Comprehensive Clinical Decision Support for the Primary Care of Premature Infants

描述（由申请人提供）： 早产儿是一个脆弱的人群，有多种相互关联的健康问题，使他们面临不良后果的风险。电子健康记录捕获大量的信息，可以帮助指导决策，但现有的警报和基于警报器的临床决策支持（CDS）框架没有充分应用多个重叠的护理指南，以复杂的病人的历史，以产生连贯的临床建议。我们提出的研究直接解决了10信息技术处理医疗保健数据的研究和特定的挑战主题10-LM-102复杂临床决策的高级决策支持的广泛挑战领域。本研究将使用嵌入在电子健康记录（EHR）中的基于规则的专家系统来提取、解释和呈现与早产儿医疗保健相关的突出事实和建议。这一提议也满足了通过保留或创造就业机会和增加特拉华州山谷地区的支出来立即刺激经济的挑战。费城儿童医院（CHOP）为当地经济做出了巨大贡献。2008年，CHOP的业务在该地区创造和支持了超过16，882个工作岗位，CHOP的总经济影响超过20.1亿美元。此外，通过私人捐款、国家卫生研究院的资助和医院业务的拨款，CHOP获得的研究资助总额超过美国任何其他儿童医院-2007-2008财政年度为1.8亿美元。目前的计划将创造或保留5个工作岗位。早产儿的护理是美国迅速增长的公共卫生问题，2006年有超过60，000名婴儿出生时体重低于1500克1。随着新生儿护理的最新进展，越来越多的早产儿从新生儿重症监护室（NICU）存活出院。尽管这一高风险患者群体对他们出院后接受的护理特别敏感，但确保高质量出院后护理的努力却杂乱无章。电子健康记录与嵌入式临床决策支持工具，如基于规则的专家系统提供了一个自然的机会，有利地影响这些脆弱的婴儿的医疗保健。这些婴儿的复杂决策必须考虑大量随时间变化的变量。覆盖所有可能的变量组合所需的规则矩阵是巨大的。到目前为止，还没有这样的决策支持工具已经实施或评估，以处理早产儿的医疗保健所需的决策的复杂性。本项目将采用前瞻性分组随机设计，评估健康信息技术干预的成功性，以提高低出生体重（LBW）和极低出生体重（VLBW）早产儿从重症监护室出院到24个月大的护理质量。干预措施将包括（1）电子健康记录中的实时数据挖掘工具，该工具将以简洁的时间线格式直观地组织大量健康信息;（2）基于规则的专家系统，其将预测适合于LBW和VLBW婴儿的即将到来的预防性健康评估和干预的时间表，以及（3）一套以问题为中心的专家系统，以指导与该弱势群体经历的最普遍的合并症和并发症相关的决策。评价将集中在三个方面：（1）自动数据挖掘工具和专家系统实施前的可用性;（2）干预前后临床医生对LBW和VLBW婴儿标准护理过程的了解的变化;（3）与对照实践相比，干预实践中护理过程结果的变化。拟议的工作将提高在复杂的临床领域提供临床决策支持（CDS）的能力。设计阶段和干预阶段的严格评价将支持未来CDS开发的最佳实践出版物。在电子健康记录（EHR）中具有适度自定义编程“足迹”的基于标准的网络服务模型的方法将促进将此和未来的CDS干预分发给其他医疗保健组织和EHR供应商。拟议的工作将提高在复杂的临床领域提供临床决策支持（CDS）的能力。设计阶段和干预阶段的严格评价将支持未来CDS开发的最佳实践出版物。在电子健康记录（EHR）中具有适度自定义编程“足迹”的基于标准的网络服务模型的方法将促进将此和未来的CDS干预分发给其他医疗保健组织和EHR供应商。

DESCRIPTION (provided by applicant): Premature infants are a vulnerable population with multiple inter-related health problems that put them at risk for poor outcomes. Electronic health records capture large amounts of information that may help guide decisions, but existing alert and reminder-based clinical decision support (CDS) frameworks do not adequately apply multiple overlapping care guidelines to complex patient histories to produce coherent clinical recommendations. Our proposed study directly addresses the broad challenge area of 10 Information Technology for Processing Health Care Data for Research and the specific challenge topic 10-LM-102 Advanced Decision Support for Complex Clinical Decisions. This study will use a rules-based expert system embedded in an electronic health record (EHR) to extract, interpret, and present salient facts and recommendations related to the healthcare of premature infants. This proposal also meets the challenge of providing an immediate stimulus to the economy through the retention or creation of jobs and increased spending in the Delaware Valley region. The Children's Hospital of Philadelphia (CHOP) contributes substantially to the local economy. In 2008, CHOP's operations created and supported over 16,882 jobs in the region, and CHOP's total economic impact was over $2.01 billion. Moreover, through a combination of private donations, NIH funding, and allocations from its hospital operations, CHOP receives more total research support than any other children's hospital in the United States-$180 million in fiscal year 2007-2008. The current proposal will create or retain 5 jobs. The care of premature infants is a rapidly growing public health concern in the United States, with over 60,000 infants born in 2006 with a birth weight under 1500 grams1. With recent advances in neonatal care, more premature infants are surviving to discharge from the neonatal intensive care unit (NICU). Even though this high-risk group of patients is particularly sensitive to the care they receive after discharge, efforts to ensure high quality post-discharge care are haphazardly implemented. Electronic health records with embedded clinical decision support tools such as rules-based expert systems provide a natural opportunity to favorably affect the healthcare for these vulnerable infants. The complex decision-making for these infants must consider large numbers of variables that change over time. The matrices of rules required to cover all possible combinations of variables are huge. To date no such decision support tools have been implemented or evaluated to handle the complexity of decision-making required for the healthcare of premature infants. This project will use a prospective cluster randomized design to evaluate the success of a health information technology intervention to improve care quality for low birth weight (LBW) and very low birth weight (VLBW) premature infants from the time of intensive care nursery discharge through 24 months of age. The intervention will consist of (1) a real-time data mining tool functioning in the EHR that will organize large amounts of health information visually in a succinct time-line format; (2) a rules-based expert system that will forecast a schedule for upcoming preventive health assessments and interventions appropriate for LBW and VLBW infants, and (3) a suite of problem focused expert systems to guide decision-making related to the most prevalent co-morbidities and complications experienced by this vulnerable population. Evaluation will focus on three areas: (1) usability of the automated data-mining tool and expert system pre-implementation; (2) change in clinician knowledge of standard care process for LBW and VLBW infants pre- and post-intervention; and (3) change in care process outcomes in the intervention practices compared to the control practices. The proposed work will improve the capacity for delivering clinical decision support (CDS) in complex clinical domains. Rigorous evaluation of both the design phase and intervention phase will support publications regarding best practices for future CDS development. The approach of a standards-based web-service model with a modest custom programming "footprint" in the electronic health record (EHR) will facilitate the distribution of this and future CDS interventions to other healthcare organizations and EHR vendors. The proposed work will improve the capacity for delivering clinical decision support (CDS) in complex clinical domains. Rigorous evaluation of both the design phase and intervention phase will support publications regarding best practices for future CDS development. The approach of a standards-based web-service model with a modest custom programming "footprint" in the electronic health record (EHR) will facilitate the distribution of this and future CDS interventions to other healthcare organizations and EHR vendors.