CIOs already understand that, in a value-based marketplace, they need analytics to improve clinical outcomes and maintain financial sustainability. What they may not know is that they can make major strides in cost savings, medical error prevention, risk prediction, and population health management without investing in an enterprise-wide analytics platform.
On its own, the medical laboratory is a goldmine of actionable information that can drive these financial and performance metrics. Given that lab data is structured, timely and can be used to track patient "care-flow," they represent arguably the most powerful analytic input for intelligent, value-based care delivery. Once integrated, the lab is a powerful and cost-effective tool for cutting costs and improving care—a new central nervous system for healthcare organizations.
Driving clinical decision-making and current trends in medicine
Currently, even some of the top health systems rely on their EHR to measure performance and determine the appropriate clinical decision-making support. But the reach of the EHR is actually narrower than the lab's, which performs and records more touch points with patients than any other department or unit in healthcare. Driving 70% of all clinical decisions, the lab's testing extends across settings (from home to the emergency room, from clinics to hospitals); across providers (from primary care physicians to specialists); and across time (from patient test histories to real-time). In fact, its wide span puts laboratory data at the heart of two very different trends: personalized medicine and population health (see Figure 1).
Figure 1. Lab Data as a Nexus for Health System Value Initiatives
How can medical laboratory data increase clinical value?
Lab data—and importantly, lab data as the primary input—can provide the basis for four key value initiatives:
1. Clinical decision support
Lab data analytics platforms can use current and historical lab data representing patient diagnosis, monitoring, and outcome to determine the most effective treatment pathways for different diseases, optimal testing protocols, and the boundaries of appropriate utilization1. They can also alert clinicians (via the EHR) to departures from these protocols and pathways, thus identifying care gaps and helping minimize certain categories of medical errors. These tools are especially useful for the high-stakes care of chronically ill patients.2
2. Predictive analytics tools
Predictive analytics transforms a vast trove of information into actionable insights that change performance for the better. Predicting a patient's likelihood of readmission, his or her length of stay, and the costs for an episode of care (or for a cohort) helps clinicians and hospitals plan strategically and proactively. Lab data analytics also have an advantage over other approaches in enabling disease-specific solutions, which tend to be more easily implemented. Further, as investment increases and so does the data to be analyzed, these tools are only becoming more powerful.3 Currently, predictive analytics platforms are using lab data to predict a patient's risk of a specific disease progression, including profiling for chronic kidney disease (risk of kidney failure) and the risk of cardiovascular events.4
3. Population health
In the US, the market for tools to promote population health management is booming: a Deloitte report put the American healthcare analytics market at over $2 billion, with a projected annual growth rate of 8% to 11% through 2020.5 One of the highest reported priorities for these organizations was population health. Smaller and mid-sector health systems should know that they can make significant strides in population health management with lab results and data. Lab medicine is integral to prevention and wellness screening, chronic disease management, and identifying potential adverse events and care gaps. For example, lab test analytics can be used to identify missed diabetics or to proactively identify disorders earlier, such as chronic kidney disease. Furthermore, in terms of care coordination, because lab testing connects settings, providers, and visits, the lab is a natural source for accurately tracking care and outcomes. Finally, lab values are uniquely able to integrate the coming waves of data from point-of-care and home-testing—trends that must be accounted for, and tests that must be integrated, if analytics are going to capture the full picture of disease progression and prevention among patient groups.
4. Personalized medicine
The lab is the heart of molecular testing, companion diagnostics, and genomic sequencing: the advances that are animating personalized medicine. This new era has just begun, with advanced analytics integrating the suite of lab testing now being performed on cancer patients, for example, to synthesize therapeutic recommendations.
Results from the front lines
What is possible when lab data is used to drive decision-making? One exciting development is a recently validated risk predictor for all-cause mortality that requires only lab test variables.6 As hospital leaders and front-line clinicians will recognize, the ability to predict "mortality and long-term outcomes for specific health disorders"—rather than a general patient cohort —would provide much more actionable results, including patient placement, resource allocation, and monitoring. This predictive instrument "exhibited a robust association with concurrent chronic conditions, recent hospital utilization, and current health status as assessed by self-rated health."7
What it takes to activate this new "central nervous system"
There are three reasons why CIOs must take a lab-centered analytics approach to fully exploit the value benefits to their healthcare system:
1. The lab incorporates outpatient, inpatient, and multi-provider testing for a patient, whereas an EHR or EMR provides only a partial view. An EMR may only include inpatient or outpatient results, and only from the provider who ordered those tests. These limitations preclude implementing a number of the analytics described above.
2. Of all the departments in a health system, pathology and laboratory medicine comprise a range of testing domains that can best leverage clinical value. Because labs typically maintain separate databases (e.g., chemistry, microbiology, anatomic pathology, molecular testing), those will need to be integrated to unlock that value. Also, anatomic pathology frequently utilizes narrative reporting, which may require natural language processing to extract discrete data for analytics.
3. An integrated, lab-centered analytics strategy enables the data-mining that generates lab-based predictive analytics; health system population applications such as infectious disease management, blood management, wellness, and provider utilization management; and provider-based comparisons by clinician, practice, specialty, service line and institution.
Conclusion
It is understood that data analytics are vital for unlocking improvements to health system operations and care. Health system CIOs can look to their existing assets in the medical laboratory to achieve those insights and improvements. A lab-centered analytics strategy delivers much-needed structure and context, transforming big data into actionable, insightful, smart data.
1 Jennifer Musumeci. Clinical Informatics in the Laboratory: The New Value Generator. G2 Intelligence. Kennedy Information: Keene, NH: 2014, 35.
2 Musumeci, 39.
3 http://www2.deloitte.com/us/en/pages/life-sciences-and-health-care/articles/health-system-analytics.html.
4 See for instance, Tangri N. et al. JAMA. 2016;315(2):164-174.
5 http://www2.deloitte.com/us/en/pages/life-sciences-and-health-care/articles/health-system-analytics.html.
6 Bello et al. Development and Validation of a Clinical Risk-Assessment Tool Predictive of All-Cause Mortality. Bioinformatics and Biology Insights 2015:9(S3) 1–10.
7 Bello et al.
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.