Have you ever wondered what insights are hiding in your lab’s data? The key to unlocking these gems of knowledge is building institutional capacity for data analytics. Clinical laboratories in their daily operations generate mountains of data that describe not just orders and results, but also myriad aspects of hospital and laboratory operations.

Unfortunately, many laboratories do not have the infrastructure to readily access and analyze this data. The scientific session on Tuesday, December 15 at 2:00 pm Central, “Essential Elements to Build Capacity for Lab Analytics” delves into the steps required to build such an infrastructure and apply these tools in laboratory medicine practice.

Joseph Rudolf, MD, views data and analytics as important motivators, as data are what first drew him to the field of lab medicine. Importantly, Rudolf states, data analytics are most powerful when implemented as proactive and longitudinal initiatives, providing opportunities for increasing lab quality and efficiency.

Availability of meaningful data often makes the difference between a confident, informed decision, and a difficult, uneasy decision. Rudolf hopes that after this session, attendees will view building capacity and expertise in all facets of data analytics as sound long-term investments in lab quality.

Speaker Patrick Mathias, MD, PhD, first found medical informatics during medical school and residency training, where he saw the impressive potential of machine learning for healthcare data, but also encountered many of the practical barriers to obtaining, curating, cleaning, and transforming data. Mathias explains that the “laboratory arguably generates the highest proportion of structured observations of any field of medicine,” and yet most labs devote a very small portion of their budgets to analytics.

With the right tools, laboratories could maximize their existing data and enable facile investigations into clinical and operational issues. Databases are a fundamental element of healthcare information systems, so Mathias shares key concepts about databases, as well as available tools for interacting with databases.

Presenter Amrom Obstfeld, MD, PhD, describes the positive impact that even simple data analyses can have. For example, knowledge of total sign-out volumes per year could make the difference in opening another pathologist position. With the amount of data generated by labs, clinicians and upper level administrators often expect that labs have access to key operational, quality, and efficiency analyses.

Obstfeld argues that laboratory information systems and electronic health records form the starting blocks for gaining these insights, and that lab leaders should focus on building new capabilities out of these systems. Obstfeld’s presentation aims to take some of the mystery out of data analytics to increase accessibility for a broader lab audience. He believes that a lot can be accomplished with existing resources and a little bit of additional knowledge.

Speaker Shannon Haymond, PhD, notes that healthcare leaders increasingly recognize the value of data analytic infrastructure, and that laboratorians can pioneer its creation.

Haymond’s presentation focuses on advanced analytics, including machine learning, and how these tools can be applied to healthcare data. Importantly, she draws a distinction between measuring a model’s performance and measuring its impact. Haymond explains the phases of data analytic maturity as well as strategies to overcome challenges in implementing data analytic capacity.

With an understanding of the importance of data analytics, laboratorians can make the case to build analytics capacity and pave the way to even higher quality laboratories.