The growing popularity of LIS modules such as Epic's Beaker inexorably leads to hospital/health system EHR personnel assuming greater responsibility for most of lab computing. This is in contrast to the past three or four decades when a small group of pathology-based informaticians and their staff exercised control over the LISs running in pathology departments. This turn of events prompted the following question in my mind: what professional activities will occupy the time and mind-share of the growing body of pathology informaticians now?
Fortunately, there are two large challenges that will now be assigned in part to pathology informaticians. First will be the need to automate anatomic pathology to the same level as clinical pathology, accompanied by the deployment and LIS integration of digital pathology. The second challenge will be the development and testing of the four types of analytics tools that are now coming to the forefront. Here is an excerpt from a recent note on this topic (see: Healthcare Will Contribute a Sizable Portion of Future Earnings for Apple):
...Apple has launched various projects that will provide the company with broad access to anonymized patient EHR records....The value of these records for Apple is that they can be the basis for the development of four types of analytics research: descriptive, diagnostic, predictive, and prescriptive (see: 4 types of data analytics to improve decision-making). Descriptive analytics provides information about what has already happened to a patient, diagnostic analytics provides the root cause of what has happened, predictive analytics predicts what will happen in the future, and prescriptive analytics goes beyond predictive analytics by specifying both the actions necessary to achieve predicted outcomes and the interrelated effects of each decision.
I have previously blogged on what I am now calling descriptive analytics. In these older notes I referred to this category as operational analytics (see: vRad Actively Pursues Operational and Clinical Radiology Analytics). Here is a quote from that note four years ago:
I like to divide lab analytics into two components that I refer to as clinical analytics and operational analytics. The former is patient-centric and addresses questions about testing such as the who, why, when. Operational analytics, as the name implies, addresses operational issues such as workflow, distributed testing, and cost-per-test. The interest in operational analytics is currently greater than clinical analytics with the task usually assigned to supervisors and lab managers rather than pathologists.
There is one overarching reason for pathology informaticians to occupy a large portion of their time with lab-oriented descriptive analytics. Hospital labs are being continuously tasked with the pursuit of value-based care, one of the central pillars of which is to provide higher quality healthcare at a lower price. Lab testing has been one of the greatest bargains in healthcare for decades due largely to increasing lab automation. Descriptive analytics provides a continuous view of key aspects of lab operations such as workflow, labor contribution, and automated instruments. Hence and with various analytics tools, pathology informaticians will be able to make continuous recommendations about how to wring waste out of lab operations and thus pursue even greater efficiency and effectiveness.
Comments