including MI, HF, and stroke associated with BP control at a greater percentage of visits in two separate analyses, from the INVEST (International Verapamil SR-Trandolapril) trial which enrolled participants with hypertension in addition to coronary artery disease and the VALUE (Valsartan Antihypertensive Long-term use Evaluation) trial which included those with hypertension and three or more CVD risk factors.26,27 A previous analysis using data from the ALLHAT-CMS linkage identified an inverse association between sustained SBP control and progression of multimorbidity defined as the rate of accumulation of chronic conditions over time.9,10 In that analysis, sustained SBP control at 100% of visits versus <50% of visits corresponded to estimated delays in the development of six or more chronic conditions from 68 years to 76 years of age. The current study extends these findings of lower risk with sustained SBP control to include nursing home admission.
Findings from the current study have potential implications for reducing health care costs and providing patient-centered care. Nursing home care is expensive and as most older adults do not have long-term care insurance, much of this care is paid for out-of-pocket or through Medicaid.3,4 Prior studies have shown that the average private-pay prices for a semi-private nursing home room in the US to be more than $70,000 per year, although there is considerable variability in prices by location, ownership type, and CMS rating.28 As Medicare does not cover long-term care, Medicaid is the largest payer in the US with an annual long term care spending of over $100 billion.29 In addition to the high cost of nursing home care, more than 90% of older adults prefer to live somewhere other than a nursing home, and more than half of seriously ill older adults report that they are unwilling to live permanently in a nursing home.1,2 Aging in place may be a common health goal for older adults, but discussions about BP management may often focus on CVD events, rather than the possibility of nursing home admission. Findings from the current analysis may support shared decision-making about setting and maintaining SBP treatment goals among older adults who report remaining in the community as one of their prioritized health goals.
Strengths of the current analysis include the large sample size of ALLHAT participants taking antihypertensive medication, the availability of multiple standardized measurements to estimate sustained SBP, and the linkage to CMS MDS data for identification of nursing home admission over an average of nearly 10 years of follow-up. There are, however, limitations that must be considered. Although data were from a large randomized trial, this was a non-randomized analysis. Limitations of this approach include potential unmeasured confounders for the association between sustained SBP control and nursing home admission. Misclassification of nursing home admission was also possible. Participants who died with <90 days in a nursing home were considered to have a nursing home admission but may have been discharged from post-acute care if they had survived.
Among ALLHAT participants with Medicare health insurance, sustained SBP control was associated with lower rates of long-term nursing home admission. Older adults often frame their health goals around maintaining independence, not in terms of risk factor management. Evidence from the current study may facilitate patient-centered discussions about BP control as a strategy to prevent nursing home admission.
Supplementary Material
Funding Source
C.B.B. had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Support was provided through the National Heart, Lung, and Blood Institute (R01HL133618) and the National Institute on Aging (R01AG062502) to C.B.B. This work was also supported by the Durham Center of Innovation to Accelerate Discovery and Practice Transformation (ADAPT), (CIN 13-410) at the Durham VA Health Care System. The views expressed here/in this manuscript are those of the authors and do not necessarily represent the views of the National Heart, Lung, and Blood Institute; the National Institutes of Health; or the Department of Health and Human Services. C.B.B. had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Footnotes
Conflicts of interest
The authors report to conflicts of interest.
References
- 1.Kasper JD, Wolff JL, Skehan M. Care Arrangements of Older Adults: What They Prefer, What They Have, and Implications for Quality of Life. Gerontologist. 2019;59(5):845–855. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Mattimore TJ, Wenger NS, Desbiens NA, et al. Surrogate and physician understanding of patients' preferences for living permanently in a nursing home. J Am Geriatr Soc. 1997;45(7):818–824. [DOI] [PubMed] [Google Scholar]
- 3.National Institute on Aging. Paying for Care. https://www.nia.nih.gov/health/paying-care. US Department of Health and Human Services. Published 2017. Accessed February 23, 2021. [Google Scholar]
- 4.Mommaerts C Are coresidence and nursing homes substitutes? Evidence from Medicaid spend-down provisions. J Health Econ. 2018;59:125–138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Abrams HR, Loomer L, Gandhi A, Grabowski DC. Characteristics of U.S. Nursing Homes with COVID-19 Cases. J Am Geriatr Soc. 2020;68(8):1653–1656. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Brown KA, Jones A, Daneman N, et al. Association Between Nursing Home Crowding and COVID-19 Infection and Mortality in Ontario, Canada. JAMA Intern Med. 2021;181(2):229–236. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Bowling CB, Davis BR, Luciano A, et al. Sustained blood pressure control and coronary heart disease, stroke, heart failure, and mortality: An observational analysis of ALLHAT. J Clin Hypertens (Greenwich). 2019;21(4):451–459. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ferrucci L, Guralnik JM, Pahor M, Corti MC, Havlik RJ. Hospital diagnoses, Medicare charges, and nursing home admissions in the year when older persons become severely disabled. JAMA. 1997;277(9):728–734. [PubMed] [Google Scholar]
- 9.Bowling CB, Sloane R, Pieper C, et al. Association of Sustained Blood Pressure Control with Multimorbidity Progression Among Older Adults. J Am Geriatr Soc. 2020;68(9):2059–2066. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bowling CB, Sloane R, Pieper C, et al. Association of sustained blood pressure control with lower risk for high cost multimorbidities among Medicare beneficiaries in ALLHAT Journal of General Internal Medicine. 2021;In Press. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Davis BR, Cutler JA, Gordon DJ, et al. Rationale and design for the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). ALLHAT Research Group. American journal of hypertension. 1996;9(4 Pt 1):342–360. [DOI] [PubMed] [Google Scholar]
- 12.Officers ALLHAT and Coordinators for the ALLHAT Collaborative Research Group. Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA. 2002;288(23):2981–2997. [DOI] [PubMed] [Google Scholar]
- 13.Franklin SS, Jacobs MJ, Wong ND, L'Italien GJ, Lapuerta P. Predominance of isolated systolic hypertension among middle-aged and elderly US hypertensives: analysis based on National Health and Nutrition Examination Survey (NHANES) III. Hypertension. 2001;37(3):869–874. [DOI] [PubMed] [Google Scholar]
- 14.SPRINT Research Group Wright JT Jr., Williamson JD, et al. A Randomized Trial of Intensive versus Standard Blood-Pressure Control. N Engl J Med. 2015;373(22):2103–2116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Centers for Medicare and Medicaid Servies. MDS 2.0 Public Quality Indicator and Resident Reports. https://www.cms.gov/Research-Statistics-Data-and-Systems/Computer-Data-and-Systems/MDSPubQIandResRep. Updated 12/06/2013. Accessed February 24, 2021.
- 16.Goodwin JS, Li S, Middleton A, Ottenbacher K, Kuo YF. Differences Between Skilled Nursing Facilities in Risk of Subsequent Long-Term Care Placement. J Am Geriatr Soc. 2018;66(10):1880–1886. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Koroukian SM, Xu F, Murray P. Ability of Medicare claims data to identify nursing home patients: a validation study. Med Care. 2008;46(11):1184–1187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kane RL. Finding the right level of posthospital care: "We didn't realize there was any other option for him". JAMA. 2011;305(3):284–293. [DOI] [PubMed] [Google Scholar]
- 19.Goodwin JS, Howrey B, Zhang DD, Kuo YF. Risk of continued institutionalization after hospitalization in older adults. J Gerontol A Biol Sci Med Sci. 2011;66(12):1321–1327. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Burke RE, Whitfield EA, Hittle D, et al. Hospital Readmission From Post-Acute Care Facilities: Risk Factors, Timing, and Outcomes. J Am Med Dir Assoc. 2016;17(3):249–255. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.MedPAC. A Data Book: Health Care Spending and the Meicare Program, July 2020. Washington, D.C. 2020. http://www.medpac.gov/-documents-/data-book. Published 2020. Accessed February 25, 2021. [Google Scholar]
- 22.Centers for Medicare and Medicaid Servies. Chronic Conditions Data Warehouse. https://www2.ccwdata.org/web/guest/condition-categories. Published 2020. Accessed February 10, 2021.
- 23.Korn EL, Graubard BI, Midthune D. Time-to-event analysis of longitudinal follow-up of a survey: choice of the time-scale. Am J Epidemiol. 1997;145(1):72–80. [DOI] [PubMed] [Google Scholar]
- 24.Thiebaut AC, Benichou J. Choice of time-scale in Cox's model analysis of epidemiologic cohort data: a simulation study. Stat Med. 2004;23(24):3803–3820. [DOI] [PubMed] [Google Scholar]
- 25.Thomeer MB, Mudrazija S, Angel JL. How do race and Hispanic ethnicity affect nursing home admission? Evidence from the Health and Retirement Study. J Gerontol B Psychol Sci Soc Sci. 2015;70(4):628–638. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Mancia G, Kjeldsen SE, Zappe DH, et al. Cardiovascular outcomes at different on-treatment blood pressures in the hypertensive patients of the VALUE trial. Eur Heart J. 2016;37(12):955–964. [DOI] [PubMed] [Google Scholar]
- 27.Mancia G, Messerli F, Bakris G, Zhou Q, Champion A, Pepine CJ. Blood pressure control and improved cardiovascular outcomes in the International Verapamil SR-Trandolapril Study. Hypertension. 2007;50(2):299–305. [DOI] [PubMed] [Google Scholar]
- 28.Loomer L, Gandhi A, Geng F, Grabowski DC. Secret Shopper Data on Private Prices in the Nursing Home Industry From 2008 to 2010. Med Care Res Rev. 2019:1077558719879022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Kaiser Family Foundation. Distribution of Medicaid Spending by Service. State Health Facts Web site. www.kff.org/medicaid/state-indicator/distribution-of-medicaid-spending-by-service/. Published 2019. Accessed February 25, 2021. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.