Arbor Research Collaborative for Health, Ann Arbor, Michigan.
JAMA Health Forum. 2022 Sep 2;3(9):e222723. doi: 10.1001/jamahealthforum.2022.2723.
The original Home Health Value-Based Purchasing (HHVBP) model provided financial incentives to home health agencies for quality improvement in 9 randomly selected US states.
To evaluate quality, utilization, and Medicare payments for home health patients in HHVBP states compared with those in comparison states.
DESIGN, SETTING, AND PARTICIPANTS: This cohort study was conducted in 2021 with secondary data from January 2013 to December 2020. A difference-in-differences design and multivariate linear regression were used to compare outcomes for Medicare and Medicaid beneficiaries who received home health care in HHVBP states with those in 41 comparison states during 3 years of preintervention (2013-2015) and the subsequent 5 years (2016-2020).
Home health care provided by a home health agency in HHVBP states and comparison states.
Utilization (unplanned hospitalizations, emergency department visits, skilled nursing facility [SNF] visits) for Medicare beneficiaries within 60 days of beginning home health, Medicare payments during and 37 days after home health episodes, and quality of care (functional status, patient experience) during home health episodes.
Among 34 058 796 home health episodes (16 584 870 beneficiaries; mean [SD] age of 76.6 [11.7] years; 60.5% female; 11.2% Black non-Hispanic; 79.5% White non-Hispanic) from January 2016 to December 2020, 22.6% were in HHVBP states and 77.4% were in non-HHVBP states. For the HHVBP and non-HHVBP groups, 60.4% and 61.0% of episodes were provided to female patients; 10.0% and 13.6% were provided to Black non-Hispanic patients, and 82.4% and 75.2% were provided to White non-Hispanic patients, respectively. Unplanned hospitalizations decreased by 0.15 percentage points (95% CI, -0.30 to -0.01) more in HHVBP states, a 1.0% decline compared with 15.7% at baseline. The use of SNFs decreased by 0.34 percentage points (95% CI, -0.40 to -0.27) more in HHVBP states, a 6.9% decline compared with the 4.9% baseline average. There was an association between HHVBP and a reduction in average Medicare payments per day of $2.17 (95% CI, -$3.67 to -$0.68) in HHVBP states, primarily associated with reduced inpatient and SNF services, which corresponded to an average annual Medicare savings of $190 million. There was greater functional improvement in HHVBP states than comparison states and no statistically significant change in emergency department use or most measures of patient experience.
In this cohort study, the HHVBP model was associated with lower Medicare payments that were associated with lower utilization of inpatient and SNF services, with better or similar quality of care.
最初的家庭保健价值为基础的购买（HHVBP）模式为 9 个随机选择的美国州的家庭保健机构提供了质量改进的财务激励。
评估 HHVBP 州与比较州的家庭保健患者的质量、利用率和医疗保险支付情况。
设计、设置和参与者：这是一项在 2021 年进行的队列研究，使用了 2013 年 1 月至 2020 年 12 月的二级数据。采用差异-差异设计和多元线性回归比较了 HHVBP 州和 41 个比较州的医疗保险和医疗补助受益人在 3 年的干预前（2013-2015 年）和随后的 5 年（2016-2020 年）期间接受家庭保健服务的结果。
HHVBP 州和比较州的家庭保健机构提供的家庭保健。
医疗保险受益人在开始家庭保健后 60 天内的利用情况（无计划的住院、急诊、熟练护理设施[SNF]就诊）、家庭保健期间和 37 天后的医疗保险支付情况以及家庭保健期间的护理质量（功能状态、患者体验）。
在 2016 年 1 月至 2020 年 12 月期间，34058796 个家庭保健事件（16584870 名受益人；平均[SD]年龄 76.6[11.7]岁；60.5%为女性；11.2%为黑人非西班牙裔；79.5%为白人非西班牙裔）中，22.6%来自 HHVBP 州，77.4%来自非-HHVBP 州。在 HHVBP 和非-HHVBP 组中，分别有 60.4%和 61.0%的病例为女性患者；10.0%和 13.6%为黑人非西班牙裔患者，82.4%和 75.2%为白人非西班牙裔患者。HHVBP 州的无计划住院率下降了 0.15 个百分点（95%置信区间，-0.30 至-0.01），降幅为 15.7%的基线水平的 1.0%。HHVBP 州的 SNF 使用量下降了 0.34 个百分点（95%置信区间，-0.40 至-0.27），降幅为 4.9%的基线平均水平的 6.9%。HHVBP 与医疗保险日平均支付减少 2.17 美元（95%置信区间，-3.67 至-0.68）之间存在关联，主要与减少住院和 SNF 服务有关，这相当于每年医疗保险节省 1.9 亿美元。HHVBP 州的功能改善优于比较州，急诊使用或大多数患者体验指标没有统计学上的显著变化。
在这项队列研究中，HHVBP 模式与较低的医疗保险支付有关，这与住院和 SNF 服务的利用率较低有关，同时也与较好或类似的护理质量有关。
