Determining Barriers to Achieving Optimal Post-Acute Care Destinations

确定实现最佳急性后护理目的地的障碍

基本信息

批准号：
10226410
负责人：
Erin Kennedy
金额：
$ 3.74万
依托单位：
UNIVERSITY OF PENNSYLVANIA
依托单位国家：
美国
项目类别：
财政年份：
2021
资助国家：
美国
起止时间：
2021-09-01 至 2022-04-30
项目状态：
已结题

来源：
https://reporter.nih.gov/project-details/10226410
关键词：
Address Agreement Algorithms Bioinformatics Caring Characteristics Clinical Clinical Decision Support Systems Communication Communities Community Hospitals Complex Data Data Set Decision Making Destinations Discharge Plannings Discipline of Nursing Elderly Eligibility Determination Expert Systems Face Family Future Goals Guidelines Health Health Care Costs Health care facility Health system Healthcare Home Home Care Services Hospitalization Hospitals Individual Inpatients Insurance Intelligence Knowledge Learning Literature Location Medicare Mentors Methods Mission National Institute of Nursing Research Natural Language Processing Nature Nurses Outcome Patient-Focused Outcomes Patients Persons Population Process Prospective Studies Recommendation Rehabilitation therapy Replacement Arthroplasty Research Research Training Risk Risk Reduction Services Skilled Nursing Facilities Standardization Strategic Planning System Techniques Technology Testing Text Time United States Urban Hospitals Variant Work acute care base beneficiary care systems cost economic disparity evidence base experience hospital readmission hospital utilization improved inferential statistics innovation innovative technologies novel point of care prevent readmission rates readmission risk social disparities suburb suburban communities training opportunity unnecessary treatment

项目摘要

Project Summary 13 million Medicare beneficiaries are discharged from acute care hospitals annually, and approximately 42% of these older adults receive referrals to post-acute care (PAC) services including long term acute care hospitals, inpatient rehabilitation, skilled nursing facilities, and home health care. Effective referrals that promote patient health and prevent negative outcomes rely on coordinated discharge planning. However, this coordination is difficult to achieve when interprofessional discharge planning teams frequently face time constraints, team communication issues, variance in risk tolerance in decision making, and inconsistent assessments. Therefore, significant variation in discharge planning practices exists at the individual and hospital level and there are no clinical guidelines for this common but crucial process. Without standardized discharge planning practices in place, patients are at risk for negative outcomes after discharge including social and economic disparities in PAC referral location, unnecessary treatments, unplanned hospital readmissions, and increased healthcare costs. Clinical decision support systems (CDSS) equip clinicians with evidence-based, individualized information about their patients at the point of care, and address the urgent need for standardized solutions to improve discharge planning decisions. The Discharge Referral Expert System for Care Transitions (DIRECT) is a recently developed CDSS algorithm (RO1-2-NR007674) that identifies which patients need PAC and suggests the level of care as home health care or facility-level care based on patient needs. Use of DIRECT in discharge planning is associated with a reduction in hospital readmissions, however, hospital clinicians referred 26% fewer patients to PAC than DIRECT. This discordance has been historically difficult to study due to the unstructured nature of discharge planning data in clinical notes, making data abstraction and analysis difficult to achieve. The proposed study prepares the applicant to advance the DIRECT algorithm and expand it to a new clinical setting through two specific aims: 1) Among patients discharged without PAC, compare patient characteristics and 30-day readmission rates between those identified by DIRECT as needing PAC and those where DIRECT and clinicians agreed on no referral for PAC and 2) Identify the reasons associated with discharge home without services when the DIRECT CDSS recommends PAC. The proposed study will expand an existing CDSS developed in a suburban community hospital to a new population in a large urban hospital and utilize natural language processing methods to advance the understanding of why some patients do not receive the recommended level of PAC. The findings from this study will illuminate possible implementation and algorithm refinement strategies for future prospective study, and align with the applicant’s long term research goals to improve transitions in care for older adults by developing and implementing CDSS. The results of this study address the National Institute of Nursing Research’s mission to improve the health of older adults by supporting innovative technology.

项目摘要每年有1300万Medicare受益人与急诊医院脱节，约占42％这些老年人接受了急性后护理（PAC）服务的转诊，包括长期急性护理护理医院，住院康复，熟练的护理设施和家庭保健。健康并防止负面结果取决于协调的排放计划。当跨专业分期出院计划团队经常面临时间限制时，DIFFICALT将实现沟通，决策中的风险承受能力差异以及评估不一致。在个人和医院级别上的出院计划实践的显着差异，否否。普通但至关重要的过程的临床准则。地方，患者有出院后有负面结果的风险，包括社会和经济差异 PAC推荐地点，不必要的治疗，未植入的医院再入院和增加医疗保健费用。有关患者护理时的信息，并解决了对标准化解决方案的迫切需求改善出院计划的决定。最近开发的CDSS算法（RO1-2-NR007674），它注视着患者需要PAC和PAC的CDSS算法建议作为医疗保健或设施级别的照顾水平。排放计划与霍斯疗再入院的修订有关，但是，Hosspital临床医生将患者少26％，而不是直接研究。临床注释中排放计划数据的非结构化性质，使数据抽象和分析 DIFFICALT实现了支撑性研究，准备了直接算法的应用通过两个特定目标进行新的临床环境：1）在没有PAC的患者中，比较直接赋予thosenting thsection的患者特征和30天的重新启动率 PAC和那里的直接和临床医生不同意PAC的转诊，2）确定原因当直接CDS记录PAC时，与无服务的出院相关研究将将在郊区社区医院开发的现有CDS扩展到新的人群大型城市医院并利用本质语言处理方法托付理解一些患者未收到建议的PAC水平。未来前瞻性研究的可能的IMACTION和算法完善策略，并与之保持一致申请人的长期研究目标，以改善通过发展Anding来改善老年人的护理过渡实施CDS。通过支持创新技术来改善老年人的健康。

项目成果

期刊论文数量（2）

专著数量（0）

科研奖励数量（0）

会议论文数量（0）

专利数量（0）

Identifying Barriers to Post-Acute Care Referral and Characterizing Negative Patient Preferences Among Hospitalized Older Adults Using Natural Language Processing.

使用自然语言处理识别住院老年人中急性后护理转诊的障碍并描述负面患者偏好。

DOI：
发表时间：
2022
期刊：
AMIA ... Annual Symposium proceedings. AMIA Symposium
影响因子：
0
作者：
Kennedy,ErinE;Davoudi,Anahita;Hwang,Sy;Freda,PhilipJ;Urbanowicz,Ryan;Bowles,KathrynH;Mowery,DanielleL
通讯作者：
Mowery,DanielleL