Data Accessibility and Analytics Enables Data-Driven Emergency Care

Thirty percent of the entire world’s data is generated in the healthcare industry, with valuable information often locked in the EMR. Many times, time-consuming manual processes are required to abstract the data and transform learnings into a usable format for clinicians and administrators to understand performance, drive operations, and inform improvement plans.

对于奥兰多卫生，业务领导有效运行急诊(ED)业务所需的数据不容易获取。获取医疗系统的ED性能数据消耗了分析师近90%的FTE，对使用及时数据驱动和改善ED操作的能力产生了负面影响，限制了分析师执行更有意义工作的能力。通过利用其分析平台，奥兰多健康领导层扩大了对数据的访问世界杯厄瓜多尔vs塞内加尔波胆预测和可见性，以推动改进工作。通过获得接近实时的可操作数据，奥兰多健康能够以及时、有意义的方式规划ED改善工作。

DATA ACCESSIBILITY LACKING IN EMERGENCY CARE

In the U.S., over 130 million people access the emergency department (ED) each year for care. As a result of increasing patient volumes, EDs must optimize resources to provide care—requiring organizations to measure and improve access to timely care, improve emergency preparedness, and reduce costs of care.¹

Of the world’s data, thirty percent is created in the healthcare industry. These data are of tremendous value but are often locked in the EMR, requiring laborious manual processes to abstract the data and transform it into a format that clinicians and administrators can use to understand current performance, drive operations, and inform improvement plans.²

Orlando Health is one of Florida’s most comprehensive private, not-for-profit healthcare networks. Based in Orlando, its seven acute care hospitals serve nearly two million Central Florida residents and 10,000 international visitors, with over 98,600 hospital admissions, 380,000 emergency visits, and 5,200 trauma cases annually. The healthcare network sought a way to improve data accessibility to help advance improvement efforts.

MULTIPLE DATA SOURCES REDUCE IMPROVEMENT BENEFITS

Orlando Health has consistently sought to use data to support improvement, leveraging lean concepts to enhance outcomes and reduce cost. However, data were contained within multiple sources, making it difficult to obtain meaningful information that could be used to drive operations and inform improvement efforts.

The data required by operational leaders to effectively run ED operations were not easily accessible and developing meaningful reports required analysts to manually abstract data from multiple sources. Varying data definitions across several sources resulted in discrepancies, causing clinicians to doubt the validity of data.

Access was limited to specific users who had the knowledge and skills required to abstract data, limiting its use to drive decisions across the entire organization. Non-analyst users were limited in the type of data they could obtain, and reports were not stable. Users could attempt to gather data for an entire month or several months, but the return of that data was inconsistent. The requested information may be received a couple of hours after submitting the query, although typically the report would time out and never return the data of interest.

领导和改进专家联系了分析师，后者会手动提取所需的数据，复制并粘贴到电子表格中，然后手动创建可视化，以便其他人可以使用和理解这些信息。分析师每周要花好几天时间获取数据，这限制了他们进行更有意义工作的能力。

时间限制、有限的数据可见性和文件大小限制对改进工作产生负面影响。延迟，分析趋势和比较时间的能力有限，或分析地点和供应商之间的差异，限制了组织使用数据来驱动和改善ED运营的能力。Orlando Health领导层需要一个复杂的分析解决方案，以提高数据的可见性和访问能力，推动改进工作。

DATA ACCESS DRIVES EMERGENCY CARE IMPROVEMENTS

To improve data accessibility and visibility, Orlando Health invests in the Health Catalyst^®Data Operating System (DOS™) platform, and a robust suite of analytics accelerators, including the Department Explorer: Emergency Services Analytics Accelerator to provide sophisticated analytics, engendering data-driven operations. The analytics accelerator pulls data from the analytics platform into a user-friendly visualization, which displays the ED data necessary to track, trend, and predict resources required to deliver high-quality emergency care. These visualizations are easy to understand and support operational leaders in using data to drive operations (see Figure 1).

The analytics accelerator serves as the organizational source of truth and is used to understand ED performance. To eliminate data discrepancies and improve clinician confidence, standard definitions within the analytics accelerator define each metric and the sources of data for each, including arrival times, door to provider, door to discharge, and door to disposition, improving confidence in the data and willingness to use the data to drive improvements. Near real-time information is accessible to operational leaders, clinicians, quality improvement specialists, and analysts, who can drill down and make comparisons over time.

Armed with visibility into timely actionable data, emergency services clinicians can analyze ED volumes by day of the week, month of the year, and hour of the day. Orlando Health can identify sub-cycles for overall length of stay, as well as track, trend, and predict resource utilization. The organization can also analyze the impact of ancillary services on department throughput and quality. Accountabilities are clearly assigned for each process that patients encounter.

Using the analytics accelerator, operational leaders no longer wait weeks or months for reports, instead making data-driven continuous improvement a part of daily work. Leaders use the analytics accelerator to review the previous day volumes and throughput measures, enabling better planning for the day-to-day operations and are better able to adjust to meet patient needs. Data is integrated into the ED’s visual management system and used as part of daily quality huddles. Teams receive immediate feedback regarding performance, with the data illuminating where they have achieved the organizational goal and where they still have room for improvement (see Figure 2).

Orlando Health engages frontline clinicians in its improvement work. The analytics accelerator is readily available for access on most devices in the ED. End-users participate in a 20-minute personal training session, ensuring they understand how the analytics accelerator works and have the skills to use data as part of their routine activities.

The improvement team is now able to use the analytics accelerator to drill into performance data during team meetings and can develop an action plan in under 30 minutes, leading to more nimble responsiveness to improvement efforts. Teams can also immediately evaluate the impact of improvement plan interventions on outcomes. In one instance, Orlando Health piloted a trial of a provider in triage to reduce wait times. The organization questioned which type of provider would be best suited to fill that role, so it completed two pilots with different provider types. Incorporating data within the analytic application at weekly meetings, the benefits of the staffing change and the role that leads to the greatest improvement became clear. Use of a provider in triage is now standard practice across the system.

RESULTS

通过获得接近实时的可操作数据，奥兰多健康能够以更及时、更有意义的方式规划ED改善工作。The healthcare network has experienced a substantial reduction in time required to obtain performance data and visualizations, including a:

• 90 percent reduction in time analysts spend collecting, analyzing, and visualizing data, freeing up availability for more meaningful improvement activities.
• 访问和使用分析应用程序的不同用户数量增加了20倍，增加了数据透明度，并进一步加速了数据驱动决策的使用。
• Increased efficiency, timeliness, and accuracy of data has improved morale for ED clinicians, resulting in higher engagement in improvement efforts.

对数据的更好访问还为奥兰多健康提供了对其性能的新见解，使领导者能够更好地优先考虑改进工作并增强日常运营。组织可以提出和回答比以前更复杂的问题，消除了提出问题和回答问题之间的延迟。

“We spent a lot of time looking for areas to improve but without a clear source of data to guide our improvement efforts projects were selected based on opinion rather than actionable data. We now use data to drive our operations, helping us to take care of our patients in a timely manner without delays.”

– Carlos Carrasco, MBA
Vice President of Operations Orlando Health
Orlando Regional Medical Center

WHAT’S NEXT

奥兰多卫生应急服务团队已经对手术进行了大量数据驱动的调整，以更好地适应不断变化的患者数量。它计划进一步改善数据和分析的使用，以促进数据驱动的改进，利用整个医疗系统来协作和提高ED吞吐量。

REFERENCES

1. 舒尔，J. D.，夏瑞妍，Burstin, H.， Schull, M. J.， & Pines, J. M.(2013)。Quality measurement in the emergency department: Past and future.Health Affairs, 32(12): 2129-2138.
2. Huesch, M. D. & Mosher, T. J. (2017).Using it or losing it? The case for data scientists inside health care.NEJM Catalyst.