The brightest leaders in healthcare are enjoying improved quality and outcomes with effective digital health programs.
The need for access to clean data and analytics when implementing a remote patient monitoring strategy is pressing, as I outlined in my last post. Typically, when I explain the "crude oil vs refinery" analogy, people of all expertise levels understand why it makes more sense to access patient generated data from the refinery.
There's a sense of discomfort and rapid change in health care right now, particularly given the rapid evolution of payment and reimbursement mechanisms. CMS wants to tie 50% of Medicare payments to alternative payment models by 2018. The need to innovate and interoperate is pressing.
According to a report by the American Medical Association, failures in care delivery and over treatment are wasting $320 Billion dollars per year in the United States. The industry will likely never get this number to zero, but a dose of the right data and analytics will enable better decision making.
Let's look at a few ways actionable mHealth insights prove to be valuable for wellness and healthcare companies beyond reducing the IT spend.
Improved efficacy of population health programs
There is a major opportunity in population health for data and analytics to move programs towards improved efficacy and efficiency. Evaluating ROI per member is typically the primary way success is measured; however, these measures tell an administrator very little about how the population's health changed, and can create deceiving results about a program's actual effects on the population.
Not only do programs often measure success based on cost per member ROI, but many of these programs are very labor intensive and inefficient. By now, the industry has widely adopted connection of fitness trackers and in some cases home health devices to enable the flow of data into a system. However, few are able to use this data in a way enables better decisions.
A recent success story from Houston Methodist highlights what is possible when organizations embrace data and analytics to move the needle on population health issues. The system ran an analytics driven six-month employee population health pilot which significantly reduced visits to the emergency room and hospitalizations.
The health system was able to cut uncontrolled diabetes and hypertension rates in half with a six month program. Analytics played a key role in the program's success: according to the piece from HealthLeaders Media, "[due to proper use of analytics], one nursing care manager can manage 200 to 230 patients."
Image credit: HealthLeaders Media and Houston Methodist
The numbers don't lie. When remote patient data is properly gathered and analyzed, organizations can use it to their advantage to improve results, and given the ability to analyze results quickly, they can use what they've learned to make improvements rapidly.
Patient and member engagement
Next, let's look at a health system with a goal of engaging population segments associated with the top chronic conditions and comorbidities. Launching a program that keeps those patients healthy and out of the hospital is likely already a high priority.
With coherent access to segments and trends on patient activity, heart rate, exercise, sleep, and body metrics, you could easily create a set of rules and associated triggers for intervention when patient anomalies are detected. Instead of the patient coming in, a care manager could reach out to engage the patient and potentially re-direct the patient to applicable services or care, avoiding unnecessary readmission.
Reena Pande, MD is the Chief Medical Officer at Abilto and a cardiologist at Brigham and Women's Hospital. Dr. Pande says one of the most important aspects around engagement is reducing barriers to quality care.
“They may not wait a week or more to get into their PCP’s office or an emergency department. They may be more apt to use the new mechanism and that might make access to care more readily available,” Pande said in an interview with mHealthIntelligence.com.
Clinical decision support
One of the areas that excites me the most about patient generated data is the ability for that data to inform clinical decision making.
I imagine a world where continuous actionable insights flow from my body to the healthcare providers in my life. When I go in for an annual checkup my doctor should have a quick view that illustrates various aspects of my life, such as how active I am, what my blood pressure has been (vs. a point in time), and how much sleep I've been getting.
The idea is not to overload the provider with more data points, but to more quickly enable a recommended course of action. Perhaps my data coupled with qualitative analysis shows the provider that I don't need to come see him again for two years, ultimately saving everyone in the value chain time and money. Alternatively, if an anomaly is detected that is cause for alarm, the provider could get recommended treatment options based on my lifestyle and medical history.
To be fair, the data being collected by consumer wearables today has a long way to go before we can leverage it broadly for clinical decision support. There is a lack of trust on the part of providers, and device manufacturers are not held to the same standards that FDA medical devices are.
However, the mechanisms to get mHealth data flowing in an actionable manner need to be integrated now. The rate of technological advancement is not slowing down. Devices will continue to gain accuracy, battery life will improve, and patients will demand that their data be a part of their care continuum. While we shouldn't categorize a Fitbit as a clinical-grade medical device today, we can begin to leverage data trends and insights as a piece of the puzzle.
Mark Olschesky, Chief Data Officer at Catalyze, wrote a great piece recently that highlights the importance of using data to support decision making. In the blog post, Mark outlines why being "data-enabled" is a more realistic goal than being "data-driven," because in most cases treatment decisions should not be completely automated.
"A data-enabled organization delivers providers information and suggestions on how to better manage its population health portfolio or community to produce an improved, more consistent quality of care," says Olschesky.
The bar is higher than ever
On the population health side, employers and vendors alike struggle to prove that their programs actually work. Measuring efficacy of a program purely by per-member ROI is an outdated and misguided approach, and often results in programs being discontinued simply because they were deemed ineffective.
For health systems, the mandate for EHRs has perceivably not driven the desired quality, and health systems must begin to invest in data and insights in a meaningful way.
If the industry is truly going to be transformed anytime soon, it must take advantage of the ever-present and growing ability to take action on patient generated data.
If any these challenges resonate with you, I would love to chat about how you're thinking about overcoming them. Feel free to send me an email to steve -at- straphq.com, DM me on our slack community, or take the conversation to Twitter by following @getstrap and @stevecaldwell.
Let's Put mHealth Data to Work
At Houston Methodist, Population Health is the Real Deal. HealthLeaders Media, May 3, 2016.
How the Telehealth Field Strengthens Patient Engagement. mHealthIntelligence, October 9, 2015.
Data enabled Does Not Equal Data Driven. Catalyze blog, Mark Olschesky. April 26, 2016. http://content.catalyze.io/blog/data-enabled-does-not-equal-data-driven
Eliminating Waste In US Healthcare. Berwick DM et al. https://www.icsi.org/_asset/y74drr/BerwickWedges2012.pdf
Thanks to Ben Marston, Rohit Nayak, Tom Hills, and Marco Della Torre for reading drafts and providing feedback.
Co-Founder / CEO