This content is part of the Essential Guide: Conference guide: News and analysis from HIMSS 2014
Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

Clinical analytics automation still requires the human touch

Providers are stacking third-party clinical analytics providers' services on top of EHR systems to get more insight into patient population data.

At HIMSS 2014, several hospital systems showcased examples of how they have been able to leverage clinical analytics and dramatically increase the value of all the data collected by electronic health records and other systems over the years. In many cases, these hospitals used third-party data analytics providers for assistance.

After the conference, it was time to sit and reflect on the vendors and the capabilities of the products they have introduced in 2014. EHR vendors were very eager to offer HIMSS attendees a look into their booths for a sneak peek of new modules and reports, and to confirm their compliance with certification requirements. Despite most vendors offering booth goodies and colorful handouts describing their products, it became difficult to say what could be taken away from these EHR products in terms of new capabilities and what these vendors' next move is in the marketplace.

On the exhibit hall floor, amid the smell of popcorn and other treats handed out at various booths, a few vendors took a different approach to seeking new business in the hospital market; for example, one had their product summarized by a friendly robot, reflecting their focus on artificial intelligence. These different approaches summed up what I was looking to learn about EHRs at HIMSS 2014. From one EHR system to the next, I got a sense of what vendors are doing that has changed and will continue to change the way we interact with clinical data.

Clinical analytics is becoming a big business. Companies small and large are able to use analytics to break through and get the attention of the chief medical information officers and chief information officers at many health organizations. Providers are interested in identifying and using systems that can provide more value by utilizing the data collected by EHR systems. Third-party clinical analytics providers increasingly offer more insight into EHR data. In doing so, these analytics companies are gaining ground on EHR systems and challenging them on their own turf.

EHR vendors face tremendous pressure to stay competitive with one another. They also must consider compliance regulations and other client demands when building or updating their products. These decisions drive EHR developers to continue to do what they do best and not make too many changes to their systems that are out of their core business model. Focusing on their strengths -- capturing, managing and exchanging electronic health data -- has opened the door for clinical data analytics providers to focus on thinking outside the database box to explore more uses for providers' data. Some call it big data and others label it meaningful use of data. Whatever the title, these analytics can impact the quality and costs of healthcare.

More on clinical
data analytics

Backups should be at the forefront of data analytics plan

Take this quiz to see how your knowledge of analytics measures up

Improved analytics and quality measures pave the road to accountable care

A new process of performing analytics on health records that is gaining popularity is one that taps into both structured and unstructured data. Numerous health records are held in unstructured documents, such as audio or transcription files. Clinical analytics tools must be able to take the unstructured data and analyze it, something Stuart Covit, chief operating officer and principal at Artificial Medical Intelligence, addressed.

This process does not stop there. Many of the smart analytics tools contain artificial intelligence, combined with machine learning, that is able to adapt and offer valuable insights into a patient's health. This cannot be left to a computer system alone to perform.

"Machine [learning] must be done with some human touch," Covit said. Most experts recommend involving humans in a process of checks and balances and not leaving all the work to computers. This is a valid point, because a machine can only learn what it's given -- as the old computer adage goes, if you put garbage in, you get garbage out.

Clinical artificial intelligence will always need human intervention. Companies such as IBM, Nuance and Artificial Medical Intelligence are partnering with hospitals to address the lack of strong analytics offered by some EHR products. These organizations have been able to optimize natural language processing and clinical decisions support tools, combined with data scientists, to offer solutions for better patient outcomes. Using all of these analytics tools gives hospitals a chance to seek a Level 8 placement atop the current analytics adoption model.

About the author:
Reda Chouffani is vice president of development with Biz Technology Solutions Inc., which provides software design, development and deployment services for the healthcare industry. Let us know what you think about the story; email or contact @SearchHealthIT  on Twitter.

Dig Deeper on Clinical data analytics software and systems

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.