As the remarkable, eventful year of 2017 lumbers to its cold and snowy close, healthcare organizations may feel as if they are starting 2018 on an unstable footing.
From tragic natural disasters and devastating ransomware attacks to the first year of MACRA and last-minute drama as a Republican Congress manages to scrape through a repeal of the ACA’s individual mandate, the industry has certainly seen its fair share of shakeups over the past twelve months.
The shape of consumer interactions with healthcare providers is changing, too, as a series of shrewd, strategic alliances bring retail clinics, pharmacies, providers, payers, and technology companies ever closer together.
Data-driven deals like the CVS acquisition of Aetna, Optum’s takeovers of the Advisory Board and DaVita Medical Group, and Humana’s $800 million purchase of Kindred at Home are creating new networks for patients and providers to navigate – and new treasure troves of big data to make that process easier and much more personalized.
The closer integration of healthcare’s parallel tracks may be just what the industry needs to withstand the tremors of change, which are bound to keep getting stronger as the calendar rolls over again.
As rumors grow that lifestyle experience companies like Apple, Google, and Amazon are eyeing the clinical sphere, providers must start rising to the challenges of value-based care, an increasingly educated and cost-conscious consumer base, and a growing expectation that healthcare engagements harness the intuitive and intelligent capabilities of everyday technologies.
Many are starting to make progress in this area, says Thomas Laur, CEO of SAP Health, albeit somewhat slowly.
“We are finally starting to establish a data-driven ecosystem in healthcare,” he told HealthITAnalytics.com.
“But we are still very early in the process of leveraging data to its full potential. It still takes so much time, energy, and capital to put all of the necessary data together and curate it.”
Too many organizations are still clinging to the traditional fee-for-service methodologies of the past – but often because of the basic necessity of keeping the lights on, he swiftly points out, and not because they are willful luddites.
“In the context of a very cash-strained organization, most of the investment is going towards hardware that will help them survive in the fee-for-service world they’re living in, and there is little to spare for making the transition to value-based reimbursement,” said Laur.
“But we are coming to the point where the current economics will just not work anymore. Medical costs aren’t slowing down, and chronic diseases are rising. The platform we’re standing on is just getting smaller and smaller, and pretty soon it won’t support us anymore.”
Organizations will be forced to make the leap to value-based, consumer-centered care once their historical foundations really start to crumble, and that may happen sooner rather than later.
Despite a slight slackening in healthcare spending rates due to the successes of the Affordable Care Act, costs are still slated to increase by an average of 5.8 percent a year until 2025 – 1.3 percent faster than the nation’s GDP.
Within the next ten years, full fee-for-service will simply become untenable, and organizations that have not embraced big data analytics as the lifeline to guide them to more stable ground will start to wish they had planned their transition a little sooner.
“We are coming to the point where the current economics will just not work anymore.”
Now may be their window of opportunity, says Niyum Gandhi, Executive Vice President and Chief Population Health Officer at New York’s Mount Sinai Health System.
“They say you always overestimate the amount of change that happens in two years, and underestimate the amount that will happen in ten,” he said. “Our progress will continue to be incremental, but it will be inexorable, and there will come a point when things will have changed so much that there is no going back.”
“Over the next one or two years, we must focus on making sure our big building blocks of data analytics and quality measurement are in place so that our models can continue to improve and optimize over time.”
“If we don’t do that now, we might find ourselves trying to build new financial structures and more consumer-friendly care techniques without the right foundations in place.”
Data, data everywhere…
The healthcare industry is certainly not short on the data it needs to get started, says Busy Burr, VP and Head of Health Care Trend and Innovation at Humana.
The majority of organizations have petabytes of the stuff, with more pouring in from EHRs, medical devices, claims, patients, apps, and other IT systems every single day.
But despite the emerging trend towards payer, provider, and consumer integration and the data transparency that tends to come with it, most organizational data assets still remain dormant in isolated repositories or aging legacy systems.
“Over the next few years, we absolutely have to figure out how to connect our disparate data sources so that we can truly take advantage of the information we’re already generating, not to mention all the new data sources we need to start harnessing,” Burr stressed.
“Eventually, some of the analytics work will be completed by machine learning and artificial intelligence. But as an industry, we are relatively far away from bringing that to scale because we still lack so much of what we need to make those exciting innovations work well for us.”
Growing interest in artificial intelligence, and a flurry of successful use cases and pilots, may make it appear as if every other organization in the country now has automated access to all the data they need to stamp out all their inefficiencies and save millions in wasteful spending.
But self-conscious providers still struggling with the basics of defining what they want to improve, let alone how they can use their data to accomplish it, should feel (slightly) comforted by the fact that they may be the norm, not the exception.
The fundamentals are still extremely challenging for nearly everyone, says Larry Burnett, RN, Principal with KPMG’s Healthcare Solutions practice, and many organizations are still working through the first stages of preparing to make data-driven decisions.
“Most institutions are able to use data to identify mortality rates or their current average length of stay,” he said. “But as soon as you get past that very first level, analytics immediately become a lot more challenging.”
“You might understand that your length of stay is longer than the average for your type of organization, but you might not be able to figure out why it’s that long. Is it related to a specific physician? A nursing unit? Is there a group of DRGs that are resulting in longer stays?”
In order to identify and eventually address those issues, organizations need to start harnessing their unstructured data, Burnett explained, and that can quickly become a complex task.
“As soon as you get past that very first level, analytics immediately become a lot more challenging.”
“It’s not such a simple process to figure out that your length of stay is two or three days longer than it should be because you couldn’t get a patient scheduled for a stress test over the weekend, so they stayed those two extra days before they could have the test done on Monday,” he said.
“That’s not only a financial issue for the organization, but it likely also has a negative impact on patient satisfaction.”
“You need more than basic descriptive patient data from the EHR to discover those operational patterns. And it’s not that easy to address those issues, because it’s something that the patient’s physician has little control over. It will require a concerted operational effort, which means time and resources.”
Most organizations are still working in the world of descriptive analytics, Burnett said, and have not yet progressed to prediction or prescription.
“On the whole, we’re not at a point where we can predict readmissions based on a patient’s presenting conditions, let alone control those factors to prevent the readmission,” he said.
“As an industry, we are nowhere near being sophisticated enough to make that happen reliably and deploy those analytics at scale. It will take some time before we get there.”
The patient-centered promise of the Internet of Things
Opening up existing data siloes is just one part of the solution for this less-than-rosy portrait of the industry’s analytics progress.
New data sources, including patient-generated health information and data from the Internet of Things (IoT), might also make it easier for organizations to make the most of the relatively restricted EHR and claims data they currently have.
“Health happens outside the doctor’s office, so we need data from outside the doctor’s office, too,” said Burr. “Chronic disease management is about what the patient does when they’re out with friends, or in their homes, or at their jobs. That’s when the decisions happen, and those choices are not really captured very well in clams data or the EHR.”
“Once we get access to that data, it will create an explosion of brand new analytics models and new data assets that will allow us to develop brand new platforms to help manage the health of individuals.”
“I believe that over the next three years or so, we will start to be able to combine those data sources with traditional data sources to create extraordinarily valuable new insights.”
The Internet of Things is an extremely promising connection point between consumers and providers, agrees Laur, and its continuing expansion will soon start to impact the way healthcare organizations start to do business.
“Over the next three years or so, we will start to be able to combine those data sources with traditional data sources to create extraordinarily valuable new insights.”
“The number of sensors we are starting to carry around or use in our houses is growing very quickly,” he said. “I have a device on my wrist that tells me my heartbeat. It reminds me when it’s time to get up and move. It tells me how I slept that night, and gives me some basic tips on how to get better rest.”
“I am managing my health outside of the actual healthcare system in a way I can’t do by visiting my doctor once a year. Healthcare is becoming what happens between doctor’s visits, not what happens in the clinic. The IoT and its data are redefining the contours of healthcare.”
As consumer devices and new attitudes start to reshape what constitutes “care,” strong yet flexible provider-patient relationships are becoming more central to reimbursement, says Burr.
“As providers start to have more information about what’s happening with patients in real-time, they are able to have much more meaningful conversations supported by real-life context. A continuous stream of data can help providers create really effective, personalized interventions,” she said.
“Traditionally, that has been outside the scope of our analytics capabilities. But the proliferation of smart, connected devices is changing the paradigm and giving us access to the data we really need to make proactive, individualized recommendations. That is going to produce a major change in how we care for patients.”
This approach, supported by the ongoing development of technology in the consumer sphere, will start to bring healthcare into alignment with the way other sectors have started to woo, engage, and guide their consumers towards idealized outcomes.
“The challenge for traditional healthcare organization business models will be to embrace direct-to-consumer devices to extend their conversations with patients,” Laur predicted.
“The proliferation of smart, connected devices is changing the paradigm and giving us access to the data we need to make proactive, individualized recommendations.”
“There are going to be more and more IoT options coming to consumers, and they’re only going to get more sophisticated.”
“We have to learn, right now, how to manage that information, make sure patients are educated, and create mechanisms that will allow providers to intervene when something doesn’t look quite right with the data.”
Letting big data lead the way to personalized value
Personalized interventions, whether informed by clinical data, the Internet of Things, or retail-style consumer engagement profiling, are becoming a requirement of success in the value-based reimbursement ecosystem.
Payers aren’t the only ones who are expecting to get more value out of their interactions with providers: patients are increasingly demanding more bang for their bucks – especially when those bucks are coming out of their own pockets much more frequently than before.
A recent survey from the University of Utah Health found that patients have high baseline expectations for quality of care: few were likely to be willing to pay any more than usual for care that is timely, courteous, and positive for their health.
Source: University of Utah Health
Instead, they assume that providers will have access to all of their data, give preference to organizations with online appointment scheduling, and are likely to leave negative reviews on social media for providers that fail to come up to scratch.
Meeting those expectations is a very high bar for many providers, especially those still stuck in mired in a fee-for-service survival mode with little luxury to explore the world of big data analytics and personalization technologies.
But providers are starting to find help from a surprising ally in the battle to win and retain patient loyalty: their payers.
“The value-based care model is changing the traditional dynamic between providers and payers,” Burr asserted. “We all know that the fee-for-service world encouraged a little more defensiveness, because the financial incentives for the two players were in opposition.”
“It’s important to realize that value-based care brings providers and payers onto the same side – the patient’s side – and allows them to work in partnership for the benefit of the consumer.”
Payers have access to an important set of engagement and monitoring tools that individual providers can't always leverage, such as comprehensive pharmaceutical data, explained Burr.
While a patient may forget to tell her cardiologist that her rheumatologist prescribed a specific drug, the payer has access to all of the patient’s pharmaceutical claims from every one of her providers.
That can help payers identify potential problem areas with medication reconciliation or ensure that patients understand their regimens after a hospital discharge.
“At Humana, we’re growing an initiative that takes a look at the member’s entire regimen, including over-the-counter medications,” said Burr.
“We incorporate knowledge from our pharmacists as well as clinical data and claims data so we can understand what can happen, for example, when someone is discharged from the hospital.”
A hospital discharge tends to be a high-risk event for patients, she said. Not only is the individual coping with the aftermath of an acute event, but she is likely to leave the inpatient setting with new medications, discontinuations of previous drugs, or changes to her dosages.
“Those first few days after discharge can be difficult for members, especially when they’re adjusting to a new routine,” Burr said. “There is a risk of negative interactions or falling away from the regimen.”
“We have created a very rich layer of analytics over that data that helps us serve up a highly personalized experience involving both a care manager and a pharmacist who connect with that member in the first days after discharge.”
The care manager usually makes a home visit, she added, and works closely with the member’s pharmacist and primary care provider to address any emerging issues or patient safety concerns with the drug regimen.
“It’s the combination of analytics and the high-touch, human centered experience that makes this initiative work,” said Burr.
“We’ve been somewhat slow to realign the industry in general, but as more and more of these models take root, they will produce big changes in the way we conduct our relationships.”
Using data analytics to put the individual in the driver’s seat
Providers and payers don’t just have to collaborate with each other, added Laur. They both need to let patients take the lead when it comes to engagement and value.
“Much of our push towards patient engagement has been centered on the idea of letting patients access the records that providers have created,” he said. “We have been trying very hard to push our data to the patient, with less than stellar results.”
Patient portal adoption among providers has been high, thanks to the EHR Incentive Programs, but the number of consumers who access their data through these tools has remained consistently lackluster.
Although close to 90 percent of providers have implemented the technology, just 15 percent of hospital patients and 30 percent of ambulatory patients have actually used them, the Government Accountability Office says.
Patients find the interfaces confusing, are frustrated by slow responses from medical staff, get annoyed at automated emails, and can’t understand the medical jargon, said a 2016 survey.
“We can’t keep focusing on pushing data to patients,” said Laur. “It has to be the other way around.”
“Patients aren’t getting pulled along by the health system nearly as much as they used to. They are driving their own care with their own out-of-pocket funds, and much of what they are doing is happening outside of what we consider traditional channels.”
Patients may want their data and their interactions with the healthcare system to happen on their terms, but that doesn’t mean that payers and providers should remain completely passive while they wait for consumers to come to them.
Providers and payers have an obligation to ensure that consumers are informed about costs and spending, that they understand the need for routine preventive care, and that they hold up their end of the healthcare bargain by completing recommended protocols.
“We can’t keep focusing on pushing data to patients. It has to be the other way around.”
Balancing the right amount of consumer education and messaging with patients’ personal preferences and lifestyle choices is no easy feat, even for heavyweight players like Humana, Burr noted.
Humana has a robust member engagement program that relies heavily on digital tools and analytics that tailor engagement methods to the individual, she said.
“It’s easier to have a meaningful relationship with a person when you are communicating them in their preferred format.”
“For some people, face-to-face is still the best way to have a conversation. Some like phone calls. Some want video conferences, or emails, or text messages. Having more digital options and a way to track preferences allows us to be sensitive to these desires.”
“But you have to be careful that you don’t turn having more options into creating more complicated pathways to getting the information consumers need, or more communications than they desire,” she cautioned.
“You don’t want to overwhelm them with the emails and the video chats and the text messages…especially if they really want a phone call instead.”
The key to creating an impactful, consumer-focused environment is to create personalized experiences, not just more experiences, Burr said.
“You have to be careful that you don’t turn having more options into creating more complicated pathways to getting the information consumers need.”
“Organizations have to be very careful that they aren’t getting overly enthusiastic about having these tools at their fingertips. They have to align their outreach to what their patients or members want and need.”
“Analytics are supposed to help organizations be more careful about what messaging they deliver and how they deliver it. Ultimately, you are trying to influence behavior and create better human interactions. There’s no point in getting excited about all the data if you’re not leveraging it for that core purpose.”
The same goes for physicians and other care providers, agrees KPMG’s Burnett.
“If you want to effect change in an organization, you need to start with getting the right data and positioning your conversations in a positive, collaborative light,” he said.
“There is a natural inclination for physicians to resist being told they have to treat their patients differently than they have been, whether for reasons of cost, quality, or patient satisfaction. The only way to get past that is to use good, clean, accurate, risk-adjusted data to present a story of their variability.”
“Here is where you’re similar; here’s where you’re different – and more importantly, here is where your differences make sense and where they might need to be adjusted.”
Organizational leaders should stress that eliminating variability for variability’s sake isn’t the goal of showing clinicians their performance data, Burnett said.
“Physicians always need to have flexibility when treating their patients. Cookie-cutter medicine simply doesn’t work, and it’s not what these professionals and experts were trained to do. We don’t want to take away their ability to apply clinical judgement, because that’s what they get paid for.”
“The problem is when there are radically different practice patterns that come with increased costs and no improvement in outcomes. Then they need to take a look at what they could change to improve – and it turns out that most providers are really excited to be able to do that.”
Whether addressing providers or patients, using tailored messaging backed by big data analytics is a competency organizations must develop in order to make the switch to value-based care.
“I’m always reminded of the quote from Paul Batalden that says, ‘every system is perfectly designed to get the results it gets,’” said Gandhi.
“Our system is a sick-care system, and we’re designed to get the mediocre health outcomes that we get in America.”
“We need a system-level change that will move us towards population health, because that’s the outcome we want. And it’s going to require that literally everything is done differently.”
Enacting consumer-centered changes across the entire industry will take time, commitment, and a little bit of persuasion, he added, throughout 2018 and well beyond.
“You can’t mandate value-based care or population health or even consumer behaviors from the top down,” he said.
“We have 38,000 employees at Mount Sinai and 6,500 physicians. We have 3.1 million outpatient visits a year. It’s just not possible to cram everyone into the same mold by executive degree. And even if it was possible, it would be the wrong way to do it.”
“Success will come from expanding the mindset of our providers and staff. As they make their day-to-day decisions that involve trade-offs and optimizations and workflow planning, they will orient those decisions around patient-centered preventive care rather than around sick care. That is what real, lasting change will look like. That is how we get value out of what we do.”
This article was originally published on December 21, 2017.