Background: Hospital at Home (HaH) delivers hospital-level care to eligible patients in their homes, offering an alternative to traditional inpatient hospitalization. Longitudinal research has demonstrated that HaH can enhance patient outcomes, increase satisfaction, and decrease rates of hospital readmissions. Patients are first identified for HaH program enrollment through referrals by inpatient providers or our internal programmatic algorithm. Some patients are not enrolled in HaH due to clinical ineligibility or patient preference for inpatient hospital-based care. To better understand the reasons for decline from HaH, we analyzed the demographics of those who were screened but not subsequently transferred to HaH.
Methods: Patient records were included for all 7179 individuals screened for HaH between November 1, 2023 and October 31, 2024. Demographic data is unavailable (unknown or patient declined) for 5% of patients screened and 1% of those transferred into HaH, and these individuals were excluded from the analysis. We investigated the reasons for non-enrollment and conducted a demographic analysis of these patients by race/ethnicity and Medicaid coverage status using the two-proportion z-test.
Results: Among patients not transferred into HaH, the primary reasons for non-enrollment were clinical ineligibility (38.0%) and a planned discharge within 24 hours (25.5%). The third most common reason was patient preference for hospital-based care or lack of interest in the program, with an overall patient decline rate of 14.6%. Patients identifying as Asian were the most likely to decline participation (19.4%), while those identifying as Black were the least likely (13.9%). The fourth leading reason for non-enrollment was psychosocial status, such substance abuse disorders or uncontrolled mental illness (6.3%). Patients identifying as Black had the highest rate of psychosocial-related declines (9.2%), whereas those identifying as Asian had the lowest (1.6%). When analyzed by Medicaid Insurance Status, 51.6% of the patients screened and not enrolled in HaH have Medicaid as their primary or secondary insurance. Patients with Medicaid as their primary or secondary insurance had significantly higher rates of psychosocial-related declines (9.2%) versus those without Medicaid coverage (3.2%) (Z = 8.91, P < 0.001). There was no significant difference in patients’ declining participation between those who are covered by Medicaid (14.3%) vs those not covered (14.9%) (Z = -0.61, P = 0.540).
Conclusions: Our analysis demonstrated disparities in the reasons for HaH between racial / ethnic groups. Given these differences, future efforts should focus on enhancing inclusivity by investigating the reasons behind the disproportionate patient declines for HaH transfers, such as in the Asian population, and the disproportionate HaH program decline due to psychosocial barriers in the black population. In particular, psychosocial barrier exclusion to HaH should be further investigated for potential biases behind the status determination and for potential interventions or resources that can mitigate these barriers. These efforts will ensure Hospital at Home is more inclusive and reaches those who may have the most benefit from this service.