Abstract
Objectives:
This study examined urban/rural differences in the frequency of preventable Emergency Department (ED) visits among patients with Alzheimer’s Disease and Related Dementias (ADRD), with a focus on the variation of Accountable Care Organization (ACO) participation status for hospitals in urban and rural areas.
Design:
We performed a cross-sectional study using the 2015 State Emergency Department Databases, the American Hospital Association Annual Survey of Hospitals, and the Area Health Resource File. Individual-, county-, and hospital-level characteristics and state fixed effects were used for model specification.
Setting:
Patients with ADRD from seven states who visited the ED and had routine discharges.
Participants:
Our sample consists of 117,196 patients with ADRD.
Measurements:
The outcome was preventable ED visits which was classified using the New York University (NYU) Emergency Department visit algorithm. We performed a multivariable logistic regression to estimate the variation of preventable ED visits by urban and rural areas.
Results:
Rural patients with ADRD had 1.13 higher adjusted odds (P = 0.007) of going to the ED for a preventable visit compared to their urban counterparts. In addition, ACO-affiliated hospitals had 0.91 lower adjusted odds (P = 0.005) of preventable ED visits for ADRD patients compared to hospitals not affiliated with an ACO. Whole County Mental Health Care Health Professional Shortage Area (HPSA) (OR = 1.14, P = 0.002) designation was also an indicator of higher preventable ED rates.
Conclusions:
ACO delivery systems have the potential to decrease rural preventable ED visits among ADRD patients.
Keywords: Alzheimer’s Disease and Related Dementias, preventable hospitalizations, emergency department, access to care, rural health
Health outcomes and health care costs for patients with Alzheimer’s Disease and Related dementias (ADRD) are of increasing concern.1 Compared to patients without dementia, patients with dementia have higher rates of emergency department (ED) visits and revisits.2,3 ED dementia care research has been prioritized by the American Geriatrics Society since 2006,4 and the American College of Emergency Physicians developed guidelines for the geriatric ED accreditation program to improve geriatric emergency health care delivery.5
A multidisciplinary team approach is critical to improving the health of patients with dementia.6 The Accountable Care Organization (ACO) framework was established to improve care coordination among doctors, hospitals, and other sources of healthcare with the goals of increasing quality of care while mitigating costs.7 ACOs have focused their cost-reduction strategies on reducing ED visits by engaging in efforts like primary care redesign to centralize patient health management through primary care providers and creating ED alternatives.8 By putting greater emphasis of care coordination on primary care and geriatricians, ACOs are suggested to provide greater patient care coordination which leads to better care management for ADRD patients.9 However, this is a challenge since the number of geriatricians has declined amidst an expanding population of older adults.10
Compared to urban residents, rural ADRD patients have significantly higher preventable ED visit rates due to a lack of cognitive impairment specialists and county mental health resources.11 Rural areas are home to greater numbers of residents 65 years and older and persons with multiple chronic conditions and are also vulnerable to shortages of health professionals, specialists, and social services.12 To combat these resource deficits, rural ACOs have made strides in care management, cost-saving incentives, and population health management for rural Medicare beneficiaries.13 However, hospitals currently utilizing the ACO model tend to be large, urban, and have more funding and data for care coordination programs.14 The ACO infrastructure can increase administrative/bureaucratic burden and be expensive for smaller hospitals,15 while there are observed differences in ACO participation based on practice differences.16 Additionally, ACOs may potentially exacerbate racial disparities due to differences in quality performance.17
Our objective is to examine associations between hospital ACO affiliation and preventable ED visits for urban and rural ADRD patients, hypothesizing that patients with ADRD were less likely to encounter a preventable ED visit if treated in an ACO-affiliated hospital. Furthermore, given that there are fewer ACO-affiliated hospitals in rural areas, we hypothesize that increasing rural hospital ACO participation can potentially reduce preventable ED rates and reduce urban/rural disparities. Our hypotheses are developed based on the ACO model’s mission to improve health care quality and reduce health care costs.18 ACOs accomplish this by providing financial incentives to providers for promoting care coordination across care settings and care transitions among providers, patients, and caregivers.19 All of our analyses control for the availability of health care resources in the community.20
METHODS
Our primary dataset was the 2015 State Emergency Department Databases (SEDD) which captured all the discharge information from the participating states’ EDs. ED discharges from seven states (Arizona, Florida, Kentucky, Maryland, North Carolina, Vermont, and Wisconsin) were selected due to the availability of SEDD files and necessary variables, such as patients’ race/ethnicity and geographic linkage to the hospitals’ county. To capture county geographic variations and hospital characteristics, we further linked the SEDD data with the 2015 Area Health Resources File (AHRF) and the 2015 American Hospital Association (AHA) Annual Survey of Hospitals.
We used ICD-9 codes in Quarters 1–3 and ICD-10 codes in Quarter 4 to identify the ADRD diagnosis (see the full list in Supplementary Tables S1 and S2). ICD-9 codes for any ADRD diagnosis code (primary, secondary, etc.) were identified from the Alzheimer’s Association and previously published studies.11,21 ICD-10 codes used for any ADRD diagnosis were cited by the Alzheimer’s Association’s Cognitive Impairment Care Planning Toolkit.22
We employed the New York University (NYU) ED visit algorithm to assess preventable ED visits based on the diagnoses. This algorithm classified each ED visit with a probability of requiring emergent ED care but was potentially preventable if timely ambulatory care was provided. The algorithm has been independently validated and was patched in 2017 to improve the precision of the visit classification.23,24 We identified ED visits as preventable if the probability of being a preventable ED visit was at least 50% based on an existing measure from the literature.25
Our key independent variable is the latest available urban/rural county status of the hospital, as determined by the 2013 Rural-Urban Continuum Codes (RUCC) obtained from AHRF. Among the nine categories in the RUCC, we designated the first three as urban and the last six as rural. The second key explanatory variable is the ACO affiliation of the hospital, captured using the 2015 AHA Annual Survey of Care Systems and Payment. The survey question asks if the “hospital has established a separate legal entity for an ACO, is part of an ACO, or is actively working to establish an ACO in the future” with yes or no as the response.
Our chosen covariates were informed by the ADRD and rural health literature, as well as the adapted Andersen Healthcare Utilization Model which included geographic predictors.26 We controlled for race/ethnicity, age, gender, Elixhauser Comorbidities, and primary insurance/payer. We were not able to measure person-level wealth directly in our data given the data limitations. Instead, we used the patient’s ZIP code-level income as a proxy for patient’s income-level. A time index comparing the first three quarters of 2015 and the last quarter of 2015 was created to account for the ICD-9 to ICD-10 switch in diagnostic codes.
Geographic characteristics included the median household income percentiles based on the patient’s zip code and county percent of African Americans. We also assessed the availability of health care resources by county Health Professional Shortage Area (HPSA) and Mental Health HPSA (MHPSA) status which shows the ratio of the population to primary care physicians and mental health professionals, respectively. We used these indicators as proxies for whether residents in the county lack access to timely and affordable primary and/or mental health care. We created indicators of Whole County HPSA and MHPSA if the whole county was designated as a shortage area. Finally, we included the following hospital characteristics: ACO affiliation, number of hospital beds, and hospital ownership status.
Our final sample size was 117,169 visits that had an ADRD diagnosis, home discharge, and complete data; over 90% of patients were 65 years or older. We excluded non-home discharges to create a more homogenous sample since non-home discharge suggests increased disease severity.
Study Design
First, we compiled the characteristics of the ADRD patient population by urban/rural county status and hospital ACO affiliation to compare population differences. Second, we performed a logistic regression of preventable ED visits using the covariates listed above while including state fixed effects and reported adjusted odds ratios.
We performed a sensitivity analysis using different cutoffs for preventable ED visits and primary care-related ED utilization.11 We also tested different definitions of urban and rural areas based on the RUCC.11 We used the Oaxaca decomposition to identify and quantify the characteristics that contribute to the observed urban/rural differences in preventable ED visit rates (results shown in Supplementary Table S3). All analyses were performed using Stata version 15 following approval by the University of Maryland IRB.
RESULTS
Table 1 presents the demographics of our study sample. Most rural hospitals had less than 200 beds and had patients from zip codes representing the bottom quartile of income. Urban ACO hospitals had a preventable ED visit rate of 4.9%, compared to 5.4% for urban non-ACO hospitals, 6.3% for rural ACO hospitals, and 6.8% for rural non-ACO hospitals (Figure 1). Over half of all rural ADRD patients went to a hospital in an MHPSA county, compared to less than 25% of urban ADRD patients. About 12% of rural patients of non-ACO hospitals were in an HPSA county while less than 2% of urban patients and rural ACO-hospital patients lived in an HPSA county.
Table 1: Sample Characteristics of ADRD Patients.
ADRD Sample Characteristics by Urban/Rural Classification and Hospital ACO Affiliation
Urban | Rural | |||
---|---|---|---|---|
Variable | Non-ACO (%) | ACO (%) | Non-ACO (%) | ACO (%) |
Preventable ED Visit Status | ||||
Not Preventable | 45,172 (94.6) | 49,420 (95.1) | 12,299 (93.2) | 4,021 (93.7) |
Preventable | 2,572 (5.4) | 2,544 (4.9) | 898 (6.8) | 270 (6.3) |
Race | ||||
Non-Hispanic White | 37,055 (77.6) | 38,561 (74.2) | 10,718 (81.2) | 3,895 (90.8) |
Non-Hispanic Black | 7,662 (16.0) | 7,541 (14.5) | 1,990 (15.1) | 218 (5.1) |
Hispanic | 2,365 (5.0) | 4,821 (9.3) | 180 (1.4) | 130 (3.0) |
Non-Hispanic Asian/Pacific Islander | 224 (0.5) | 405 (0.8) | 20 (0.2) | 7 (0.2) |
Non-Hispanic Native American | 121 (0.3) | 200 (0.4) | 236 (1.8) | 35 (0.8) |
Non-Hispanic Other | 317 (0.7) | 446 (0.8) | 53 (0.4) | 6 (0.1) |
Gender | ||||
Male | 16,873 (35.3) | 18,291 (35.2) | 4,551 (34.5) | 1,598 (37.2) |
Female | 30,871 (64.7) | 33,673 (64.8) | 8,646 (65.5) | 2693 (62.8) |
Age group | ||||
30–49 years old | 392 (0.8) | 456 (0.9) | 93 (0.7) | 38 (0.9) |
50–64 years old | 2,495 (5.2) | 2,603 (5.0) | 657 (5.0) | 233 (5.4) |
65–74 years old | 6,802 (14.3) | 7,303 (14.0) | 2,149 (16.3) | 663 (15.5) |
75+ years old | 38,055 (79.7) | 41,602 (80.1) | 10,298 (78.0) | 3,357 (78.2) |
Income Quartile by Zip code | ||||
0–25th percentile | 15,565 (32.6) | 14,457 (27.8) | 8,050 (61.0) | 2,203 (51.3) |
26th-50th percentile | 12,451 (26.1) | 14,433 (27.8) | 3,746 (28.4) | 1,351 (31.5) |
51st-75th percentile | 11,140 (23.3) | 12,262 (23.6) | 1,183 (9.0) | 631 (14.7) |
76th-100th percentile | 8,588 (18.0) | 10,812 (20.8) | 218 (1.7) | 106 (2.5) |
Primary Payer | ||||
Medicare | 40,244 (84.3) | 45,363 (87.3) | 11,275 (85.4) | 3,779 (88.1) |
Medicaid | 1,052 (2.2) | 1,151 (2.2) | 233 (1.8) | 70 (1.6) |
Other | 6,448 (13.5) | 5,450 (10.5) | 1,689 (12.8) | 442 (10.3) |
Elixhauser Comorbidities | ||||
≤2 | 33,663 (70.5) | 35,876 (69.0) | 9,458 (71.7) | 2,930 (68.3) |
>2 | 14,081 (29.5) | 16,088 (31.0) | 3,739 (28.3) | 1,361 (31.7) |
Time Index (Quarter of Year) | ||||
Q1–Q3 | 35,889 (75.2) | 39,213 (75.5) | 10,100 (76.5) | 3,232 (75.3) |
Q4 | 11,855 (24.8) | 12,751 (24.5) | 3,097 (23.5) | 1,059 (24.7) |
County Average Percent African American | ||||
Below Median (<13.92%) | 22,017 (46.1) | 24,840 (47.8) | 8,464 (64.1) | 3,840 (89.5) |
Above Median (≥13.92%) | 25,727 (53.9) | 27,124 (52.2) | 4,733 (35.9) | 451 (10.5) |
County Health Professional Shortage Area | ||||
Not/Part of County HPSA | 47,725 (99.9) | 51,598 (99.3) | 11,553 (87.5) | 4,212 (98.2) |
Whole County HPSA | 19 (0.1) | 366 (0.7) | 1,644 (12.5) | 79 (1.8) |
County Mental Health HPSA | ||||
Not/Part of County Mental Health HPSA | 39,934 (83.6) | 39,796 (76.6) | 6,492 (49.2) | 1,579 (36.8) |
Whole County Mental Health HPSA | 7,810 (16.4) | 12,168 (23.4) | 6,705 (50.8) | 2,712 (63.2) |
Hospital Number of Beds | ||||
<200 | 15,603 (32.7) | 8,845 (17.0) | 9,522 (72.2) | 3,610 (84.1) |
200–299 | 9,403 (19.7) | 7,442 (14.3) | 1,889 (14.3) | 188 (4.4) |
300–499 | 13,614 (28.5) | 14,976 (28.8) | 1,786 (13.5) | 339 (7.9) |
⩾500 | 9,124 (19.1) | 20,701 (39.9) | 0 (0.0) | 154 (3.6) |
Hospital Ownership | ||||
For-profit | 6,071 (12.7) | 2,039 (3.9) | 1,883 (14.3) | 193 (4.5) |
Non-for-profit | 33,271 (69.7) | 44,333 (85.3) | 8,658 (65.6) | 3,987 (92.9) |
Government | 8,402 (17.6) | 5,592 (10.8) | 2,656 (20.1) | 111 (2.6) |
States | ||||
Arizona | 2,153 (4.5) | 9,379 (18.1) | 252 (1.9) | 151 (3.5) |
Florida | 18,260 (38.3) | 14,141 (27.2) | 516 (3.9) | 246 (5.7) |
Kentucky | 4,244 (8.9) | 2,992 (5.8) | 3,245 (24.6) | 826 (19.3) |
Maryland | 7,744 (16.2) | 7,193 (13.8) | 679 (5.2) | 0 (0.0) |
North Carolina | 13,859 (29.0) | 12,213 (23.5) | 7,431 (56.3) | 1,299 (30.3) |
Vermont | 0 (0.0) | 356 (0.7) | 123 (0.9) | 162 (3.8) |
Wisconsin | 1,484 (3.1) | 5,690 (10.9) | 951 (7.2) | 1,607 (37.4) |
Overall | 47,744 | 51,964 | 13,197 | 4,291 |
Notes. Data Sources: 2015 State Emergency Department Databases, AHA Annual Survey of Hospitals, and Area Health Resources File. Preventable ED visits are defined as ED care was required but the nature of the condition was preventable if timely and effective ambulatory care had been received during the episode of illness. Numbers may not add up to 100 percent due to rounding. N = 117,169, the final sample consists of >90% of patients older than or equal to 65 years.
Figure 1:
Preventable ED Visit % by Urban/Rural and ACO Status
Table 2 shows the multivariable logistic regression results. Rural patients with ADRD had 1.13 higher adjusted odds (95% confidence interval (CI) = 1.04–1.24) of going to the ED for a preventable visit compared to urban counterparts. Additionally, ACO-affiliated hospitals had 0.91 lower adjusted odds (CI = 0.86–0.96) of preventable ED visits for ADRD patients compared to hospitals with no ACO affiliation. Non-Hispanic Black (OR = 1.25, CI = 1.16–1.34) and Hispanic patients (OR = 1.16, CI = 1.04–1.30) had higher rates of preventable ED visits compared to Non-Hispanic White patients. Patients with more than two Elixhauser comorbidities had significantly greater odds of preventable ED visits (OR = 3.04, CI = 2.89–3.20). Patients whose visit was in the fourth quarter of 2015 had 1.07 higher odds of a preventable ED visit (CI = 1.01–1.14) compared with those in the first three quarters of 2015. Hospitals with greater than 300 beds had significantly reduced odds of preventable ED visits compared to hospitals with less than 200 beds (300–499 beds had OR = 0.90, ≥500 beds had OR = 0.81). Whole county MHPSA (OR = 1.14, CI = 1.05–1.24) was also an indicator of higher preventable ED rates. Compared to the reference state of Florida, Kentucky (OR = 1.30, CI = 1.18–1.44) and Vermont (OR = 1.77, CI = 1.25–2.49) had higher odds of preventable ED visits.
Table 2: Logistic Regression Results of Preventable Emergency Department Visits.
Logistic Model of Preventable Emergency Department Visits for ADRD Patients
Variable | Odds Ratio | 95% CI | P value |
---|---|---|---|
Hospital ACO Affiliation | |||
Non-ACO | Ref | ||
ACO | 0.91 | (0.86–0.96) | .007 |
Race | |||
Non-Hispanic White | Ref | ||
Non-Hispanic Black | 1.25 | (1.16–1.34) | <.001 |
Hispanic | 1.16 | (1.04–1.30) | .008 |
Non-Hispanic Asian/Pacific Islander | 1.16 | (0.83–1.63) | .383 |
Non-Hispanic Native American | 1.20 | (0.87–1.65) | .272 |
Non-Hispanic Other | 0.99 | (0.71–1.38) | .942 |
Gender | |||
Male | Ref | ||
Female | 0.92 | (0.87–0.97) | .002 |
Age group | |||
30–49 years old | Ref | ||
50–64 years old | 1.03 | (0.78–1.37) | .828 |
65–74 years old | 0.90 | (0.68–1.20) | .482 |
75+ years old | 0.82 | (0.62–1.07) | .145 |
Income Quartile by Zip code | |||
0–25th percentile | Ref | ||
26th-50th percentile | 0.96 | (0.90–1.03) | .230 |
51st-75th percentile | 0.97 | (0.90–1.05) | .504 |
76th-100th percentile | 0.97 | (0.89–1.06) | .521 |
Primary Payer | |||
Medicare | Ref | ||
Medicaid | 0.96 | (0.80–1.15) | .660 |
Other | 1.02 | (0.94–1.11) | .679 |
Elixhauser Comorbidities | |||
≤2 | Ref | ||
>2 | 3.04 | (2.89–3.20) | <.001 |
Time Index (Quarter of Year) | |||
Q1–Q3 | Ref | ||
Q4 | 1.07 | (1.01–1.14) | .020 |
Urban/Rural Status | |||
Urban | Ref | ||
Rural | 1.13 | (1.04–1.24) | .005 |
County Average Percent African American | |||
Below Median (<13.92) | Ref | ||
Above Median (≥13.92) | 1.02 | (0.96–1.09) | .487 |
County Health Professional Shortage Area | |||
Not/Part of County HPSA | Ref | ||
Whole County HPSA | 1.03 | (0.86–1.23) | .756 |
County Mental Health Professional Short Area | |||
Not/Part of County Mental Health HPSA | Ref | ||
Whole County Mental Health HPSA | 1.14 | (1.05–1.24) | .002 |
Hospital Number of Beds | |||
<200 | Ref | ||
200–299 | 0.99 | (0.91–1.07) | .763 |
300–499 | 0.90 | (0.83–0.97) | .004 |
≥500 | 0.81 | (0.75–0.88) | <.001 |
Ownership | |||
For profit | Ref | ||
Non-for-profit | 1.01 | (0.91–1.12) | .845 |
Government | 1.09 | (0.97–1.23) | .165 |
States | |||
Florida | Ref | ||
Arizona | 1.08 | (0.95–1.23) | .218 |
Kentucky | 1.30 | (1.18–1.44) | <.001 |
Maryland | 1.09 | (0.98–1.22) | .116 |
North Carolina | 0.97 | (0.90–1.06) | .545 |
Vermont | 1.77 | (1.25–2.49) | .001 |
Wisconsin | 1.07 | (0.95–1.21) | .236 |
Notes. Data Sources: 2015 State Emergency Department Databases, AHA Annual Survey of Hospitals, and Area Health Resources File. We defined an ED visit as preventable if it has ≥50% probability of being preventable, we tested different cutoffs of 40% and 60% and saw that our results are robust. P value is based on χ2 test. N = 117,169
DISCUSSION
Our study found significant differences in health care provider availability and state variations in rural and urban regions, which helped explain higher preventable ED visits among ADRD patients in rural counties compared to urban ones. This finding is consistent with literature about the positive association between ambulatory care continuity and decreased preventable hospitalizations for Medicare beneficiaries.27
Our study shows that factors contributing most to the urban/rural disparities were hospital and county characteristics such as hospital bed size, the prevalence of mental health professionals in the county, and hospital ACO affiliation (the decomposition explained 45.8% of the urban-rural differences, see Supplementary Table S3). Mental health is important for ADRD since psychiatric disturbances affect as many as 90% of patients with ADRD and also many ADRD caregivers.28 An MHPSA implies that patients in this county may have problems with obtaining timely psychiatric care or any mental health care at all.
Our study had several limitations. First, it followed a cross-sectional design and cannot be used to make causal inferences. Second, we used hospital claims data which might contain recording errors. For example, we were not able to capture under-diagnosis of ADRD using the claims data. Future studies that improve early ADRD diagnosis and timely access to prevention and patient education will be critical. Key factors identified in the ADRD ED literature such as measures of quality of care, patient satisfaction, geriatric medical specialties, and training and accreditation were not included due to the data limitation.29 Using claims-based analysis, we were not able to observe patients’ frailty and disease progression. Third, the measure of hospital ACO affiliation indicated the hospitals’ intention and enrollment in the ACO but lacked additional details of ACO features and could not distinguish between hospital- and physician-based ACOs or between currently established ACOs and working to establish an ACO in the near future. Fourth, our dataset consisted of seven states and may not be representative of nationwide trends. Finally, mental health conditions were not classified as preventable in the NYU algorithm.
Conclusion
Older adults with ADRD living in rural areas face numerous challenges compared to their urban counterparts. This study adds to the literature in demonstrating that hospitals affiliated with ACOs were associated with lower odds of preventable ED visits compared to hospitals that were not ACO-affiliated. Other factors that explained the rural-urban difference included hospital bed size and county-level availability of mental health professionals. These results suggest that the ACO model’s incentives for care coordination may be a promising mechanism by which to eliminate urban-rural disparities in ADRD care.
Supplementary Material
Supplementary Table S1: ICD-9 Diagnostic Codes for ADRD
Supplementary Table S2: ICD-10 Diagnostic Codes for ADRD
Supplementary Table S3: Oaxaca Decomposition Results
ACKNOWLEDGMENTS
Sponsor’s Role
There are no sponsors for this paper.
Funding:
This research was supported by National Institutes of Health, National Institute on Minority Health and Health Disparities Grant R01MD011523 and 3R01MD011523-03S1; National Institutes of Health, National Institute on Aging Grant R56AG062315 awarded to Jie Chen.
Footnotes
Disclosures: The authors have no conflicts of interest to disclose.
The abstract was submitted to ASHEcon 2020 and AcademyHealth Annual Research Meeting 2020.
REFERENCES
- 1.2020 Alzheimer’s disease facts and figures. Alzheimer’s & Dementia. 2020;16(3):391–460. [DOI] [PubMed] [Google Scholar]
- 2.LaMantia MA, Stump TE, Messina FC, Miller DK, Callahan CM. Emergency Department Use Among Older Adults with Dementia. Alzheimer Dis Assoc Disord. 2016;30(1):35–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kent T, Lesser A, Israni J, Hwang U, Carpenter C, Ko KJ. 30-Day Emergency Department Revisit Rates among Older Adults with Documented Dementia. Journal of the American Geriatrics Society. 2019;67(11):2254–2259. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Carpenter CR, Gerson LW. Geriatric Emergency Medicine.; 2006:28. http://www.newfrontiers.americangeriatrics.org/chapters/pdf/rasp_16.pdf.
- 5.Carpenter C, Hwang U, Biese K, et al. ACEP Accredits Geriatric Emergency Care for Emergency Departments. ACEP Now. https://www.acepnow.com/article/acep-accredits-geriatric-emergency-care-emergency-departments/. Published April 10, 2017. Accessed August 12, 2020.
- 6.Carpenter CR, McFarland F, Avidan M, et al. Impact of Cognitive Impairment Across Specialties: Summary of a Report From the U13 Conference Series. J Am Geriatr Soc. 2019;67(10):2011–2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Berwick DM. Making Good on ACOs’ Promise — The Final Rule for the Medicare Shared Savings Program. New England Journal of Medicine. 2011;365(19):1753–1756. [DOI] [PubMed] [Google Scholar]
- 8.Lin MP, Muhlestein D, Carr BG, Richardson LD, Wiler JL, Schuur JD. Engagement of Accountable Care Organizations in Acute Care Redesign: Results of a National Survey. J Gen Intern Med. 2018;33(10):1601–1603. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Anderson AC, Chen J. ACO Affiliated Hospitals Increase Implementation of Care Coordination Strategies. Med Care. 2019;57(4):300–304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Lo AX, Carpenter CR. Balancing Evidence and Economics While Adapting Emergency Medicine to the 21st Century’s Geriatric Demographic Imperative. Acad Emerg Med. April 2020. [DOI] [PubMed] [Google Scholar]
- 11.Wang N, Albaroudi A, Chen J. Decomposing Urban and Rural Disparities of Preventable ED Visits Among Patients With Alzheimer’s Disease and Related Dementias: Evidence of the Availability of Health Care Resources. The Journal of Rural Health. July 2020. https://onlinelibrary.wiley.com/doi/abs/10.1111/jrh.12465. Accessed August 13, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Alfero C, Coburn AF, Lundblad JP, et al. Care Coordination in Rural Communities: Supporting the High Performance Rural Health System.; 2015:29. https://www.ruralhealthresearch.org/publications/972.
- 13.Baloh J, MacKinney C, Mueller K, Vaughn T, Ullrich F. Developmental Strategies and Challenges for Rural Accountable Care Organizations. https://www.ruralhealthresearch.org/publications/960. Published February 2015. Accessed March 26, 2020. [PubMed]
- 14.Colla CH, Lewis VA, Tierney E, Muhlestein DB. Hospitals Participating In ACOs Tend To Be Large And Urban, Allowing Access To Capital And Data. Health Aff (Millwood). 2016;35(3):431–439. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Berenson RA, Burton RA, McGrath M. Do accountable care organizations (ACOs) help or hinder primary care physicians’ ability to deliver high-quality care? Healthc (Amst). 2016;4(3):155–159. [DOI] [PubMed] [Google Scholar]
- 16.Resnick MJ, Graves AJ, Buntin MB, Richards MR, Penson DF. Surgeon Participation in Early Accountable Care Organizations. Ann Surg. 2018;267(3):401–407. [DOI] [PubMed] [Google Scholar]
- 17.Lewis VA, Fraze T, Fisher ES, Shortell SM, Colla CH. ACOs Serving High Proportions Of Racial And Ethnic Minorities Lag In Quality Performance. Health Affairs. 2017;36(1):57–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.McWilliams JM, Hatfield LA, Landon BE, Hamed P, Chernew ME. Medicare Spending after 3 Years of the Medicare Shared Savings Program. New England Journal of Medicine. 2018;379(12):1139–1149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Tan ZS, Jennings L, Reuben D. Coordinated Care Management For Dementia In A Large, Academic Health System. Health Aff (Millwood). 2014;33(4):619–625. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Yeager VA, Menachemi N, Savage GT, Ginter PM, Sen BP, Beitsch LM. Using resource dependency theory to measure the environment in health care organizational studies: A systematic review of the literature. Health Care Management Review. 2014;39(1):50–65. [DOI] [PubMed] [Google Scholar]
- 21.Alzheimer’s Association. 2018 Alzheimer’s disease facts and figures. Alzheimer’s & Dementia. 2018;14(3):367–429. [Google Scholar]
- 22.Cognitive Impairment Care Planning Toolkit. Alzheimer’s Association; 2018:47. https://www.alz.org/careplanning/downloads/care-planning-toolkit.pdf. [Google Scholar]
- 23.Ballard DW, Price M, Fung V, et al. Validation of an algorithm for categorizing the severity of hospital emergency department visits. Med Care. 2010;48(1):58–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Johnston KJ, Allen L, Melanson TA, Pitts SR. A “Patch” to the NYU Emergency Department Visit Algorithm. Health Serv Res. 2017;52(4):1264–1276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Guthrie K, Casselman B, Grice J, McManus D, Larkin T. Using SAS® to Analyze Emergency Department Visits: Medicaid Patients Compared to Other Pay Sources.; 2017:16. https://support.sas.com/resources/papers/proceedings17/1248-2017.pdf.
- 26.Andersen RM. Revisiting the Behavioral Model and Access to Medical Care: Does it Matter? Journal of Health and Social Behavior. 1995;36(1):1–10. [PubMed] [Google Scholar]
- 27.Nyweide DJ, Anthony DL, Bynum JPW, et al. Continuity of Care and the Risk of Preventable Hospitalization in Older Adults. JAMA Intern Med. 2013;173(20). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3877937/. Accessed March 23, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Lyketsos CG, Olin J. Depression in Alzheimer’s disease: overview and treatment. Biol Psychiatry. 2002;52(3):243–252. [DOI] [PubMed] [Google Scholar]
- 29.Hogan TM, Losman ED, Carpenter CR, et al. Development of Geriatric Competencies for Emergency Medicine Residents Using an Expert Consensus Process. Academic Emergency Medicine. 2010;17(3):316–324. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Table S1: ICD-9 Diagnostic Codes for ADRD
Supplementary Table S2: ICD-10 Diagnostic Codes for ADRD
Supplementary Table S3: Oaxaca Decomposition Results