Division of Community Internal Medicine, Mayo Clinic, Rochester, MN, USA; Division of Geriatric Medicine and Gerontology, Mayo Clinic, Rochester, MN, USA.
Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery, Mayo Clinic, Rochester, MN, USA.
J Am Med Dir Assoc. 2021 Oct;22(10):2154-2159.e1. doi: 10.1016/j.jamda.2020.12.009. Epub 2021 Jan 11.
Health care providers at hospitals and skilled nursing facilities (SNFs) are increasingly expected to optimize care of post-acute patients to reduce hospital readmissions and contain costs. To achieve these goals, providers need to understand their patients' risk of hospital readmission and how this risk is associated with health care costs. A previously developed risk prediction model identifies patients' probability of 30-day hospital readmission at the time of discharge to an SNF. With a computerized algorithm, we translated this model as the Skilled Nursing Facility Readmission Risk (SNFRR) instrument. Our objective was to evaluate the relationship between 30-day health care costs and hospital readmissions according to the level of risk calculated by this model.
This retrospective cohort study used SNFRR scores to evaluate patient data.
The patients were discharged from Mayo Clinic Rochester hospitals to 11 area SNFs.
We compared the outcomes of all-cause 30-day standardized direct medical costs and hospital readmissions between risk quartiles based on the distribution of SNFRR scores for patients discharged to SNFs for post-acute care from April 1 through November 30, 2017.
Mean 30-day all-cause standardized costs were positively associated with SNFRR score quartiles and ranged from $9199 in the fourth quartile (probability of readmission, 0.27-0.66) to $2679 in the first quartile (probability of readmission, 0.07-0.13) (P ≤ .05). Patients in the fourth SNFRR score quartile had 5.68 times the odds of 30-day hospital readmission compared with those in the first quartile.
The SNFRR instrument accurately predicted standardized direct health care costs for patients on discharge to an SNF and their risk for 30-day hospital readmission. Therefore, it could be used to help categorize patients for preemptive interventions. Further studies are needed to confirm its validity in other institutions and geographic areas.
医院和熟练护理机构（SNF）的医疗保健提供者越来越需要优化对急性后患者的护理，以降低医院再入院率并控制成本。为了实现这些目标，提供者需要了解患者的医院再入院风险以及这种风险与医疗保健成本的关系。先前开发的风险预测模型可在患者出院至 SNF 时确定其 30 天内医院再入院的概率。通过计算机算法，我们将该模型转化为熟练护理机构再入院风险（SNFRR）工具。我们的目标是根据该模型计算的风险水平评估 30 天内医疗保健成本与医院再入院之间的关系。
这项回顾性队列研究使用 SNFRR 评分评估患者数据。
患者从梅奥诊所罗切斯特医院出院到 11 个地区的 SNF。
我们根据 2017 年 4 月 1 日至 11 月 30 日出院到 SNF 进行急性后护理的患者的 SNFRR 评分分布，比较了所有原因 30 天标准化直接医疗费用和医院再入院率的风险四分位数。
30 天内所有原因标准化费用与 SNFRR 评分四分位数呈正相关，范围从第四四分位数（再入院概率，0.27-0.66）的 9199 美元到第一四分位数（再入院概率，0.07-0.13）的 2679 美元（P≤0.05）。第四 SNFRR 评分四分位数的患者与第一四分位数相比，30 天内医院再入院的可能性高 5.68 倍。
SNFRR 工具准确预测了出院至 SNF 的患者的标准化直接医疗保健成本及其 30 天内医院再入院的风险。因此，它可以用于帮助对患者进行分类以进行预防性干预。需要进一步的研究来确认其在其他机构和地理区域的有效性。
