When thinking about the business value of predictive analytics, hospitals often view it as an evolutionary technology and look for things like use cases, accuracy, cost, and return on investment. While those are all valid points to consider, a better way to look at it is to focus on what truly matters for patients and caregivers, work backwards, and explore how predictive analytics can make them better. Having worked with dozens of hospitals, I can tell you that — when viewed from this perspective — the answer is predictive analytics is a necessity, not a luxury.
Before I explain why, let me highlight some observations on products we love and use every day.
Millions of people use Uber every single day. Many can’t live without it. Why is that? Because they know Uber is the easiest and fastest way to get a ride: push a button, and your ride arrives in a few minutes. Behind the scenes, Uber looks at millions of data points and applies sophisticated data science to predict demand at a given location and ensure enough drivers are available to service that demand. But people don’t care about that; they care about getting a ride within a few minutes — Uber or not.
The important point here — and the reason why Uber became so successful — is that people always wanted to push a button and get a ride quickly. Uber just happened to be the first company that made it work. Uber as a company might be new, but Uber as a concept — a want — always existed. Then why didn’t someone else beat Uber to it? Because for the concept to come to life, a number of technologies, had to be in place: the cloud, GPS, mobile adoption, cheaper computing, affordable machine learning and AI. Ten years ago, the Uber concept simply couldn’t exist.
Every other company whose products we can’t live without started out the same way: Focus on what truly matters to their customers, and give them a radically better experience, with predictive analytics at its core.
Netflix started out renting DVDs because back in 1997, the internet wasn’t ready for video streaming. But it was still called Netflix because its founder, Reed Hastings, knew that people want to watch movies on the internet and that one day video streaming would enable that. That was inevitable, but what made Netflix do impossible things like killing Blockbuster and now threatening the TV networks and Hollywood itself? A relentless focus on personalization driven by predictive analytics. Today, Netflix knows more about how, why and when people watch content than anyone else.
Amazon, for decades, always believed that people want the largest selection possible at the lowest rates and fastest shipping. Amazon’s success is centered on using data and predictive analytics to boost those three metrics in a way no one else can replicate. Amazon predicts what you might like and recommends it. Amazon learns your purchase patterns, predicts what you might buy when and prefetches those products to the nearest warehouse. Amazon is now investing in drones to make the last mile even more efficient and deliver products within one hour.
We expect this kind of innovative, “out-of-the-box” thinking in high-tech, but what about healthcare? It’s complex and highly regulated, right?
Let’s look at airlines, an industry that is a lot more complicated and regulated than healthcare. Southwest Airlines, one of the most admired (and profitable) airlines in the world, understood that people want to get from point A to point B quickly and at low cost. They started out with two simple ideas: a single aircraft type (the Boeing 737) to simplify maintenance and operations and a single state (Texas) to simplify routing.
Today, Southwest flies millions of people to hundreds of destinations across the world. Yet, it is still low cost and has great on-time performance overall. Why? Because it made massive investments in predictive analytics to keep planes in the air and turn them around fast. Southwest’s centralized software keeps everyone involved in flight planning and operations informed with real-time data on takeoffs, landings and turnaround times. It predicts delays due to weather or other aircraft or airport problems, automatically figures out the best way to minimize impact, and takes corrective actions before problems spiral. While many airlines still rely on faxes, phone calls and log books, Southwest relies on predictive power, the web and mobile to give the right information to the right employee at the right time.
Take any company that has similar complexities that truly cares about what matters most to their customers — FedEx, UPS, etc. — they have turned to predictive analytics, mobile and technology in general as a necessity to improve efficiency.
If we look at these broader patterns, it is clear that increasing complexity is an existential threat. Airlines, retailers or any service organization for that matter simply cannot stay relevant by making incremental process improvements. They have no choice but to rely on predictive power and mobile to add step functions to their efficiency to improve quality and lower cost.
So, for hospitals, the question isn’t what predictive analytics can do. The question is what truly matters to their customers (patients and caregivers) and how to significantly improve them quickly. Fundamentally, hospitals are complex service organizations, so it boils down to better planning and better execution.
Let’s start with 3 key principles:
For decades, the de facto focus was on process improvement. But process improvements can only go so far. Can predictive analytics do better? The simple answer is: a lot better.
Operating rooms, a key resource at a hospital — bringing more than 60 percent of admissions and 65 percent of revenue — are a key example. Surgeons want an operating room whenever they want and wherever they want. Today, block scheduling is so complex and slow that we simply cannot give them that flexibility in a way that meets the objectives of all stakeholders involved. Block schedule changes are cumbersome, error prone and take months to effect. How can we make it better? By using predictive analytics and data science to dig deep into utilization patterns and by creating lightweight mobile experiences that let surgeons get the block time they need with a single click. We can make block scheduling as easy as Uber so they simply push a button and get the block they want. And if we do that, everyone wins — patients get treated faster, surgeons have better control and access, and the overall utilization (and revenue) increases. I have seen improvements of $500,000 to $1 million with this approach.
Specialty clinics like infusion centers are another example. Again, first principles: Patients want to be seen fast, and nurses and caregivers want better control of their schedules. We can make many process improvements to meet those goals, but can we do better with predictive analytics? A lot better. With predictive analytics, we can dig deep into historical appointment data and combine operational constraints to do a lot of number crunching and precisely calculate demand on a given day. We then can schedule patients in a way that flattens chair utilization. That significantly lowers wait times and leads to better schedules for nurses. Everyone wins.
Similarly, take any department at a hospital and ask what truly matters at that department? I’m willing to bet that compared to whatever plan you have in place to make them better, predictive analytics can offer a significant boost. Imaging equipment, ER, in-patient beds — effecting improvement in almost any unit at a hospital boils down to reducing variability and better planning. Predictive analytics can significantly help with both.
Read More: Healthcare Technologies | Analytics
There are a lot of parallels between hospitals and airports; they are there to make a journey better. Just like the air traffic control makes everything works like clockwork at airports, hospitals need an air traffic control-like functionality that predicts a patient’s journey and routes the right resource to the right patient at the right time. Such a system could fundamentally improve all the things that truly matter for a patient’s journey.
That may sound futuristic, but believe me, it’s not. We already have the core technological pieces we need to make it work — the cloud, sophisticated data science and machine learning, and mobile. We also don’t need complicated EHR integrations or months of planning and training and things that make your IT department nervous. We can do it using lightweight data extracts and well-designed experiences that are intuitive, easy and simple. In fact, it’s already happening — dozens of hospitals are already making it happen, just like Southwest, Netflix and Amazon.
Because ultimately, people want better care at a lower cost, and there’s no better way to achieve that than with predictive analytics.
Looking to reduce your patient onboarding time? Check out the award-winning case study from Mentis Neuro Health now:
As first published in Health IT Outcomes.
Sanjeev Agrawal is president and chief marketing officer of LeanTaaS iQueue. Sanjeev was Google's first head of product marketing. Since then, he has had leadership roles at three successful startups: CEO of Aloqa, a mobile push platform (acquired by Motorola); VP Product and Marketing at Tellme Networks (acquired by Microsoft); and as the founding CEO of Collegefeed (acquired by AfterCollege).
Sanjeev graduated Phi Beta Kappa with an EECS degree from MIT and also spent time at McKinsey & Company and Cisco Systems. Sanjeev is a Forbes contributor and also writes on his personal blog at
http://medium.com/@saagrawa. He is an avid squash player and has been named by Becker's Hospital Review as one of the top entrepreneurs innovating in Healthcare.