Truly meaningful chronic disease management relies on the confluence of electronic health records, big data analytics, and a new consumer favorite: the Internet of Things.
As the nation’s medical system sheds its own reactive, sick-care skin and reinvents itself as a proactive, data-driven, team-based ecosystem of wellness and health maintenance, there is a strong consensus among provider organizations that clinicians and patients need a host of new tools to meet a slew of unfamiliar challenges.
With millions of patients living with heart disease, diabetes, and hypertension – and millions more on the cusp of developing these life-altering conditions – the development of intuitive, impactful, and coordinated chronic disease management programs has become a hot ticket for the vendor community.
From consumer tech behemoths like Apple and Google to big data analytics companies to EHR vendors to the tiniest start-ups operating out of home offices and coffee shops, everyone seems to be offering the latest and greatest addition to the growing Internet of Things: a breakthrough patient engagement device to revolutionize behavioral choices, seamlessly track and integrate big data into the electronic health record, and keep patients focused on staying out of the hospital and getting on with their lives.
Unfortunately, a cursory look at many of these technologies shows that healthcare’s Internet of Things hasn’t yet achieved all the goals it has set out to conquer.
Providers are still struggling to carve out the basic health data interoperability pathways required to collect, aggregate, and analyze big data, let alone deliver actionable insights to end users without producing more problems than positive impacts.
Patients lack not only the knowledge of how to best manage their diabetes or hypertension, but the incentive to stick with gimmicky app tutorials or repeat the often cumbersome process of logging into a patient portal to update their personal health information.
How can providers, developers, consumers, and data experts work together to create a meaningful, engaging, and supportive chronic disease management environment that leverages electronic health records and the Internet of Things to maximum effect?
Laying the ground work with big data analytics
When it comes to the daily workflow, healthcare stakeholders may be split on the true benefits of the electronic health record. However, it is difficult to dispute the assertion that EHRs have significantly boosted the industry’s ability to examine vast stores of patient records to uncover new relationships and help manage populations on a broad scale.
Researchers from UCLA, for example, recently shared their work on helping to identify undiagnosed Type 2 diabetics using machine learning and EHR data to flag patients unwittingly living with the disease.
A previous study from the CDC and National Institutes of Health (NIH) found that up to 36 percent of Americans living with the metabolism disease are undiagnosed and unmanaged, with patients of Hispanic and Asian heritage most likely to be unaware of the problem.
The UCLA team, led by Ariana Anderson and Mark Cohen from the university’s Semel Institute for Neuroscience and Human Behavior, developed a pre-screening tool that draws information from a patient’s entire EHR file, resulting in a risk stratification tool that is 2.5 percent more accurate at identifying diabetics than the standard blood-pressure-and-BMI approach.
The new methodology uses ICD-9 and ICD-10 codes to predict the likelihood of an individual having diabetes, and examines such novel risk factors as previous viral infections such as chicken pox or chlamydia, intestinal infections, and sexual or gender identity issues, all of which correlate strongly with patients who also have a Type 2 diabetes diagnosis.
While the algorithms used in the study are in need of refinement to accurately deal with ICD codes that may cover more than one specific risk fact, "the overall message is that ordinary record keeping that doctors do is a very, very rich source of information," said Cohen. "If you use a computerized approach to studying patterns in that data, you can greatly improve diagnosis and medical care."
The team found that if the big data analytics method was used nationwide, it could potentially identify 400,000 patients who would otherwise miss out on critical chronic disease management opportunities.
"Given that 1 in 4 people with diabetes don't know they have the disease," Anderson said, "it's very important to be able to say, 'This person has all these other diagnoses, so we're a little bit more confident that she is likely to have diabetes. We need to be sure to give her the formal laboratory test, even if she's asymptomatic.'"
"With widespread implementation, these discoveries have the potential to dramatically decrease the number of undetected cases of Type 2 diabetes, prevent complications from the disease and save lives," she added.
Developing the “stickiness” of chronic disease management
Once identified, chronic disease patients must be brought into compliance with their treatment programs, which isn’t always easy. In a 2015 survey by Accenture, patients expressed widespread frustration about a lack of communication and coordination from their providers, stating that the pre-treatment period was one of the most stressful aspects of receiving a new diagnosis.
Patients also shared a strong desire for more comprehensive education, better predictive analytics that would warn them of risks before developing a full-blown disease, and easier access to information about resources and programs that will help them manage their conditions.
Patients still rely on their providers as a highly-trusted source of health information, but they are increasingly turning to the online resources, and virtual coaches that come along with their smartwatches and smartphone apps for help managing their diets or exercise plans.
But these tools may not encourage successful, sustained disease management, says Ben Jonash, Principal at Doblin, the design and innovation arm of Deloitte LLP – at least not without the help of providers themselves.
“A patient may have gotten a Fitbit or Apple Watch as a Christmas present, but if you talk to the vast majority of them six months later, how many are using these tools as an integrated way to change their health?”
“The challenge isn’t necessarily the front-end of a patient experience,” he told HealthITAnalytics.com. “It’s not about enticing them to play with these gadgets. It’s about creating the right habits and reinforcing the message and linking these strategies back to what matter. It’s about making sure you’re not completely dependent on the patient to self-manage the situation, because most people just can’t do that.”
A study from the Journal of Medical Internet Research backs up this assertion. While 58 percent of surveyed patients said they had downloaded an mHealth app to track some aspect of their health, a disappointing number of them had discontinued use of the tool after being turned off by hidden fees, poor usability, or fatigue.
More than forty percent of patients who had stopped using an app complained about the time involved in collecting and inputting their data, and a similar number said that they just lost interest in maintaining their participation over time, especially when the tools failed to provide specific, personalized, and actionable advice, motivation, or strategies for achieving their health goals.
Only a few patient engagement offerings really crack the code for sustained patient engagement in the Internet of Things era, says Jonash.
“There are some companies that have done a good job of combining the tracking of data and the use of it. They’ve taken several steps to make their digital therapy programs sticky. One of the ways they can do this is by adding cohorts to the program. Each user gets assigned to a small group of people who are collectively responsible for achieving a group goal. That produces a little bit of shared accountability and gamification.”
But relying on a patient’s innate competitiveness is not enough, he added. Multiple touch-points are required to draw users in and keep them there.
“It’s important to have some built-in nudging from a virtual coach or actual provider that’s built into the platform,” Jonash asserted. “That creates another set of conversations. The patient knows that someone is looking at the information they are collecting, and it gives them the opportunity to ask about what it means.”
“It doesn’t need to be an hour-long conversation, but these quick moments can imbed some added stickiness to the program.”
Healthcare providers may not be having these conversations often enough, due in part to the fact that they don’t always have access to the patient-generated health data (PGHD) stored in mHealth apps and devices.
Most electronic health records have thus far done a poor job of integrating emerging IoT technologies into the provider workflow, contributing to a growing disconnect between what patients think their providers know about them and what is actually available in the consult room.
Carolinas HealthCare is one of the organizations trying to bridge this gap by developing its own set of apps and trackers that directly integrates PGHD into the routine care process.
"It's one thing to collect all your steps on your phone," said Greg Weidner, MD, Medical Director of the Carolinas HealthCare System Proactive Health Group. "It's quite another to provide the clinical explanation and connection to the provider.”
With the release of the MyCarolinas Tracker app, the health system hopes to be able to solve the problem of bringing disparate streams of IoT data into a unified record, stored separately from the EHR but still available for decision-making with the patient’s permission.
“It’s an evolution,” said Craig Richardville, Chief Information Officer at Carolinas HealthCare to mHealthIntelligence.com. “This is just one small part of the whole 360-degree view of the patient. This adds value … to the patient’s health and wellness or care management program. It creates engagement and accountability.”
Patients are getting fit, but are providers getting paid?
These data integration efforts may bring benefits to chronic disease patients, but there’s still a major missing piece as far as providers are concerned: the ability to be reimbursed for the time and effort they spend on combing through a patient’s blood pressure fluctuations or scrolling through heart rate graphs.
Internet of Things device makers may be raking in the profits by selling flashy gadgets to a consumer market eager to boast about owning the latest and greatest toys, but the reimbursement landscape is far less secure for providers, many of whom have only taken a few tentative steps towards value-based payments.
“Fitbit gets reimbursed by selling a widget, but we’re still figuring out how to pay providers for a better health outcome on the back end,” Jonash said. “Once you shift accountability to the providers like that, they are going to start doing different things to make sure patients stay on track and stay engaged.”
The accountable care organization has been one of the more successful frameworks for making sure providers are getting their money’s worth for their chronic disease management efforts, but it isn’t yet a foolproof way of raking in revenue.
For one thing, chronic disease patients are among the biggest spenders in a healthcare system that is still largely dependent on fee-for-service payments. Shared savings accrued by participants in initiatives like the Medicare Shared Savings Program (MSSP) barely scratch the surface of what it takes to care for a single complex patient each year.
Data from the Agency for Healthcare Research and Quality (AHRQ) shows that the top five percent of chronic disease patients are responsible for nearly 30 percent of all spending in the healthcare system. The top one percent of patients with four or more chronic diseases can require close to $100,000 worth of services every twelve months.
In 2014, all of the MSSP and Pioneer ACOs combined produced just $411 million in savings, and $422 million in shared savings returned to qualifying participants. Clearly, there is a major opportunity for providers to equal the financial playing field, and they can start to do so by ensuring that patients with lower-level chronic disease problems don’t turn into complex cases.
CMS has already acknowledged that providers must have a financial incentive for performing advanced care coordination and population health management activities, and has been providing organizations with a $42.60 per-patient reimbursement for treating Medicare patients with two or more chronic diseases according to an approved care plan.
The program, which started in January of 2015, immediately galvanized healthcare organizations. Ninety-two percent of providers participating in a poll about the offering said they plan to take advantage of the extra cash.
Care management bonuses are nothing new, and the “per-member, per-month” payment strategy has seen some success with driving the adoption of innovative patient management frameworks, such as the patient-centered medical home.
But providers may not be all too interested in pouring their time, energy, and technology budgets into developing an Internet of Things mentality until digitally-driven chronic disease management is a bit more lucrative.
For that to happen, the healthcare industry must continue to concurrently pursue multiple major reforms. By making big data analytics and PGHD integration a top priority, they can conjure up meaningful patient engagement strategies that make provider-sided population health management investment an attractive – and properly reimbursed – proposition.
None of these strategies can exist without the others, and that may be why consumer enthusiasm for the IoT is outpacing its current clinical value. However, the healthcare community has never wanted anything but the best for its patients, and may soon find that embracing innovative, data-driven chronic disease management programs is imperative for the delivery of high quality, proactive, and effective patient care.