Free EMR Newsletter Want to receive the latest news on EMR, Meaningful Use, ARRA and Healthcare IT sent straight to your email? Join thousands of healthcare pros who subscribe to EMR and EHR for FREE!

The Value Of Pairing Machine Learning With EMRs

Posted on January 5, 2017 I Written By

Anne Zieger is veteran healthcare consultant and analyst with 20 years of industry experience. Zieger formerly served as editor-in-chief of FierceHealthcare.com and her commentaries have appeared in dozens of international business publications, including Forbes, Business Week and Information Week. She has also contributed content to hundreds of healthcare and health IT organizations, including several Fortune 500 companies. Contact her at @ziegerhealth on Twitter or visit her site at Zieger Healthcare.

According to Leonard D’Avolio, the healthcare industry has tools at its disposal, known variously as AI, big data, machine learning, data mining and cognitive computing, which can turn the EMR into a platform which supports next-gen value-based care.

Until we drop the fuzzy rhetoric around these tools – which have offered superior predictive performance for two decades, he notes – it’s unlikely we’ll generate full value from using them. But if we take a hard, cold look at the strengths and weaknesses of such approaches, we’ll get further, says D’Avolio, who wrote on this topic recently for The Health Care Blog.

D’Avolio, a PhD who serves as assistant professor at Harvard Medical School, is also CEO and co-founder of AI vendor Cyft, and clearly has a dog in this fight. Still, my instinct is that his points on the pros and cons of machine learning/AI/whatever are reasonable and add to the discussion of EMRs’ future.

According to D’Avolio, some of the benefits of machine learning technologies include:

  • The ability to consider many more data points than traditional risk scoring or rules-based models
  • The fact that machine learning-related approaches don’t require that data be properly formatted or standardized (a big deal given how varied such data inflows are these days)
  • The fact that if you combine machine learning with natural language processing, you can mine free text created by clinicians or case managers to predict which patients may need attention

On the flip side, he notes, this family of technologies comes with a major limitation as well. To date, he points out, such platforms have only been accessible to experts, as interfaces are typically designed for use by specially trained data scientists. As a result, the results of machine learning processes have traditionally been delivered as recommendations, rather than datasets or modules which can be shared around an organization.

While D’Avolio doesn’t say this himself, my guess is that the new world he heralds – in which machine learning, natural language processing and other cutting-edge technologies are common – won’t be arriving for quite some time.

Of course, for healthcare organizations with enough resources, the future is now, and cases like the predictive analytics efforts going on within Paris public hospitals and Geisinger Health System make the point nicely. Clearly, there’s much to be gained in performing advanced, liquidly-flowing analyses of EMR data and related resources. (Geisinger has already seen multiple benefits from its investments, though its data analytics rollout is relatively new.)

On the other hand, independent medical practices, smaller and rural hospitals and ancillary providers may not see much direct impact from these projects for quite a while. So while D’Avolio’s enthusiasm for marrying EMRs and machine learning makes sense, the game is just getting started.

Uncovering Hidden Hospital Resources: Hospital IQ Shows that the First Resource is Data

Posted on January 4, 2016 I Written By

Andy Oram is an editor at O'Reilly Media, a highly respected book publisher and technology information provider. An employee of the company since 1992, Andy currently specializes in open source, software engineering, and health IT, but his editorial output has ranged from a legal guide covering intellectual property to a graphic novel about teenage hackers. His articles have appeared often on EMR & EHR and other blogs in the health IT space. Andy also writes often for O'Reilly's Radar site (http://oreilly.com/) and other publications on policy issues related to the Internet and on trends affecting technical innovation and its effects on society. Print publications where his work has appeared include The Economist, Communications of the ACM, Copyright World, the Journal of Information Technology & Politics, Vanguardia Dossier, and Internet Law and Business. Conferences where he has presented talks include O'Reilly's Open Source Convention, FISL (Brazil), FOSDEM, and DebConf.

We’ve become accustomed to the results that data mining have produced in fields ranging from retail to archeology to climate change. According to Rich Krueger, CEO of HospitalIQ, the health care industry is also ripe with data that could be doing a lot more for strategic operations and care.

Many companies (perhaps even too many) already offer analytics for grouping patients into risk pools and predicting the most unwanted events, such as readmissions within 30 days of a hospital discharge. According to Krueger, Hospital IQ is the first company offering analytics to improve operations. From a policy standpoint, the service has three benefits:

  • Improving operations can significantly lower the cost of providing health care, and the savings can be passed on to payers and consumers.
  • Better operational efficiency means better outcomes. For instance, if the ER is staffed properly and the hospital has enough staff resources to meet each patient’s needs, they will get the right treatment faster. This means patients will recover sooner and spend less time in the hospital.
  • Hospital IQ’s service is an intriguing example of the creative use of clinical data, inspiring further exploration of data analysis in health care.

One of the essential problems addressed by Hospital IQ is unexpected peaks in demand, which strains emergency rooms, ICUs and inpatient units, and ultimately can contribute to poor patient outcomes. The mirror images of this problem are unexpected troughs in usage, where staff is underutilized and costing the healthcare system.

“There’s no reason for surprise,” says Krueger. While it is difficult to predict demand for any given day, it is possible to forecast upper and lower boundaries for demand in a given week. Hospital staff often have an intuitive sense of this. Krueger says with his software, they can mine clinical data to make these forecasts based on actual live data.

There are firms who make sensors that can tell you when a bed is occupied, when someone has come to clean the room or visit a patient, and so forth. The interesting thing about Hospital IQ is that it does its work without needing to install any special equipment or IT builds. Its inputs are the everyday bits of data that every hospital already collects and stores in their electronic record systems: when someone enters the ED, when they are seen, where they are transferred, when surgery started and ended. The more at-risk patients in telemetry beds generate information that is also used for analytics.

Hospital IQ uses mature business analytics tools such as queuing theory that have been popularized for decades since the age of W. Edwards Deming. Even though EHRs differ, Krueger has discovered that it’s not hard to retrieve the necessary data from them. “You tell us where the data is and we’ll get it.”

The service also provides dashboards designed for different types of staff, and tools for data analysis. For instance, a manager can quickly see how often patients are sent to the ICU not because they need ICU care but because all other beds are full. While this doesn’t compromise patient safety, it’s an unnecessary cost and also reduces patient satisfaction because ICUs generally have a more restricted visiting policy. This situation becomes more problematic when a patient arrives through the ED with a real need for an ICU bed and the unit is full (Figure 1). Hospitals can look at trends over time to optimize staffing and even out the load by scheduling elective interventions at non-peak times. There’s a potentially tremendous impact on patient safety, length of stay, and mortality.

solutions-patient-safety-and-quality
Figure 1: Chart showing placements into the ICU. “Misplaced here” are patients that don’t belong in the ICU, whereas “Misplaced elsewhere” are patients that should have gone to the ICU but were sent somewhere else such as the PACU.

Taken to another level of sophistication, analytics can be used for long-term planning. For instance, if you are increasing access to a service, hospitals can forecast the additional number of beds, operating rooms, and staff the service will need based on historic demand and projected growth.. Hospitals can set a policy such as a maximum wait of three hours for a bed and see the resources necessary to meet that goal. Figure 2 shows an example of a graph being tweaked to look at different possible futures.

solutions-capacity-planning
Figure 2: Simulation using historical demand data showing the relationship between bed counts and average wait time for a patient to get a bed.

Hospital IQ is a fairly young company whose customers cover a wide range of hospitals: from safety net hospitals to academic institutions, large and small. It also works across large systems with multiple institutions, aiding in such tasks as consolidating multiple hospitals’ services into one hospital.

My hope for Hospital IQ is that it will open up its system with an API that allows hospitals and third parties to design new services. It seems to offer new insights that were hidden before, and we can only guess where other contributors can take it as well.