Disparities in Successful Discharge to the Community Following Use of Medicare Home Health by Level of Neighborhood Socioeconomic Disadvantage

Abstract

Considering the importance of social and structural support and resources in recovering health, where people reside could lead to differences in health outcome in Medicare home health care. We used the 2019 Outcome and Assessment Information Set and Area Deprivation Index to examine the association between neighborhood context and successful discharge to community among older Medicare home health care users. Based on the multivariable logistic regression (OR: 0.84; 95% CI, 0.83–0.85) and conditional logistic regression models stratified by home health agency (OR: 0.95; 95% CI, 0.94–0.95), patients living in the most disadvantaged neighborhoods were less likely to experience successful discharge to community than others. Furthermore, the predicted probability of successful discharge to community decreased as the percentage of patients from the most disadvantaged neighborhoods within a home health agency increased. Policymakers should consider using area-level interventions and supports to reduce disparities in Medicare home health care.

Keywords

health disparities home care quality of care socioeconomic status

What this paper adds

• Our study results suggest that Medicare home health care users in disadvantaged neighborhoods face more challenges in being successfully discharged to their home and community than others despite using the same home health agency.

• This study shows home health agencies with a higher percentage of patients from the most disadvantaged neighborhoods were associated with a lower probability of successful discharge to community.

Applications of study findings

• Our study results suggest area-level interventions and programs be considered to target and support patients in disadvantaged neighborhoods and home health agencies mainly serving such patients to improve and reduce neighborhood-based gaps in Medicare home health care and outcomes.

• We need further research to identify potential external factors outside the home health agencies’ control contributing to such differences and clarify the effect of such factors on the association between neighborhood context and successful discharge to community to produce helpful information on designing plans to mitigate differences in home health care and outcomes by neighborhood disadvantage.

Introduction

Medicare home health provides needed care services, such as intermittent skilled nursing care, skilled therapy, and home health aide services, to homebound Medicare beneficiaries (MedPAC, 2021). Home health care services allow older adults (ages 65 or older) to stay at their home and community, fulfilling individual preferences and reducing Medicare spending (Dick et al., 2019; Geng et al., 2020; Gregory et al., 2010; Werner et al., 2019). Based on these benefits, the use of home health care has grown substantially, of which the number of Medicare beneficiaries using home health care increased by more than 30% from 2002 to 2019 (MedPAC, 2021). Several reports have also shown that home health care is one of the most rapidly increasing post-acute care settings used by Medicare patients after discharge from hospitals (Cuckler et al., 2018; Mcwilliams et al., 2017; MedPAC, 2021). Despite the shift toward Medicare home health care and increasing preference for home-based care, it is not clear whether the quality of care and patient outcomes vary across Medicare home health care users. Overlooking certain subgroups with poor health outcomes and low-quality care could limit the benefits of home health care and lead to unnecessary health care spending. Thus, identifying whether home health outcomes vary across patients and what factors relate to this variation is a crucial step in developing policies that improve the overall health of home health care users and reduce unnecessary spending on Medicare home health care.

Health outcomes of Medicare home health care patients could vary by neighborhood context. According to Mohnen et al. (2019), neighborhood can affect healthcare utilization and health through three mechanisms: supply-side, need, and demand-side factors of healthcare (Mohnen et al., 2019). In this study, we are interested in the “supply-side” pathway that focuses on neighborhood differences such as in distance, reachability, accessibility, as well as quantitative and qualitative characteristics of healthcare facilities. Studies have in fact shown that individuals in socioeconomically disadvantaged neighborhoods are more likely to use and receive care from lower care quality hospitals and home health agencies (HHAs) based on quality measures such as the star rating (Fahrenbach et al., 2020; Rahman & Foster, 2015; Smith et al., 2008). Previous studies have also highlighted the importance of care quality in health outcomes (Agency for Healthcare Research and Quality, 2020; Centers for Medicare & Medicaid Services, 2021), in which using healthcare institutions offering low-quality care could lead to worse health outcomes. Additionally, considering the importance of social and structural support and resources in recovering from an adverse health symptom or event (Li et al., 2013; Roberts et al., 2019), where people reside could affect differences in health outcomes among those who receive needed care at their home and community. Yet, we have limited information on neighborhood-level variations in health outcomes and the role of neighborhood context on health outcomes for patients receiving Medicare home health care.

This study, therefore, examined differences in successful discharge to community by neighborhood socioeconomic status (SES) using 2019 Outcome and Assessment Information Set (OASIS) data and Area Deprivation Index (ADI). Furthermore, we examined whether successful discharge to community varies by neighborhood SES even when patients use the same HHA. Understanding the role of neighborhood context in Medicare home health care can guide policies designed to reduce disparities and achieve equitable home health outcomes, as well as improve overall health outcomes and reduce healthcare costs.

Materials and Methods

Study Design and Data Sources

The study used the 2019 OASIS, the 2019 Medicare Beneficiary Summary File (MBSF), and the 2019 Centers for Medicare & Medicaid Services (CMS) providers of services data and home health star rating data. The OASIS data is collected by a home care clinician and provides a comprehensive assessment (more than 100 items), including home health care users’ demographic information, clinical status, functional status, and service needs (O’Connor & Davitt, 2012). Also, we used the MBSF base and chronic condition segments to identify demographic characteristics (sex and race/ethnicity), Medicare-managed care enrollment, and chronic conditions. In addition, we used the CMS providers of service data to identify HHA characteristics, including ownership, affiliation, and accreditation status.

We also used the 2019 ADI to identify Medicare home health care users’ neighborhood SES (Kind & Buckingham, 2018; University of Wisconsin School of Medicine and Public Health, 2019). ADI is a composite index of 17 measures of poverty, education, housing, and employment at the census block group level developed based on American Community Survey data. ADI is a standardized score ranging from 0 to 100, in which 50 indicates the mean, and a higher ADI score indicates greater area deprivation. We merged the OASIS and ADI based on users’ 9-digit zip codes.

Study Population

We focused on Medicare-enrolled home health care patients aged 65 or older. We limited our analysis to Medicare Fee-for-Service patients, by using HMO indicators in the MBSF data, due to the difference in home health care use patterns and HHAs by Medicare program (Schwartz et al., 2019; Waxman et al., 2016). We excluded home health care patients without an identifiable 9-digit zip code as linking with ADI data in the absence of a geographic identifier was not possible. Based on the calculation of home health care quality measures used by CMS (Centers for Medicare & Medicaid Services, 2019b), we used home health care episodes ending with a discharge to community or transfer to an inpatient facility, as both as needed to calculate successful discharge to community (please refer to 2.3. Variables) and excluded those that end in patient death. Our final analytic sample included 1,884,069 patients from 9,522 HHAs (detail on the analytic sample after including the study criteria is outlined in Supplementary Table S1).

Variables

The study outcome was successful discharge to community, which is one of the home health care quality measures developed by CMS. Recovering the ability to return and remain in the community is a desired patient outcome, the primary goal of home health care, and the currently used quality measure by CMS (Centers for Medicare and Medicaid Services, 2016, 2019a; Kus et al., 2011; van Seben et al., 2019). We defined “successful discharge to community” consistent with CMS definition. Accordingly, this measure indicates home health episodes from patients with a completed assessment at discharge that indicates the patient remained in the community after discharge and that they were not transferred to an inpatient facility during home health care (Centers for Medicare & Medicaid Services, 2019b). The primary predictor of interest was the patients/home health care users’ neighborhood SES measured using ADI. We used ADI to classify the patients’ neighborhoods into two groups: less disadvantaged neighborhoods versus most disadvantaged neighborhoods. In line with previous studies, the most disadvantaged neighborhoods are defined as neighborhoods with an ADI score above the 85^th percentile (ADI ≥85; Kind et al., 2014). We included covariates based on the Aday–Andersen health behavior model representing predisposing, enabling, or health need factors (Aday & Andersen, 2005). First, we included age, gender, race, and living arrangements as predisposing factors. We also included dual enrollment in Medicare and Medicaid as enabling factors. We captured health needs by indicators of chronic conditions, limitations in functional abilities, cognitive impairment, and risk for hospitalization. Using the CMS data, we included HHA-level variables, which include HHA ownership, HHA size, accrediting status, CMS program enrollment, and hospice care programs. Covariate categories are described in Supplementary Table S3.

Analytic Approach

We first compared the characteristics of Medicare home health care patients by neighborhood SES using $χ^{2}$ test for categorical variables and t-test for continuous variables. Next, we used multivariate logistic regression models to examine the association between neighborhood SES and successful discharge to community adjusting for individual- and HHA-level characteristics. Statistical significance was set at p < .05 for all analyses. Moreover, to examine whether successful discharge to community varies by neighborhood SES for those patients using the same HHA, we performed a conditional logistic regression model stratifying by HHAs. Conditional logistic regression is a type of logistic regression that estimates the odds ratio of the outcome based on the predictor variables while taking into account the clusters. Based on the regression model adjusted by individual- and HHA-level characteristics, we also calculated and compared the average adjusted predicted probabilities of outcomes by neighborhood SES. In addition, we stratified and compared the predicted probabilities of successful discharge to community by the percentage of patients from the most disadvantaged neighborhoods in HHA.

We conducted several sensitivity analyses to assess the robustness of our results. First, we tested whether the study results vary by HHAs in the states in which the Home Health Value-Based Purchasing (HHVBP) demonstration program was implemented to encourage HHAs to improve care quality. Since 2016, all Medicare-certified HHAs in nine randomly selected states (Arizona, Florida, Iowa, Maryland, Massachusetts, Nebraska, North Carolina, Tennessee, and Washington) have been participating in the HHVBP model. While we have limited information on the effect of HHVBP on care quality disparities, we aimed to test whether the pattern of study results is consistent between HHVBP and non-HHVBP states. Also, we conducted the analyses without patients living in a congregate setting (assisted living facility, residential care home, or personal care home) due to the relationships between the living facility and specific HHAs. In addition, we examined different thresholds of ADI to define neighborhood SES. Finally, we divided the sample into four groups to define neighborhood socioeconomic status (ADI <15th percentile, 15th–50th percentile, 51st–85th percentile, and ADI >85^th percentile). We use R Studio 4.4.1 to conduct all analyses.

Results

Table 1 describes the characteristics of the 1,884,069 Medicare home health care patients in 2019 included in this study. Patients in the most disadvantaged neighborhoods were less likely to be White (most disadvantaged neighborhoods: 70.1%; less disadvantaged neighborhoods: 81.9%), and more likely to live alone (most disadvantaged neighborhoods: 29.2%; less disadvantaged neighborhoods: 23%) and to be dually enrolled in Medicare–Medicaid (most disadvantaged neighborhoods: 29.9%; less disadvantaged neighborhoods: 18.3%) than those from less disadvantaged neighborhoods. Also, the most disadvantaged neighborhood sample was more likely to have chronic conditions, including chronic obstructive pulmonary disease (most disadvantaged neighborhoods: 39.2%; less disadvantaged neighborhoods: 31.8%), diabetes (most disadvantaged neighborhoods: 45.5%; less disadvantaged neighborhoods: 39%), and heart failure (most disadvantaged neighborhoods: 42.2%; less disadvantaged neighborhoods: 37.4%). In addition, the most disadvantaged neighborhood sample was less likely to use non-profit HHAs (most disadvantaged neighborhoods: 28.7%; less disadvantaged neighborhoods: 33.4%), HHAs accredited by the Accreditation Commission for Health Care, Community Health Accreditation Program, or Joint Commission (most disadvantaged neighborhoods: 30.4%; less disadvantaged neighborhoods: 38.3%), and more likely to use HHAs with multiple branches (most disadvantaged neighborhoods: 43.8%; less disadvantaged neighborhoods: 37.8%) than the less disadvantaged neighborhood sample.

Table 1.

Characteristics of Patients by Neighborhood Disadvantage.

	Overall	LDN	MDN	p-Value
N	1,884,069	1,601,485	282,584
Age (%)				<.001
65–69	14.4	13.7	18.1
70–74	17.1	16.8	18.7
75–79	18.3	18.2	18.7
80–84	18.5	18.6	17.8
85–89	16.8	17.1	14.9
90–94	10.9	11.3	8.8
95+	4	4.2	3
Gender, female (%)	38.6	38.8	37.1	<.001
Race (%)				<.001
White	80.1	81.9	70.1
Black	9.5	7.4	21.5
Hispanic	5.8	5.7	6.2
Asian/PI	2.7	3	0.6
Other	2	2.1	1.6
Living alone (%)	24	23	29.2	<.001
Medicare–Medicaid duals (%)	20	18.3	29.9	<.001
ADL score (mean (SD))	3.78 (1.19)	3.77 (1.19)	3.82 (1.22)	<.001
Risk for hospitalization (%)
History of falls	36.2	36.2	36.1	.302
Multiple hospitalizations in the past	23.8	23.3	26.7	<.001
Multiple ED visits in the past	21.3	20.8	24.6	<.001
Currently taking 5 or more medications	92.8	92.7	93.1	<.001
Currently reports exhaustion	47.7	47.8	47.1	<.001
History of difficulty in medical instructions compliance	23.9	23.7	25.2	<.001
Others	27.4	27.6	26.1	<.001
Discharge from (%)
Skilled nursing facility	16.1	16.3	14.4	<.001
Short-stay acute hospital	45.6	45.6	45.7	.435
Cognitive impairment (%)	49.4	49.4	49.5	.316
Chronic conditions (%)
Anemia	52	51.7	53.6	<.001
Asthma	11.9	12	11.8	.015
Atrial fibrillation and flutter	33.5	33.7	32.5	<.001
BPH	18.3	18.7	16.2	<.001
Cataract	20	20.4	17.9	<.001
Chronic kidney	39.3	38.7	43	<.001
COPD	32.9	31.8	39.2	<.001
Depression	38.4	38.6	37.5	<.001
Diabetes	40	39	45.5	<.001
Glaucoma	16.7	17	15.5	<.001
Heart failure	38.2	37.4	42.2	<.001
Hyperlipidemia	73.6	73.6	73.6	.693
Hypertension	84.7	84.4	86.4	<.001
Hypothyroidism	30.9	31.2	29.3	<.001
Ischemic heart disease	44.5	44	47.5	<.001
Non-Alzheimer’s dementia	28	28.3	26.2	<.001
Osteoporosis	20.8	21.3	18.1	<.001
Pneumonia	22.5	22.2	24.4	<.001
Rheumatoid arthritis/osteoarthritis	63.7	63.5	64.4	<.001
Stroke	19.5	19.4	19.9	<.001
HHA ownership (%)				<.001
For-profit	67.3	66.6	71.3
Non-profit	32.7	33.4	28.7
HHA affiliation (%)				<.001
Freestanding	87.5	87.5	87.7
Hospital-based	12.5	12.5	12.3
HHA CMS program (%)				<.001
Medicare only	16.7	16.9	15.4
Medicare and Medicaid	83.3	83.1	84.6
HHA accreditation (%)				<.001
Unaccredited	62.9	61.7	69.6
Accredited	37.1	38.3	30.4
HHA hospice (%)				<.001
Without hospice	91.4	91.1	92.9
With hospice	8.6	8.9	7.1
HHA size (%)				<.001
Single branch	61.3	62.2	56.2
Multiple branches	38.7	37.8	43.8

Note. LDN = less disadvantaged neighborhood; MDN = most disadvantaged neighborhood; ED = Emergency Department; BPH = benign prostatic hyperplasia.

*We only displayed risks for hospitalization with >20%, discharge location and chronic conditions with >10%.

Table 2 shows results from the multivariate logistic regression models, which examine the association between neighborhood SES and successful discharge to community. The unadjusted odds of successful discharge to community for patients in the most disadvantaged neighborhoods were lower than those for patients in less disadvantaged neighborhoods (OR, 0.75; 95% CI, 0.74–0.75; p < .001). Adjusting for individual-level characteristics narrowed the differences in odds for successful discharge to community by neighborhood disadvantage (OR, 0.82; 95% CI, 0.82–0.83; p < .001), and the odds ratio was similar with the addition of HHA-level characteristics (OR, 0.84; 95% CI, 0.83–0.85; p < .001).

Table 2.

Association Between Neighborhood Disadvantage and Successful Discharge to Community (reference group: Patients in less disadvantaged neighborhoods).

	Multivariate Logistic Regression Models		Conditional Logistic Regression Models^a
	Odds ratio (95% CI)	p-Value	Odds ratio (95% CI)	p-Value
Unadjusted	.75 (.74–.76)	<.001	.93 (.92–.93)	<.001
Model 1^b	.82 (.82–.83)	<.001	.96 (.95–.96)	<.001
Model 2^c	.84 (.83–.85)	<.001	—	—

^aConditional logistic regression models are clustered by home health agency.

^bAdjusted for individual-level factors (age, gender, race, living status, Medicaid enrollment, ADL score, risk for hospitalization, cognitive impairment, and chronic condition indicators (details described in Supplementary Table S2)).

^cAdjusted for individual- and home health agency-level factors (ownership, affiliation, CMS program enrollment, size, hospice care program, and accreditation status (details described in Supplementary Table S2)).

Note. CI = confidence interval.

The study results from the conditional logistic regression models show a similar pattern to that from multivariate logistic regression models. In the unadjusted conditional logistic regression model, the odds of successful discharge to community for patients in the most disadvantaged neighborhoods were lower than those for patients in less disadvantaged neighborhoods (OR, 0.93: 95% CI, 0.92–0.93; p < .001). Adjustment for individual-level characteristics showed a consistent pattern to the unadjusted model (OR, 0.96; 95% CI, 0.96–0.97; p < .001). Our sensitivity analyses (Supplementary Tables S4–S10) also demonstrated the robustness of the main findings, which were based on 1) patients living in a congregate setting, 2) subsample stratified by HHVBP program, and 3) different thresholds of ADI to define neighborhood SES.

Figure 1 displays the average adjusted predicted probabilities of successful discharge to community by neighborhood disadvantage. The average adjusted probability of successful discharge to community for people in less disadvantaged neighborhoods was 75.8%, which was approximately 3 percentage points higher than that for those in the most disadvantaged neighborhoods (72.9%).

Figure 1.

Average Predicted Probabilities of Successful Discharge to Community by Neighborhood Disadvantage. Note: Probabilities were calculated based on the model fully adjusted for individual- (age, gender, race, living status, Medicaid enrollment, ADL score, risk for hospitalization, cognitive impairment, and chronic condition indicators) and home health agency-level (ownership, affiliation, CMS program enrollment, size, hospice care program, and accreditation status (details described in Supplementary Table S2)) factors.

Figure 2 shows the average adjusted predicted probabilities of successful discharge to community stratified by percentage of patients from the most disadvantaged neighborhoods within an HHA. The average adjusted predicted probability of successful discharge to community was 76% in HHAs, with the lowest percentage of patients from the most disadvantaged neighborhoods (25% or less). The predicted probability of successful discharge to community decreased as the percentage of patients from the most disadvantaged neighborhoods in an HHA increased, with the probability of successful discharge to community being 71.3% among patients in HHAs with more than 75% of patients from the most disadvantaged neighborhoods (HHAs with ≤25% of patients from the most disadvantage neighborhoods: 76%; HHAs with 26–50% of patients from the most disadvantage neighborhoods: 73%; HHAs with 51–75% of patients from the most disadvantage neighborhoods: 71.2%). All differences were statistically significant (p < .001), except the difference between HHAs with 51–75% and with >75% of patients from the most disadvantaged neighborhoods (p = 1).

Figure 2.

Average Predicted Probabilities of Successful Discharge to Community by Percentage of Patients From the Most Disadvantaged Neighborhoods in Home Health Agency. Note: Probabilities were calculated based on the model fully adjusted for individual- (age, gender, race, living status, Medicaid enrollment, ADL score, risk for hospitalization, cognitive impairment, and chronic condition indicators) and home health agency-level (ownership, affiliation, CMS program enrollment, size, hospice care program, and accreditation status) factors (Table 2 –Model 2)); MDN = most disadvantaged neighborhoods; HHA = home health agency.

Discussion

This study shows the relationship between home health outcomes, namely, successful discharge to the community and neighborhood SES in Medicare home health care. We found that patients in the most disadvantaged neighborhoods were more likely to be racial/ethnic minorities, report worse health status, and live alone than those in less disadvantaged neighborhoods. Also, consistent with previous studies on other care settings (Bhavsar et al., 2018; Kind et al., 2014; Nacht et al., 2022; Scaria et al., 2020), we found that patients living in less disadvantaged neighborhoods were associated with better home health outcomes than those in the most disadvantaged neighborhoods. Our findings suggest that, for Medicare beneficiaries, remaining in the community and home through home health care is more challenging in the most disadvantaged neighborhoods than in less disadvantaged neighborhoods.

Our findings align with previous studies on neighborhood context. Several studies have shown that people in more disadvantaged neighborhoods are associated with worse health outcomes than their counterparts in less disadvantaged neighborhoods (Bhavsar et al., 2018; Kind et al., 2014; Nacht et al., 2022; Scaria et al., 2020). For example, Kind and colleagues have shown that living in the most disadvantaged neighborhoods is associated with a higher risk of hospital readmission among Medicare patients discharged with congestive heart failure, pneumonia, or myocardial infarction (Kind et al., 2014). Similarly, we found that patients living in the most disadvantaged neighborhoods experienced worse home health outcomes than their counterparts in less disadvantaged neighborhoods. Identifying whether home health outcome measures vary by neighborhood context is important to clarify whether Medicare beneficiaries are receiving the same level of care quality and also to improve our understanding of the factors that are conducive to successful discharge to the community. Policymakers should consider developing programs and interventions targeting patients and HHAs based on area-level indicators to improve Medicare home health care effectively.

The differences in home health outcomes by neighborhood context in Medicare home health care could be due to differences in accessibility to and use of high-quality HHAs. Previous studies have shown that patients in disadvantaged neighborhoods are more likely to use lower quality HHAs (Fashaw-Walters et al., 2022) as well as institutional care settings (Fahrenbach et al., 2020; Rahman & Foster, 2015; Smith et al., 2008). At the same time, the association may also raise the possibility that HHAs disproportionately serving patients in disadvantaged neighborhoods face more challenges in providing high care quality and achieving targeted quality performance. It is known that people in disadvantaged neighborhoods are more likely to have poorer health with a greater number of chronic conditions, functional limitations, and cognitive impairments (Basta et al., 2008; Robert, 1998; Ross & Mirowsky, 2001) with fewer resources to support them (Cutrona et al., 2006; Schieman, 2005; Soltero et al., 2015; Turney & Harknett, 2010). Our study results also present that HHAs serving more patients in disadvantaged neighborhoods were associated with a lower probability of successful discharge to community. While these findings suggest the difficulties faced by HHAs disproportionately serving patients in disadvantaged neighborhoods, the link between neighborhood context and the quality of home health care requires further examination. Future research on underlying mechanisms linking neighborhood context and low-quality HHAs could help identify further opportunities to improve care and support HHAs.

In addition, our results suggest the possibilities that the initiatives the CMS introduced to encourage HHAs to improve care quality and health outcomes may affect existing health disparities. For instance, the HHVBP program financially rewards or penalizes HHAs based on their quality performance and will be expanded nationwide in 2023 (Pozniak et al., 2022). HHAs in disadvantaged neighborhoods could face more challenges in improving care quality and health outcomes under HHVBP. By shifting funding from poor-performing HHAs to better-performing HHAs, HHVBP can have unintended consequences of further reducing the quality of care at HHAs that mainly serve patients from disadvantaged neighborhoods. Consequently, patients in disadvantaged neighborhoods may experience worse care quality and health outcomes. Considering the differences in home health outcomes by neighborhood disadvantage, we should ensure that such initiatives do not inadvertently penalize specific HHAs and that there is appropriate support to overcome systematic or structural barriers to improve care quality and health outcomes.

Our study results further raise the possibility that the differences in home health outcomes by neighborhood SES could be partially explained by factors beyond the care provided by HHAs. Based on our conditional logistic regression model, even when patients use the same HHA, those living in the most disadvantaged neighborhoods are less likely to achieve desired outcomes than others. This could be due to external factors outside the HHAs’ control, inadvertently affecting patients’ health outcomes. Previous studies have highlighted the importance of external factors in post-acute care, including social support (Calvillo–King et al., 2013; Herrin et al., 2015). Considering that home health care involves receiving episodic/intermittent care at home and community without 24/7 total care from medical professionals and relying largely on informal caregivers, the impact of external factors on health outcomes may be undermined in home health care settings. Home health care patients living in disadvantaged neighborhoods have scarce access to these external resources, such as lower levels of social, financial, and caregiver support that may facilitate successful discharge to the community (Cutrona et al., 2006; Schieman, 2005; Settels, 2021; Soltero et al., 2015; Turney & Harknett, 2010); thus, home health users in more disadvantaged neighborhoods are more likely to experience worse health outcomes than others in less disadvantaged neighborhoods. Built environment such as housing could also contribute to poorer health in disadvantaged neighborhoods (Bosma et al., 2001; Fedorowicz et al., 2020). Therefore, there is a possibility that, despite receiving equal quality of care from the same HHA, there could be variations in home health outcomes by neighborhood context. We need further research to identify potential external factors contributing to such differences and clarify the effect of such factors on the association between neighborhood context and health outcomes to produce helpful information on designing plans to mitigate differences in health outcomes by neighborhood disadvantage.

Limitations

Our study has several limitations. This study focused on Medicare fee-for-service beneficiaries; thus, our findings may not be generalizable to Medicare Advantage or other health insurance beneficiaries. Moreover, our results may not be generalizable to the minority of HHA users who are not enrolled in Medicare nor non-Medicare HHAs. Third, this study was based on an observational study design, which limits our ability to make causal inferences. Fourth, this study excluded episodes from patients without available 9-digit zip codes or ADI. The excluded sample (Supplementary Table S2) due to non-available 9-digit zip codes or ADI was more likely to use HHAs that are non-profit, unaccredited, with hospice, and with multiple branches. Thus, our exclusion criteria may limit the generalizability of study results to the overall Medicare home health care users. Also, as ADI is developed based on the American Community Survey, study results are dependent on the accuracy and errors contained within the American Community Survey. Lastly, this study was based on 2019 data. CMS has announced to expand the HHVBP nationwide in 2022, which could affect the study results. While our sensitivity analyses have shown that the association between neighborhood context and successful discharge to community is consistent between states in which the HHVBP demonstration program was and was not implemented, further studies are needed to reflect this change.

Conclusion

Our study provides evidence of the association between neighborhood disadvantage and successful discharge to community among older adults in the context of Medicare home health care. As more Medicare beneficiaries are willing to receive needed care at their home, further efforts to reduce the gaps in home health outcomes and the use of HHAs are critical to ensure all Medicare beneficiaries can acquire the same home health care benefits. Our study results suggest area-level interventions and programs be considered to target and support patients in disadvantaged neighborhoods and HHAs mainly serving such patients to improve and reduce neighborhood-based gaps in Medicare home health care and outcomes.

Supplemental Material

Supplemental Material - Disparities in Successful Discharge to the Community Following Use of Medicare Home Health by Level of Neighborhood Socioeconomic Disadvantage

Supplemental Material for Disparities in Successful Discharge to the Community Following Use of Medicare Home Health by Level of Neighborhood Socioeconomic Disadvantage by Daniel Jung, Janani Rajbhandari-Thapa, and Zhuo Chen in Journal of Applied Gerontology.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Owens Institute for Behavioral Research, University of Georgia and National Institute on Aging, RF1AG054071.

IRB Approval

University of Georgia, Protocol Number: PROJECT00004674.

ORCID iD

Daniel Jung

Supplemental Material

Supplemental material for this article is available online.

References

Aday

L. A.

Andersen

R. M.

(2005). Health care utilization and behavior, models of. Encyclopedia of Biostatistics, 4, 2–5. https://doi.org/10.1002/0470011815.b2a4a010

Agency for Healthcare Research and Quality (2020, June). Understanding quality measurement. https://www.ahrq.gov/patient-safety/quality-resources/tools/chtoolbx/understand/index.html.

Basta

N. E.

Matthews

F. E.

Chatfield

M. D.

Brayne

, & (2008). Community-level socio-economic status and cognitive and functional impairment in the older population. European Journal of Public Health, 18(1), 48–54. https://doi.org/10.1093/EURPUB/CKM076

Bhavsar

N. A.

Gao

Phelan

Pagidipati

N. J.

Goldstein

B. A.

(2018). Value of neighborhood socioeconomic status in predicting risk of outcomes in studies that use electronic health record data. JAMA Network Open, 1(5), article e182716–e182716. https://doi.org/10.1001/JAMANETWORKOPEN.2018.2716

Bosma

Van De Mheen

H. D.

Borsboom

G. J. J. M.

Mackenbach

J. P.

(2001). Neighborhood socioeconomic status and all-cause mortality. American Journal of Epidemiology, 153(4), 363–371. https://doi.org/10.1093/AJE/153.4.363

Calvillo–King

Arnold

Eubank

K. J.

Yunyongying

Stieglitz

Halm

E. A.

(2013). Impact of social factors on risk of readmission or mortality in pneumonia and heart failure: Systematic review. Journal of General Internal Medicine, 28(2), 269–282. https://doi.org/10.1007/s11606-012-2235-x

Centers for Medicare and Medicaid Services (2016). Measure specifications for measures in the CY 2017 HH QRP final rule. https://www.cms.gov/medicare/quality-initiatives-patient-assessment-instruments/homehealthqualityinits/downloads/measurespecificationsforcy17-hh-qrp-frpdf.pdf

Centers for Medicare and Medicaid Services (2021). Quality measurement and quality improvement. https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/MMS/Quality-Measure-and-Quality-Improvement

Centers for Medicare and Medicaid Services . (2019a). What’s home health care? Medicare. https://www.medicare.gov/what-medicare-covers/whats-home-health-care

10.

Centers for Medicare and Medicaid Services (2019b). Home health quality measures – outcomes. https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/HomeHealthQualityInits/Downloads/Home-Health-Outcome-Measures-Table-OASIS-D-11-2018c.pdf

11.

Cuckler

G. A.

Sisko

A. M.

Poisal

J. A.

Keehan

S. P.

Smith

S. D.

Madison

A. J.

Wolfe

C. J.

Hardesty

J. C.

(2018). National health expenditure projections, 2017-26: Despite uncertainty, fundamentals primarily drive spending growth. Health Affairs, 37(3), 482–492. https://doi.org/10.1377/HLTHAFF.2017.1655

12.

Cutrona

C. E.

Wallace

Wesner

K. A.

(2006). Neighborhood characteristics and depression: An examination of stress processes. Current Directions in Psychological Science, 15(4), 188–192. https://doi.org/10.1111/J.1467-8721.2006.00433.X

13.

Dick

A. W.

Murray

M. T.

Chastain

A. M.

Madigan

E. A.

Sorbero

Stone

P. W.

Shang

(2019). Measuring quality in home healthcare. Journal of the American Geriatrics Society, 67(9), 1859–1865. https://doi.org/10.1111/JGS.15963

14.

Fahrenbach

Chin

M. H.

Huang

E. S.

Springman

M. K.

Weber

S. G.

Tung

E. L.

(2020). Neighborhood disadvantage and hospital quality ratings in the Medicare hospital compare program. Medical Care, 58(4), 376–383. https://doi.org/10.1097/MLR.0000000000001283

15.

Fashaw-Walters

S. A.

Rahman

Gee

Mor

White

Thomas

K. S.

(2022). Out of reach: Inequities in the use of high-quality home health agencies. Health Affairs, 41(2), 247–255. https://doi.org/10.1377/hlthaff.2021.01408

16.

Fedorowicz

Schilling

Bramhall

Bieretz

Steven Brown

(2020). Leveraging the built environment for health equity promising interventions for small and medium-size cities. Washington DC. https://www.urban.org/sites/default/files/publication/102557/leveraging-the-built-environment-for-health-equity.pdf

17.

Geng

Mansouri

Stevenson

D. G.

Grabowski

D. C.

(2020). Evolution of the home health care market: The expansion and quality performance of multi-agency chains. Health Services Research, 55(Suppl 3), 1073–1084. https://doi.org/10.1111/1475-6773.13597

18.

Gregory

Edwards

Faurot

Williams

S. W.

Felix

A. C. G.

(2010). Patient preferences for stroke rehabilitation. Topics in Stroke Rehabilitation, 17(5), 394–400. https://doi.org/10.1310/tsr1705-394

19.

Herrin

St Andre

Kenward

Joshi

M. S.

Audet

A.-M. J.

Hines

S. C.

(2015). Community factors and hospital readmission rates. Health Services Research, 50(1), 20–39. https://doi.org/10.1111/1475-6773.12177

20.

Kind

A. J. H.

Buckingham

W. R.

(2018). Making neighborhood-disadvantage metrics accessible — the neighborhood atlas. The New England Journal of Medicine, 378(26), 2456–2458. https://doi.org/10.1056/NEJMP1802313

21.

Kind

A. J. H.

Jencks

Brock

Bartels

Ehlenbach

Greenberg

Smith

(2014). Neighborhood socioeconomic disadvantage and 30-day rehospitalization: A retrospective cohort study. Annals of Internal Medicine, 161(11), 765–774. https://doi.org/10.7326/M13-2946

22.

Kus

Müller

Strobl

Grill

(2011). Patient goals in post-acute geriatric rehabilitation--goal attainment is an indicator for improved functioning. Journal of Rehabilitation Medicine, 43(2), 156–161. https://doi.org/10.2340/16501977-0636

23.

Morrow-Howell

Proctor

Rubin

(2013). Social support resources and post-acute recovery for older adults with major depression. Community Mental Health Journal, 49(4), 419–426. https://doi.org/10.1007/s10597-012-9567-1

24.

Mcwilliams

J. M.

Gilstrap

L. G.

Stevenson

D. G.

Chernew

M. E.

Huskamp

H. A.

Grabowski

D. C.

Stevenson Corresponding (2017). Changes in postacute care in the Medicare shared savings program. JAMA Internal Medicine, 177(4), 518–526. https://doi.org/10.1001/jamainternmed.2016.9115

25.

MedPAC (2021). Medicare payment advisory commission, report to the congress: Medicare payment policy. https://www.medpac.gov/wp-content/uploads/import_data/scrape_files/docs/default-source/reports/mar21_medpac_report_to_the_congress_sec.pdf

26.

Mohnen

S. M.

Schneider

Droomers

(2019). Neighborhood characteristics as determinants of healthcare utilization – a theoretical model. Health Economics Review, 9(1), 7. https://doi.org/10.1186/s13561-019-0226-x

27.

Nacht

C. L.

Kelly

M. M.

Edmonson

M. B.

Sklansky

D. J.

Shadman

K. A.

Kind

A. J. H.

Zhao

Barreda

C. B.

Coller

R. J.

(2022). Association between neighborhood disadvantage and pediatric readmissions. Maternal and Child Health Journal, 26(1), 31–41. https://doi.org/10.1007/S10995-021-03310-4

28.

O’Connor

Davitt

J. K.

(2012). The outcome and assessment information set (OASIS): A review of validity and reliability. Home Health Care Services Quarterly, 31(4), 267–301. https://doi.org/10.1080/01621424.2012.703908

29.

Pozniak

Turenne

Lammers

Mukhopadhyay

Slanchev

Green

Schur

Tomai

Arbit

Cogan

Ding

Goyat

Hanslits

Mandell

Milkovich

Repeck

Rochotte

Schrager

Bott

(2022). Evaluation of the home health value-based purchasing (HHVBP) model - fifth annual report. https://innovation.cms.gov/data-and-reports/2022/hhvbp-fifthann-rpt

30.

Rahman

Foster

A. D.

(2015). Racial segregation and quality of care disparity in US nursing homes. Journal of Health Economics, 39, 1–16. https://doi.org/10.1016/J.JHEALECO.2014.09.003

31.

Robert

S. A.

(1998). Community-level socioeconomic status effects on adult health. Journal of Health and Social Behavior, 39(1), 18–37. https://doi.org/10.2307/2676387

32.

Roberts

A. R.

Roberts

A. R.

Bowblis

J. R.

Smith

A. C.

(2019). The impact of social workers on post-acute care discharge outcomes. Innovation in Aging, 3(Supplement_1), S359–S360. https://doi.org/10.1093/GERONI/IGZ038.1309

33.

Ross

C. E.

Mirowsky

(2001). Neighborhood disadvantage, disorder, and health. Journal of Health and Social Behavior, 42(3), 258–276. https://doi.org/10.2307/3090214

34.

Scaria

Powell

W. R.

Birstler

Alagoz

Shirley

Kind

A. J. H.

Safdar

(2020). Neighborhood disadvantage and 30-day readmission risk following Clostridioides difficile infection hospitalization. BMC Infectious Diseases, 20(1), 762–810. https://doi.org/10.1186/s12879-020-05481-x

35.

Schieman

(2005). Residential stability and the social impact of neighborhood disadvantage: A study of gender- and race-contingent effects. Social Forces, 83(3), 1031–1064. https://doi.org/10.1353/sof.2005.0045, https://www.jstor.org/stable/3598268

36.

Schwartz

M. L.

Kosar

C. M.

Mroz

T. M.

Kumar

Rahman

(2019). Quality of home health agencies serving traditional Medicare vs Medicare advantage beneficiaries. JAMA Network Open, 2(9), article e1910622. https://doi.org/10.1001/jamanetworkopen.2019.10622

37.

Settels

(2021). Changes in neighborhood-level concentrated disadvantage and social networks among older Americans. Journal of Applied Gerontology: The Official Journal of the Southern Gerontological Society, 40(10), 1288–1296. https://doi.org/10.1177/0733464820969054

38.

Smith

D. B.

Feng

Fennell

M. L.

Zinn

Mor

(2008). Racial disparities in access to long-term care: The illusive pursuit of equity. Journal of Health Politics, Policy and Law, 33(5), 861–881. https://doi.org/10.1215/03616878-2008-022

39.

Soltero

E. G.

Hernandez

D. C.

O’Connor

D. P.

Lee

R. E.

(2015). Does social support mediate the relationship among neighborhood disadvantage, incivilities, crime and physical activity? Preventive Medicine, 72, 44–49. https://doi.org/10.1016/J.YPMED.2014.12.030

40.

Turney

Harknett

(2010). Neighborhood disadvantage, residential stability, and perceptions of instrumental support among new mothers. Journal of Family Issues, 31(4), 499–524. https://doi.org/10.1177/0192513X09347992

41.

University of Wisconsin School of Medicine and Public Health (2019). 2019 area deprivation index v2.0. https://www.neighborhoodatlas.medicine.wisc.edu/

42.

van Seben

Smorenburg

S. M.

Buurman

B. M.

(2019). A qualitative study of patient-centered goal-setting in geriatric rehabilitation: Patient and professional perspectives. Clinical Rehabilitation, 33(1), 128–140. https://doi.org/10.1177/0269215518791663

43.

Waxman

D. A.

Min

Setodji

C. M.

Hanson

Wenger

N. S.

Ganz

D. A.

(2016). Does Medicare advantage enrollment affect home healthcare use? The American Journal of Managed Care, 22(11), 714–720. www.ajmc.com

44.

Werner

R. M.

Coe

N. B.

Konetzka

R. T.

(2019). Patient outcomes after hospital discharge to home with home health care vs to a skilled nursing facility. JAMA Internal Medicine, 179(5), 617–623. https://doi.org/10.1001/JAMAINTERNMED.2018.7998

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.38 MB