Mike Payne, MBA, MSci

According to the CDC, the majority of deaths in the United States are due to diseases that are largely preventable.

Chronic diseases, such as diabetes, certainly have genetic components, but their etiology is also profoundly behavioral—driven by choices that individual patients make on a day-by-day, minute-by-minute basis.

  A great deal of research has been done on maladaptive behaviors that people exhibit in the near term, even when they know that a different choice is best for the long term.

 This phenomenon is accentuated among people facing alternatives that involve an addictive choice like alcohol

 These choices often are informed by cognitive biases, notably attentional bias5 and evaluative bias.

  Richard Thaler and Cass Sunstein’s behavioral economics book 

is a broad romp through social policies that can influence a wide range of human behaviors to help people overcome cognitive biases.

 The general point of the book—that behavioral science can and should be used to guide the libertarian free choices of people to better outcomes in the face of cognitive biases&mdash;has profound implications for healthcare in a world in which an increasing proportion of health gains are being pursued through behaviorial change and maintenance.

  Progressive thinkers in healthcare are certainly on the case. Health and medical services that provide continuous, on-demand care, or “nudges” to behaviors associated with chronic disease, are proliferating, often facilitated by information technology and accompanied by published clinical results. There are simple care process improvement nudges to remind patients of immunizations, like Voxiva’s text service, which was shown to produce a 30% improvement in influenza vaccination rates among women.

In another example, more focused on direct delivery of “care nudges,” Virta Health’s Diabetes Reversal Clinic provides consistent personalized, electronically delivered nudges—the frequency of which vary&mdash;to individuals with type 2 diabetes (T2D) from Virta physicians and health coaches. Virta also provides personalized education to patients on a daily basis. Recent data show that the clinic achieved a 54% reversal of diabetes among its patients with T2D in just 70 days.

  Most people believe that technology-enabled nudges make up the care model they need to address chronic disease. But traditional payment schemes for healthcare are ill-equipped to properly reimburse for nudge-based care, as recognized by many, including a recent, high-profile panel at the J.P. Morgan Healthcare Conference.

 Traditional fee-for-service (FFS) models provide reimbursement for relatively long and in-person—or, as of late, telemedicine&mdash;visits. For example, in a model in which a Virta health coach contacts a patient 3 times a day with small nudges, it is administratively burdensome and overly costly to submit 3 insurance claims because of the $8-per-claim cost for such expenses as software/hardware, processing, and back-end collections.

 On average, Virta and its patients interact 13 times per week; under a FFS model, submitting claims at this rate would not be practical in this case.

  Furthermore, many rigorous health services for chronic conditions now involve contacting patients through intelligent technology rather than having a clinician manually execute all efforts involved in the interaction. For example, ginger.io, an online care service to combat depression and anxiety, uses machine learning to evaluate device sensor data and patient survey responses to support the provision of more personalized care.

 The sensor and survey data are often supplied many times per day. In a FFS world in which there are only codes for medical devices and human—clinician visits, there are few, if any, codes that exist for these frequent high-value patient touchpoints driven by technology rather than humans.  

 Like ginger.io, the Virta model is a particularly challenging one for a FFS environment, since it involves a broad variety of touchpoints with patients: clinician interaction, technology-personalized educational curriculum delivered via text and video, and measurement and tracking of patient biometrics. Virta touchpoints with patients also happen both real-time as well as asynchronously in an online chat-style interface, allowing the patient to engage whenever it is convenient for them. Applying a FFS claims approach to this care setting would lead to drastic underappreciation of the value provided by the Virta Diabetes Reversal Clinic.   These are just 2 examples, of many. Rock Health, a venture fund dedicated to digital health, notes that of 300 digital health funding deals executed in 2016, 15% were investments into nudge-based care companies.

  • Per nudge: reformulate coding rules to allow the creation of codes for nudges as outlined above and to reduce claims transaction costs to pennies. To use a music payment analogy, this is the “pay per song” model espoused for a decade by Apple.

  • Bundle of Nudges: providers get no reimbursement for individual nudges; they are paid a per-patient amount for a bundle of services to be provided, preferably with payment based on agreed-upon clinical outcomes achieved by the nudges, rather than just payment for the bundle of nudges itself. In effect, this is a value-based bundled payment model similar to the Comprehensive Care for Joint Replacement model tested by Medicare for knee and hip-replacement surgery—but applied to treatment for chronic care.

  The Bundle of Nudges approach is a clearly preferable model. It does not require massive overhaul of claims systems or FFS coding rules and it allows for further innovation and application of technology to the achievement of desired behavioral change and mod-ification. So what is standing in the way of a Bundle of Nudges payment from catching on? Three key hurdles must be overcome.  

 First, health insurance carrier leaders must embrace the idea that technology-enabled remote healthcare services can be as good as, or better than, in-person versions of similar programs, especially in disease states for which nudge-based approaches are considered standard-of-care. A recent example of a large payer ignoring such an undeniable trend can be seen in CMS’ decision not to reimburse for digitally delivered Diabetes Prevention Programs (DPPs) during year 1 of coverage under the new Medicare Diabetes Prevention Program (MDPP)

 despite the existence of quality data from DPP providers, such as Omada Health.

 This regrettable decision by CMS is a setback, but should not deter the proliferation of technology-enabled, nudge-based healthcare services. Insurers in private markets are leading the way by agreeing to work with such companies,

 and digital providers have organized under the Council for Diabetes Prevention

 to press Medicare to include them when MDPP reimbursement rules are finalized later this year.  

 Second, we must agree on the clinical outcome that the nudges are intended to achieve and for which providers of nudge-based care models—like Virta&mdash;will be reimbursed. In the area of diabetes, Virta’s model is to get paid if, and only if, specific HbA1c targets are achieved and/or maintained. Although this may seem like common sense, agreement on value-based clinical metrics is hard to achieve in healthcare, so visionary health plan leaders need to “put a stake in the ground” on what good management of chronic disease looks like from an outcomes perspective.

 Third, health insurance carriers must pave the way for speedy signing of contracts using Bundle of Nudges schemes. Currently, most health plans have a provider network contracting group that is responsible for signing FFS network contracts and a medical management group that is responsible for evaluating more innovative models, like bundled-payment models. The problem is, the medical management groups are often not empowered or incentivized to sign these bundled-payment contracts quickly since they are new and untested. However, most health plan senior executives still say that bundled and value-based payment models, like the Bundle of Nudges concept, are the wave of the future.

 This is a particular challenge since many nudge-based treatment approaches are being pioneered by small companies, like Virta Health, that do not have the financial luxury of waiting 5 years to be evaluated and go through the tedious process of securing a bundled payment contract from large health plans. This shorter-term view should not make insurance carriers skittish, however. In general, there are surrogate endpoints for chronic healthcare services, measured at 6-12 months, that can act as reliable statistical proxies for longer-term outcomes. For example, reduction in HbA1c in Virta’s clinic has been shown to be highly correlated to 12-month HbA1c and diabetes reversal outcomes and is thus a valid proxy.19 Insurance carriers need to make it easier for the best of these innovators to secure bundled contracts quickly.

 If we are to truly address chronic diseases that have become the largest medical scourge of today’s society, we need technology-enabled, nudge-based care to become prevalent. That, in turn, requires health insurance carriers to develop models beyond classic FFS schemes to reimburse bundles of nudges and to make the new bundled-payment schemes quickly available to the clinical innovators, like Virta, that are demonstrating the immense value and potential of nudge-based approaches.

 Mike Payne, MBA, MSci, is currently head of commercial and policy at Virta Health. He last served as chief commercial officer and head of medical affairs at Omada Health. Previously, he worked at Gilead Sciences, Accenture, and McKinsey & Company. 

  1. Up to 40 percent of annual deaths from each of 5 leading US causes are preventable [press release]. Atlanta, GA: CDC; May 1, 2015. https://www.cdc.gov/media/releases/2014/ p0501-preventable-deaths.html. Accessed January 17, 2017.

  2. Wiers RW, Gladwin TE, Hofmann W, Salemink E, Ridderinkhof KR. Cognitive bias modification and cognitive control training in addiction and related psychopathology. Clin Psychological Science. 2013;1(2):192-212.

  3. Hofmann W, Friese M. Impulses got the better of me: alcohol moderates the influence of implicit attitudes toward food cues on eating behavior. J Abnorm Psychol. 2008;117(2):420- 427. doi: 10.1037/0021-843X.117.2.420.

  4. Avena NM, Rada P, Hoebel B. Evidence for sugar addiction: behavioral and neurochemical effects of intermittent, excessive sugar intake. Neurosci Biobehav Rev. 2007;32(1):20-39. doi: 10.1016/j.neubiorev.2007.04.019.

  5. Field M, Cox WM. Attentional bias in addictive behaviors: a review of its development, causes, and consequences. Drug Alcohol Depend. 2008;97(1-2):1-20. doi: 10.1016/j.drugalcdep. 2008.03.030.

  6. Bradley B, Field M, Mogg K, De Houwer J. Attentional and evaluative biases for smoking cues in nicotine dependence: component processes of biases in visual orienting. Behav Pharmacol. 2004;15(1):29-36.

  7. Thaler RH, Sunstein CR. Nudge Improving Decisions about Health, Wealth, and Happiness. New York, NY: Penguin Books; 2009.

  8. Stockwell MS, Westhoff C, Kharbanda EO, et al. Influenza vaccine text message reminders for urban, low-income pregnant women: a randomized controlled trial. Am J Pub Health. 2014;104(S1): e7-e12. doi: 10.2105/AJPH.2013.301620.

  9. Results to be published in February 2017.

10. Preston J. Many healthcare technologies are married to the fee-for-service model. MedCityNews website. http://medcitynews.com/2017/01/many-healthcare-technologies-married- fee-service-model/. Published January 10, 2017, Accessed January 17, 2017.

  11. Thorman C. Should you outsource your medical billing? The Profitable Practice website. http://profitable-practice.softwareadvice.com/when-should-you-outsource-your-medicalbilling- 1032610/. Published March 6, 2010. Accessed January 17, 2017.

  12. Captain S. Can artificial intelligence help the mentally ill? Fast Company website. https://www.fastcompany.com/3059995/mind-and-machine/can-artificial-intelligence-help-the-mentally-ill. Published July 21, 2016. Accessed January 17, 2017.

  13. Digital health funding database, 2016. Rock Health website. https://rockhealth.com/data/funding-raw-data/. Accessed January 17, 2017.

  14. Centers for Medicare & Medicaid Services. Medicare Diabetes Prevention Program (MDPP) Expanded Model. CMS website. https://innovation.cms.gov/initiatives/medicare- diabetes-prevention-program/. Updated December 13, 2016. Accessed January 17, 2016.

  15. Sepah SC, Jiang L, Peters AL. Long-term outcomes of a web-based diabetes prevention program: 2-year results of a single-arm longitudinal study. J Med Internet Res. 2015;17(4):e92. doi: 10.2196/jmir.4052.

  16. AMA, Omada Health, Intermountain Healthcare partner to reduce incidence of type 2 diabetes [press release]. Chicago, IL; San Francisco, CA; and Salt Lake City, UT: Omada Health; July 26, 2016. https://www.omadahealth.com/news/ama-omada-health-intermountain-healthcare-partner-to-reduce-incidence-of-type-2-diabetes. Accessed January 17, 2017.

  17. Council for Diabetes Prevention launched to support delivery of CDC National Diabetes Preventino Program [press release]. Washington, DC: Marketwired; September 21, 2016.

http://www.marketwired.com/press-release/council-diabetes-prevention-launched-support-delivery-cdc-national-diabetes-2160355.htm. Accessed January 10, 2017.

  18. Beck M. Hospitals push bundled care as the billing plan of the future. The Wall Street Journal website. http://www.wsj.com/articles/hospitals-promote-bundled-payments-asthe- wave-of-the-future-1402067196. Published June 4, 2014, Accessed January 17, 2017.

Articles

Using digital health to manage chronic disease means creating new payment models that recognize care is an ongoing process.

A Bundle of Nudges: Healthcare Payment in an Era of Behavioral Science

In 1941, Australian pharmacologist and pathologist Howard Florey demonstrated—for the first time&mdash;an effective process for the mass production of penicillin, the antibiotic commonly referred to as the “first wonder drug.” Although Sir Alexander Fleming discovered penicillin in 1928, it was actually Florey’s methods for large-scale production of the mold P. chrysogenum that fully unlocked the potential of the drug. His work, and the large-scale production by US companies, came just in time to address a widespread epidemic of bacterial disease proliferated by the carnage of World War II. In the 25 years following 1943, the prevalence and incidence of serious bacterial infection in the United States plummeted&mdash;a true inflection point in the epidemiology of infectious disease, largely brought about by Florey’s scaling innovation (

Seventy-five years after Florey’s ground-breaking discovery, society is facing a new epidemic. However, this time it is “metabolic” disease, characterized in affected patients by abnormal levels of sugar and certain hormones in the blood, leading to (and, in turn, perpetuated by) excess body weight. These conditions are most prominently affected by human behavior and consumption. Abnormal blood sugar and associated issues predispose people to obesity and type 2 diabetes (T2D), chronic conditions the CDC labeled “the public health challenge of the 21st century” in 2009.

  In the context of this new epidemic, March 23, 2016, may well be remembered as a day that the largest healthcare payer in the United States took a critical step toward, driving an inflection point similar to that seen in the early 1940s with infectious disease.

  On that date, HHS Secretary Sylvia Mathews Burwell declared that the delivery of the Diabetes Prevention Program (DPP) has been certified as both clinically effective and cost saving when delivered to Medicare beneficiaries. She further suggested that CMS would soon provide reimbursement for administration of DPP to Medicare beneficiaries.

  To understand the potential impact of Secretary Burwell’s landmark announcement, we must first understand the true scope of the epidemic, as well as the critical role that a new innovation called “digital behavioral medicine” can play in scaling these interventions.

  Since 2011, baby boomers have aged into Medicare at the rate of about 10,000 per day.

 Today, there are almost 45 million individuals aged 65 and older in the United States. By 2030, the US Census Bureau projects that number will be nearly 73 million.

 Given these projections, Medicare enrollment in that year will be equal to a quarter of the entire US population today.

  Coupled with this demographic mega-trend are clinical dynamics that make the situation even more frightening: more than half of those over 65—a population of more than 22 million&mdash;already have prediabetes,

 and every year, up to 10% of this population will progress to T2D. The CDC estimates that 1 in 3 American adults will have T2D by 2050.

 More than one-third of those older than age 65 also have some form of heart disease,

 a condition closely correlated to prediabetes.

  In 2014, Medicare spent more than $15,700 managing the health of each beneficiary with diabetes, which included 991 annual emergency department (ED) visits for every 1000 beneficiaries with the condition and a 30-day readmission rate of 20%.

 The numbers related to heart failure—often closely correlated with T2D profiles&mdash;were even worse: Medicare annually spends more than $41,000 on those beneficiaries, who average 1657 ED admissions per 1000 beneficiaries, with a 30-day readmission rate of 24%. The Diabetes Care Project already estimates than 1 in every 3 Medicare dollars is spent treating beneficiaries with diabetes.

  The parallels between cardiometabolic chronic disease in the 21st century and infectious disease at the dawn of the 20th century are evident—this time, with a medical expense epidemic accompanying the clinical one. How can today’s healthcare leaders address both?

  The answer lies in prevention, to mitigate both medical impact and the bulk of expense. Fortunately, appropriate clinical interventions have been shown to significantly lower the risk of T2D for those at risk. In 2002, the National Institutes of Health (NIH) published results of a clinical trial that demonstrated the effectiveness of the DPP, a behavioral intervention, in helping patients lose clinically meaningful amounts of body weight, reduce their blood sugar levels, and lower the risk for diabetes by 58%.

  Secretary Burwell’s DPP announcement implies that the NIH’s clinical work, along with recent efforts by CMS to show the medical expense impact of DPP, provide the evidence base that DPP may play a key role in addressing the diabetes “double epidemic” of disease and cost.

   For the past 3 years, the Center for Medicare & Medicaid Innovation (CMMI) has operated a demonstration project at 17 YMCA sites in 8 states. In the project, the Y administered the DPP to more than 77,000 seniors who had been diagnosed with prediabetes.

 Payments to the YMCA were dependent on how many sessions beneficiaries attended and what percentage of body weight—a recognized indicator of downstream diabetes risk reduction&mdash;those participants lost. In the final evaluation, the CMS Office of the Actuary estimated that the intervention ultimately saved Medicare $2650 per beneficiary in just 15 months..

  However, Secretary Burwell’s announcement of the CMMI/YMCA project’s excellent results does not alone guarantee victory in the battle against diabetes and its associated costs. The combination of the original NIH trial and the CMMI/YMCA project included approximately 9000 individuals with prediabetes, just 0.01% of the total US population with prediabetes. DPP delivered face-to-face- to YMCAs, churches, clinics, and other community settings will be difficult to scale to 22 million Americans seniors—let alone the additional 65 million Americans who also have prediabetes.

  Digital behavioral medicine may hold the answer to this problem of clinical impact and cost-effectiveness at scale. In fact, it may be the equivalent of Florey’s scaling methods that supported the production of 646 billion penicillin doses annually by 1945, up from fewer than 10 doses in 1942. 

  Broadly defined, digital behavioral medicine refers to the digitally-enabled delivery and continuous improvement of proven effective lifestyle interventions like the DPP, while maintaining a high level of clinical fidelity to the original research. For the past 5 years, Omada Health has done just that, developing a version of the DPP, which it delivered remotely, and leveraging insights about behavioral design, user experience, and data science to layer an already effective program with predictive analytics and personalization. Today, we are the largest CDC-recognized14 provider of the DPP in the country, in-person or remote, with an enrollment of more than 55,000 participants across all 50 states.

  Most relevantly, Omada Health’s results with seniors indicate that digital delivery of the DPP can and should be an essential part of Medicare’s DPP expansion. Administering the DPP digitally can scale the program to the scope of the problem, by lowering barriers to access and success, and by ensuring that this program reaches the populations which need it the most.

  To date, Omada Health has enrolled more than 2000 participants over the age of 65. Overall, 84% of these participants, with an average age of 69, have remained active in the program after 6 months. More impressively, they’ve lost an average of 7.8% of their body weight, reducing their 3-year risk of developing T2D by more than 58%. Every day, our data science, product, and engineering teams are refining the program to adapt it for seniors’ lifestyles, stimulate even more engagement, and generate better clinical outcomes. Digital behavioral medicine has far-ranging implications for seniors with prediabetes, as well as all those at elevated risk of an entire suite of obesity-related chronic diseases.

  For the first time in history, demographic trends are forcing a fundamental rethinking of how we orient care in our public payer programs like Medicare. The current administration—and hopefully the next&mdash;has shown a willingness to embrace prevention strategies, as well as nontraditional online providers of DPP that can demonstrate excellent clinical outcomes and program integrity at scale.

  At Omada Health, it is not lost upon us that Florey’s penicillin-based innovations are credited with saving over 82 million lives. This number is tantalizingly close to the 87 million Americans currently living with prediabetes. We are invigorated by the vision that in 30 years, we will look at an epidemiological graph showing a drastic drop in US T2D incidence starting in 2018 and know that, like those penicillin pioneers, our innovative approach to delivery of a proven health intervention had a truly historic impact of the health of this country. 

Mike Payne, MBA, MSci, is chief healthcare development officer at Omada Health. 

  1. The power of prevention: chronic disease—the public health challenge of the 21st century. CDC website. http://www.cdc.gov/chronicdisease/pdf/2009-Power-of-Prevention.pdf. Published 2009. Accessed May 4, 2016.

  2. Independent experts confirm diabetes prevention model supported by Affordable Care Act saves money [press release]. Washington, DC: HHS newsroom; March 23, 2016. http://www. hhs.gov/about/news/2016/03/23/independent-experts-confirm-diabetes-prevention-modelsupported-affordable-care-act-saves-money.html. Accessed May 4, 2016.

  3. Medicare Payment Advisory Commission. Report to the Congress: Medicare and the Health Care Delivery System, Chapter 2: the next generation of Medicare beneficiaries. http://www. medpac.gov/documents/reports/chapter-2-the-next-generation-of-medicare-beneficiaries-%28june-2015-report%29.pdf?sfvrsn=. Published June 2015. Accessed May 4, 2016.

  4. Vincent GK, Velkoff VA. The next four decades: the older population in the United States, 2010 to 2050. US Census Bureau website. https://www.census.gov/prod/2010pubs/p25- 1138.pdf. Published May 2010. Accessed May 4, 2016.

  5. National Diabetes Statistics Report, 2014: estimates of diabetes and its burden in the United States. CDC website. http://www.cdc.gov/diabetes/pubs/statsreport14/national-diabetesreport- web.pdf. Published 2014. Accessed May 4, 2016.

  6. Chan A. CDC: 1 in 3 Americans will have diabetes by 2050. NBC News website. http://www.nbcnews.com/id/39798848/ns/health-diabetes/t/cdc-americans-will-have-diabetes/. Published October 22, 2010. Accessed May 4, 2016.

 7. Go AS, Mozaffarian D, Roger VL, et al; American Heart Association Statistics Committee andvStroke Statistics Subcommittee. Older Americans and cardiovascular diseases. American Heart Association website. https://www.heart.org/idc/groups/heart-public/@wcm/@sop/@smd/documents/downloadable/ucm_319574.pdf. Published 2013. Accessed May 4, 2016.

  8. Chronic conditions. CMS website. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Chronic-Conditions/CC_Main.html. Updated January 7, 2016. Accessed May 4, 2016.

  9. Diabetes fast facts. The Diabetes Care Project website. http://www.diabetescareproject.org/diabetes-fast-facts.html. Updated 2013. Accessed May 4, 2016.

  10. Knowler WC, Barrett-Connor E, Fowler SE, et al; Diabetes Prevention Program Research Group. Reduction in the incidence of type 2 diabetes with lifestyle

intervention or metformin. N Engl J Med. 2002;346(6):393-403.

  11. About the CMS Innovation Center. CMS website. https://innovation.cms.gov/about/index. html. Accessed May 4, 2016.

  12. Hinnant L, Razi S, Lewis R, et al. Evaluation of the healthcare innovation awards: community resource planning, prevention and monitoring, annual report 2015. Awardee-level findings: YMCA of the USA. CMS website. https://innovation.cms.gov/Files/reports/hcia-ymcadppevalrpt.pdf. Published March 2016.

  13. Certification of the Medicare Diabetes Prevention Program. Office of the Actuary. CMS website. https://www.cms.gov/Research-Statistics-Data-and-Systems/Research/Actuarial- Studies/Downloads/Diabetes-Prevention-Certification-2016-03-14.pdf. Published March 14, 2016. Accessed May 4, 2016.

  14. Research-based prevention program. CDC website. http://www.cdc.gov/diabetes/prevention/prediabetes-type2/preventing.html. Updated January 14, 2016. Accessed May 4, 2016.   

Now that Medicare is poised to pay for the Diabetes Prevention Program, the next question is how to make it scalable.

Mike Payne, MBA, MSci

Articles

A Bundle of Nudges: Healthcare Payment in an Era of Behavioral Science

History Lessons in Innovation: Digital Behavioral Medicine Can Address the Diabetes "Double Epidemic" Facing Medicare and America's Seniors