America is waking up and realizing it has a diabetes problem — a massive problem — both in terms of quality of life for those who suffer the diseases, and associated monetary impact (one of every five healthcare dollars in the U.S. is spent on patients with diabetes).
Diabetes is actually a collective term for different diseases with different pathophysiology, prevalence, and solution sets. Broadly, type 1 diabetics do not produce insulin either at all or in a satisfactory amount whereas type 2 diabetics produce insulin, but it does not work. In other words, this latter group has insulin insensitivity — where the body has resistance to insulin leading to high bloodsugar — which is a continuum not a discrete moment. Those referred to as having “pre-diabetes” are on the type 2 diabetes path, where insulin production works, but its effectiveness varies.
Each set of patients with the three diseases requires different types of engagement, different types of technology, and different types of monitoring to manage. The varying nature of the diseases presents serious challenges to scaling solutions, but that does not mitigate the need for interdisciplinary work to solve problems related to each condition.
More specifically, a collective focus is required to address:
- The need for better temporal resolution of data that provides actionable intervention, management, and prevention
- The business model challenge that inhibits viable scalability of current innovations
- The soaring numbers of affected individuals and associated complexity, specifically for Type 2 and pre-diabetes, which necessitates an evolved role for the physician.
Some promising progress has been seen in the market for type 1 diabetes, which can be a life-threatening condition. Here, the relationship between glucose and insulin is well understood and the first focus requirement—better temporal resolution of actionable data—is a matter of optimization. Some solutions are coming to market, including technology like the Verily-Dexcom partnership to shrink and economize continuous blood monitoring devices for type 1 diabetics; smart alert systems like Livongo to engage in the home before significant emergencies requiring hospitalizations arise; and new analytics enabled through continuous data streams that are analyzed and tie real-time blood glucose with appropriate and precise administration of insulin via pumps.
Each of these simplifies the management, lessen the requirement for oversight, and result in a more “normal” life for those with the disease. For the 1.25 million Americans suffering from type 1 diabetes, this is certainly a good thing.
However, current innovations do not go far not enough, nor do they translate to the broader population suffering from the “other diabetes.” To solve the challenge of 20 million+ with type 2 and 85 million+ with pre-diabetes, we need to be ruthless in identifying which of the advancements and innovations actually scale and in recognizing the limitations of those that do not. Technology for Type 2 diabetes and pre-diabetes has a significant behavioral and preventive component.
Rather than addressing treatment when someone is hospitalized, engagement needs to occur continuously. The innovation requirements for Type 2 diabetics, when it comes to temporal data that is actionable, are all about altering choice and shifting behavior—or getting you to eat a salad instead of a burger and fries. It sounds simple, but this is not easy stuff. What we have learned from the Type 1 cohort and sub-segments does not directly translate to this space.
Type 2 diabetes is an epidemic. And the AMA estimates that pre-diabetes affects one-third of the U.S. population. These patients need clinical support and social support, as well as algorithmic and behavioral technology support. The good news is the use cases for impact are countless, but the bad new is that finding the points that represent scalable and real opportunities for improved treatment is complicated.
The diabetic population is a microcosm of our society. Incidence is growing at 1.4 million per year, 26 percent are seniors, and 208,000 are under 20 years old. It is present across cultures and in every state. Pre-diabetes and Type 2 diabetes are both conditions that are largely products of environment and behavior, not solely genetics. Treating these individuals requires a complex campaign against a multifaceted and multivariable disease—and that is much more difficult to do without the right teams and tools, and without incentive. Incredibly, the Health and Human Services Department (HHS) has only just recommended a shift in Medicare rules to actually cover the first diabetes prevention, as opposed to treatment, program.
In the tools arena, the notion of shrinking current tech to monitor in the same invasive manner as with type 1 diabetes (where the device has to break the skin) is not a differentiator in Type 2 or pre-diabetes. However, making a real-time, non-invasive monitor that connects to individual decisions could change the game (e.g., an optic blood glucose monitoring device connected to a smartphone with push notifications that align along familiar consumer behavior).
Technology that obviates the need for regular in-office care and works well for the 18–35-year-old Millennial demographic could provide a 10x impact; further, technology that integrates data in a way that gives individuals more effective means to to alter behavior management and reverse disease progression can change the world. Welltok, which recently announced a collaboration with IBM to leverage cognitive computing for its CafeWell application, is an example of the types of broader partnerships aimed at tackling complex problems like behavioral change. However, more is needed.
Where advancing the treatment of Type 2 diabetes is very much about reversal and slowing progression, treating pre-diabetes actually presents an opportunity to recast the operation and long-term outlook for our entire healthcare system.
Facing the exploding health needs of some 86M people whose lives could be vastly improved and whose prognosis could be completely reversed demands a pivot in the the way our healthcare system currently works. We’ve seen rapid and massive shifts in consumer industries driven by technology and bold thinking, and we can translate that into the battle against diabetes.
Machine learning and artificial intelligence, wearables and real-time analytics, as well as integrated data from across the health system (payment, clinical, consumer) together unleash new abilities around the logistics of how we care and intervene at the right time and place. And while the technologies of tomorrow will have to work in collaboration with the systems and behaviors of today, it really is not an insurmountable challenge. If you can master a chronic disease segment such as pre-diabetes, the use cases and solutions market scale across a significant target population, and the lessons may apply to other chronic diseases across similar segments.
An innovation strategy to take on our diabetes problem as a whole requires an understanding of the disease (provided by physicians), insight into behavior (leveraging data and analytics technology and ubiquity of smartphones), and, arguably most importantly, understanding how the restructuring of incentives around service delivery will alter the way revenues flow in healthcare (e.g., advanced payment models and tools that reward preventative care over reactionary care).
Collaboration between product developers, care delivery and provider teams, and payment and governmental entities are all required to “cure” this disease, as is a healthy dose of individual responsibility. However, to enable this collaboration on the scale required, clear swim lanes for companies need to be established. The business models remain a mess with consumers unwilling to pay directly, a consolidated and limited customer base in delegated payers, and a complex and often misaligned two-step sale process into self-insured companies (where the primary goal of the product does not equate with the primary goal of end user). While we will hear complaints about the time it takes for a drug approval by the FDA, there is at least a measure of expectation management and clarity of path in that process. We need similar clarity as we shift to a value-based care world, where payments will happen on a continuum. Healthcare providers, regulators, and innovators must do better at removing the impediments to innovation, while maintaining the integrity of control and safety.
Innovations may come in small steps, from basic text message reminders to old-school peer-to-peer engagement and encouragement. Education, empowerment, and a sense of community are all part of the product design needed to solve this challenge, but it must be solved, and quickly. We should be using every tool available (and, indeed, dreaming up new ones) to defeat diabetes.