The Role of Big Data in the Management of Sleep-Disordered Breathing.
The Role of Big Data in the Management of Sleep-Disordered Breathing.
复制标题
DOI：
10.1016/j.jsmc.2016.01.009
复制
发表时间：
2016-06
影响因子：
2.8
通讯作者：
Redline S
中科院分区：
文献类型：
--
作者：
Budhiraja R;Thomas R;Kim M;Redline S
尽管在医疗保健方面有相当大的投资，但我们的医疗保健系统在满足质量指标、患者满意度和临床结果目标方面往往不足。1此外，尽管在临床研究方面进行了大量投资，但我们的研究基础设施往往不能产生支持“循证”医疗保健所需的数据。缺陷的部分原因是在医疗保健研究的设计中，包括患者在内的关键利益攸关方的代表性不足，以及由于缺乏足够数量患者的有效和全面数据。这些差距在睡眠医学中尤其明显，缺乏数据来解决以下问题：哪些患者受益于替代治疗（例如，CPAP，手术，口腔器械，氧气补充）;睡眠呼吸暂停的心脏代谢和认知影响的个体间差异的基础;以及CPAP治疗后CPAP依从性或残留嗜睡的个体决定因素。这些差距是由于现有数据的局限性，这些数据往往不足以产生支持医疗保健决策的有力证据。同时存在丰富的数据（例如，合规跟踪）和数据浪费（例如，专注于保险支付要求，而丢弃其他可用数据）。生成数据的传统做法被比作在孤岛中工作。数据孤岛包括诸如临床遭遇的电子医疗记录、睡眠实验室数据库、在线气道正压跟踪数据库等，这些数据库位于独立的非通信区域。
Despite considerable investments in health care, our health care systems often fall short in meeting quality metrics, patient satisfaction, and clinical outcomes goals. 1 Furthermore, despite large investments in clinical research, our research infrastructure often does not yield the needed data to support “evidence-based” health care. Deficiencies are partially due to inadequate representation of key stakeholders, including patients, in the design of health care research, as well as due to the overall paucity of valid and comprehensive data on sufficient numbers of patients. These gaps are especially evident in sleep medicine, where data are lacking that address questions such as which patients benefit from alternative treatments (eg, CPAP, surgery, oral appliances, oxygen supplementation); the basis for inter-individual variation in cardiometabolic and cognitive effects of sleep apnea; and the individual determinants of CPAP adherence or residual sleepiness after CPAP treatment. These gaps are due to limitations of available data, which too often are inadequate to generate robust evidence for supporting health care decisions. There is simultaneously a richness of data (eg, compliance tracking) and data waste (eg, focusing on insurance payments requirements while discarding other available data). The traditional practice of generating data has been compared to working in silos. Data silos include those such as electronic medical records of clinical encounters, sleep laboratory databases, online positive airway pressure tracking databases that reside in independent non-communicating areas.