AI帮你理解科学

AI 生成解读视频

AI抽取解析论文重点内容自动生成视频


pub
生成解读视频

AI 溯源

AI解析本论文相关学术脉络


Master Reading Tree
生成 溯源树

AI 精读

AI抽取本论文的概要总结


微博一下
Beth Israel Deaconess Medical Center, an academic health care institution affiliated with Harvard University, is a typical user of big data for clinical care, education, and research

Early experiences with big data at an academic medical center.

HEALTH AFFAIRS, no. 7 (2014): 1132-1138

被引用9|浏览5
WOS
下载 PDF 全文
引用
微博一下

摘要

Beth Israel Deaconess Medical Center (BIDMC), an academic health care institution affiliated with Harvard University, has been an early adopter of electronic applications since the 1970s. Various departments of the medical center and the physician practice groups affiliated with it have implemented electronic health records, filmless imag...更多

代码

数据

简介
  • The definition of big data varies widely among industry experts. it is typically described as a large collection of disparate data sets that, taken together, can be analyzed to find unusual trends.
  • EHR data are typically stored in two ways: first, organized by patient in clinical systems for care coordination; and second, aggregated in analytic databases for purposes of population health, quality measurement, and care management.
  • The goal at BIDMC is to turn all of the collected data into actionable wisdom by applying decision support rules to actions and events such as changes in medications, patient visits, new lab results, and newly discovered allergic reactions.
重点内容
  • The definition of big data varies widely among industry experts
  • Beth Israel Deaconess Medical Center (BIDMC), an academic health care institution affiliated with Harvard University, is a typical user of big data for clinical care, education, and research
  • electronic health record (EHR) data are typically stored in two ways: first, organized by patient in clinical systems for care coordination; and second, aggregated in analytic databases for purposes of population health, quality measurement, and care management
  • Privacy can best be protected while mining big data by keeping all patient-identified data inside a secure health care data center and responding to external queries from payers, government agencies, and public health departments with aggregate numbers
  • There are challenges related to data quality, variability in capture over time, inconsistent use of terminology, privacy, and meaningful interpretation of the data
  • The future is bright, but excitement must be tempered with the possibility that more data with more analysis will only augment the noise resulting from biases that plague the scientific literature, given the availability of more data to work with and more investigators testing more hypotheses.[18]
结果
  • The statewide quality data center meets many health care operational goals, but the clinical researchers at BIDMC needed something that better supported comparative effectiveness re-
  • The building and establishment of the repository allows any authorized user who has completed BIDMC institutional training in the requirements of the Health Insurance Portability and Accountability Act (HIPAA) to run real-time population queries in support of clinical research.
  • Privacy can best be protected while mining big data by keeping all patient-identified data inside a secure health care data center and responding to external queries from payers, government agencies, and public health departments with aggregate numbers.
  • Gathering data from the patient in his or her home supports diagnosis and treatment between encounters with the health care system, enabling the ACO goals of disease prevention and wellness.[13]
  • Health services research, and clinical trials can all be enhanced by using patient-generated data to report observations that were unavailable in provider-centric records.
  • All of the approaches used at BIDMC—decision support tools, the centralization of quality data, distributed queries for clinical research, unstructured data analysis for case management, and patient-generated data—create their own new challenges and issues.
  • They amplify the challenges that have long faced data users: issues of data quality, variation in data collection over time, the inconsistent use of standard medical terminology, patient privacy concerns, and the need for expert data navigators to create meaningful queries.
  • BIDMC recently hired a third-party firm to examine the capabilities of its EHRs and health information exchange in support of quality measurement, and the frequency of data capture and the reliability of data elements.
结论
  • Variations In Data Collection Over Time BIDMC has used EHRs since 1985 to store problem lists, medication lists, and diagnostic test results.
  • Big data in health care holds great promise for empowering clinicians, health services researchers, quality measurement experts, clinical trials investigators, and care managers.
  • Success can be achieved by embracing the new possibilities that big data brings, while avoiding the mythology that bigger data means better data. ▪
研究对象与分析
clinicians: 1800
This requires new data resources that combine observations about the patient into a single, continuous, lifetime view of his or her health care experience. There are over 1,800 clinicians in the Beth Israel Deaconess Care Organization (BIDCO), one of the thirty-two Pioneer ACOs. The Pioneer ACOs are an advanced model of ACOs that was launched by the Center for Medicare and Medicaid Innovation

patients: 3
With this repository, we would be able to identify a cohort of patients with breast cancer at BIDMC who were taking ACE inhibitors. An analysis using Clinical Query yields 2,421 Æ 3 patients. Why Æ3 patients? To ensure that no query identifies a single individual, we never report the exact number

BIDMC patients: 80000
Many other novel explorations are possible. For example, 80,000 BIDMC patients had ischemic heart disease and no history of Vioxx use, while 800 patients had ischemic heart disease and took Vioxx, an anti-inflammatory drug that was withdrawn from the market in 2004 because of concerns about increased risk of heart attack and stroke with long-term, high-dosage use. Clinical Query could help investigators explore the temporal relationship between the introduction of Vioxx and the frequency of ischemic heart disease

引用论文
  • 1 Bourne PE. What Big Data means to me. J Am Med Inform Assoc. 2014;21(2):194.
    Google ScholarLocate open access versionFindings
  • 2 American College of Radiology. 2013 practice guidelines and technical standards [Internet]. Reston (VA): ACR; [cited 2014 May 27]. Available from: http://www.acr.org/QualitySafety/Standards-Guidelines
    Findings
  • 3 Centers for Medicare and Medicaid Services. Medicare and Medicaid EHR Incentive Program: meaningful use stage 1 requirements overview [Internet]. Baltimore (MD): CMS; 2010 [cited 2014 May 27]. Available from: http://www.cms.gov/ Regulations-and-Guidance/ Legislation/EHRIncentive Programs/downloads/MU_ Stage1_ReqOverview.pdf
    Findings
  • 4 Punke H. How did individual Pioneer ACOs fare in their first year? Becker’s Hospital Review [serial on the Internet]. 2013 Jul 13 [cited 2014 May 27]. Available from: http://www.beckershospitalreview.com/accountable-care-organizations/how-did-individual-pioneer-acosfair-in-their-first-year.html
    Locate open access versionFindings
  • 5 Centers for Medicare and Medicaid Services. Stage 2 overview tipsheet [Internet]. Baltimore (MD): CMS; [last updated 2012 Aug; cited 2014 May 27]. Available from: http://www.cms.gov/Regulations-andGuidance/Legislation/EHRIncentive Programs/Downloads/Stage2 Overview_Tipsheet.pdf
    Findings
  • 6 Executive Office of Health and Human Services. The Massachusetts Health Information Highway (The HIway) [Internet]. Cambridge (MA): EOHHS; c2014 [cited 2014 May 27]. Available from: http://www.mass.gov/eohhs/gov/commissions-andinitiatives/masshiway/
    Findings
  • 7 Weber GM, Murphy SN, McMurry AJ, Macfadden D, Nigrin DJ, Churchill S, et al. The Shared Health Research Information Network (SHRINE): a prototype federated query tool for clinical data repositories. J Am Med Inform Assoc. 2009;16(5):624–30.
    Google ScholarLocate open access versionFindings
您的评分 :
0

 

标签
评论
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn
小科