Overcoming Size-specific Barriers to Implementation
Healthcare organizations today are both blessed and cursed by data: to get the insights necessary for impacting patient outcomes and improving operational efficiencies, health systems need to unlock the data that is spread across multiple siloed and dispersed IT platforms. Enterprise data warehouses (EDWs), which pull together data from across an organization or health system, used to be the only way that health systems could aggregate, normalize, and store the data needed to drive their analytics programs.
Yet organizations that tried to build EDWs found their IT departments spending hours struggling with "dirty data"—data that was in the wrong shape for visualization and analytics. To make matters worse, each time the data changed or a new information source needed to be analyzed they had to redo their work. This ultimately led to an ugly cycle of resources being spent on data-plumbing rather than on what health systems really needed to be doing: analyzing, visualizing, and operationalizing the data.
Health systems have since learned that there is little value in spending IT resources on the data-plumbing problem, and they have started to look for new approaches to implement their analytics programs. Some are shifting to a more user-centric approach, focusing on tailored analytics solutions that support specific initiatives instead of a one-size-fits-all, IT-driven EDW. Especially for mid-sized hospitals and health systems with limited IT resources, fully managed analytics solutions are the more easily implemented and operated option that still yields the insights required for the mandates of improving care and enhancing efficiencies.
Reasons for Adoption
The idea of "value" permeates every aspect of healthcare, from the clinical to the operational to the financial. These aspects of health care are intertwined to a greater and greater degree; consider how Medicare and private payers alike increasingly tie reimbursements to value-based care or expected outcomes. In this context, analytics solutions are a critical tool for targeting inefficiency, inappropriate utilization, and waste.
Where EDWs Work Best
According to The Advisory Board, the enterprise data strategy works best for larger, more far-flung health system environments with disparate organizations under their umbrella. These heterogeneous systems will benefit most from the power and span of the EDW, but more importantly they can weather the approximately two years it takes to implement, build, and start seeing returns on the project.1 Mid-sized hospitals and small health systems, on the other hand, may not have the resources to wait that long for their ROI, nor the manpower to do the work without outsourcing or hiring new staff. For these organizations, a nimble solution can be installed in two days, fully implemented in six weeks, and informing the decision-making that shows returns in six months. The resulting platform can be made HIPAA secure (typically through the use of encryption), vendor agnostic, and accessible from anywhere in the hospital's network.
The Power of Lab Data
It's a commonplace that the laboratory has more points of contact with patients than any other healthcare department. What is less commonly known is that the laboratory was a pioneer in automating its workflow, including data structuring and digitization. Today up to 70% of the medical record is made up of laboratory data. Due to the critical nature of laboratory test results for diagnoses, this data is extremely potent and relevant in highlighting correlations pertaining to specific diseases, populations, or even physician performance.
Measuring performance, rooting out waste, and identifying areas for improvement are the basic functions of traditional analytics platforms. But predictive analytics is an advanced capability that more and more hospitals recognize as crucial to efforts to responsibly manage their patient population. In this realm, lab data can be an optimal foundation for predictions about disease progression and readmissions. This predictive capability is only becoming more important as performance-based reimbursement programs increase the share of their scoring tied to care episodes by patient condition groups like those with chronic kidney disease (CKD).
Examples of lab-based analytics capabilities:
1. Blood Management
The Joint Commission, the American Medical Association, the American Hospital Association, and other groups have moved to act on scientific evidence that hospitals blood transfusions are overutilized in modern medicine. Studies have demonstrated that these transfusions are associated with significant increases in hospital mortality and complications—yet in the United States, 30 to 60 percent of transfusions "are not indicated, not warranted, and not appropriate according to evidence-based transfusion guidance and best practice."2 Analytics platforms help hospitals get a handle on utilization by showing exactly who is using blood products for which procedures and in what department. Real-time access to this information can steer decision-makers directly to outliers (whether that's specific clinicians, a department, or a specific span of time), and drive an efficient, compliant use of blood products and transfusion services.
2. Utilization Management
Analytics are also yielding direct savings by reducing overutilization in the laboratory test environment, which costs the United States billions of dollars annually (including $5 billion in redundant tests alone).3 The real news, however, is that clinical analytics is addressing an even costlier, though less apparent, utilization problem: underutilization. One meta-analysis of inappropriate utilization in the lab determined that underutilization was more prevalent than underutilization and that it was the real culprit behind downstream overutilization.4 Analytics is not only helping the healthcare system determine the appropriate use of laboratory tests, but is also being used by clinical labs and their hospitals to steer clinicians to better utilization of testing through education and peer comparisons. These platforms also enable lab managers to make informed decisions about which tests should be done in-house and which should be sent out.
3. Outreach management
Outreach management is a way of leveraging in-house lab capabilities as a new source of revenue for the organization. This prospect is made possible due to hospitals' need to have a functioning laboratory in-house so that physicians of inpatients have access to critical tests on the premises. The capacity of a clinical laboratory is often greater than its institution's demands, though. Enterprising hospitals have come to realize that this unused capacity can be a huge revenue creator if they manage their "outreach" well—outsourcing their services to community providers and organizations. While the market opportunities are substantial, competition is fierce (specifically from major independent labs who depend on community provider business). From the earliest phases (assessing whether the lab has the capacity to do the outreach) to the active client management and performance tracking necessary to keep existing clients, analytics is vital. In addition, the data produced from outreach has significant strategic value for health systems that are looking to integrate care within the communities and have ongoing tracking of the patient.
4. Infectious Disease Management
What hospitals need, and what analytics helps them achieve, is the rapid identification of infectious diseases when they walk into the hospital, allowing physicians to intervene with individual patients before there is an outbreak. Further, immediate antibiograms enable institutions to see resistance patterns in real-time, and take the measures necessary to protect patients and staff. Advanced analytics can even alert staff if a pre-determined threshold has been met, which can lead to more effective containment. Much of this area coincides with the movement for better "antibiotic stewardship," an effort that has gained in urgency with the rise of antibiotic resistant bacteria, which the CDC calls "one of the most serious and growing threats to public health."5 Antibiotic resistance has arisen in part because of the inappropriate prescribing of these drugs, an overuse that the CDC estimates at between 20% and 50% of all such prescriptions.6 (Many of these prescriptions occur as a late, multi-pronged response to an infection that has become more difficult to treat.)
1 Adams, J. "Overview of the Health Care Analytics Market: BI Needs Outpace Vendor Tools and Health-Care Organizations' Capabilities." The Advisory Board Company. 2014.
2 http://www.ihi.org/communities/blogs/_layouts/15/ihi/community/blog/itemview.aspx?List=0f316db6-7f8a-430f-a63a-ed7602d1366a&ID=33.
3 Reduction in Unnecessary Testing.
4 Zhi et al. The Landscape of Inappropriate Laboratory Testing: A 15-Year Meta-Analysis. Plos One. November 2013;8(11). http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0078962.
5 https://www.cdc.gov/getsmart/healthcare/implementation/core-elements.html
6 https://www.cdc.gov/getsmart/healthcare/implementation/core-elements.html
The views, opinions and positions expressed within these guest posts are those of the author alone and do not represent those of Becker's Hospital Review/Becker's Healthcare. The accuracy, completeness and validity of any statements made within this article are not guaranteed. We accept no liability for any errors, omissions or representations. The copyright of this content belongs to the author and any liability with regards to infringement of intellectual property rights remains with them.