How to Save $200 Billion Just By Sending Text Messages
By Jim Walker
In September of 2013, a medical research study conducted in southeast India was published in The Lancet that may help save the US healthcare system from bankruptcy, and could ultimately change the way preventative healthcare measures are delivered around the world. Although it received little media attention, researchers from the India Diabetes Research Foundation have made a truly incredible discovery: using nothing more than 2-3 SMS text messages per week, delivered over the course of two years, they were able to reduce the chances that at-risk men develop type 2 diabetes. In a group of 537 patients, 18% of the participants who received regular messages containing information about healthy lifestyle and the benefits of physical activity and diet developed type 2 diabetes, compared with 27% in the control group who did not receive the messages. The number needed to treat to prevent 1 case of diabetes was 11. Put simply, sending basic text messages with health reminders was able to lower the risk of developing diabetes by 36%.
Desmond Johnston, FMedSci, an author of the study from Imperial College London, United Kingdom, noted, “We’ve known for years that you can prevent type 2 diabetes by modifying your lifestyle, but the problem is how to support people to do that. Frequent personal contact with a doctor is effective, but it’s very costly to provide. We’ve shown that you can achieve similar results using mobile phone messaging, which is a very low-cost solution. We plan to carry out more studies in other populations, but there’s no reason why this strategy shouldn’t work anywhere else in the world.”
Dr. Atul Ramachandran, a leading diabetes specialist in India who also co-authored the paper, explains that “the SMSes were sent twice a week, and we had a mechanism to ensure that they were received. We modified messages from time to time and we also know that SMSes were acceptable to 97 percent of those who received them. This was important because we were interfering in personal lives.”
So, what was the complex gaming protocol for this dramatic health intervention? Actually, the intervention was incredibly simple: the text messages were sent from 7:00 am to 9:00 am and again from 5:00 pm to 9:00 pm. The schedule was determined by polling the individuals as to preferred times to receive the SMS. The texts were computer-generated and sent automatically; they were designed to prompt physical activity—stating, for example, “Use stairs instead of a lift [elevator],” and to encourage healthy eating—”Avoid snacks while watching TV; you may overeat.” The 60 to 80 different messages were created and sent cyclically, so that participants would not be likely to receive the same message in a 6-month period.
What’s truly remarkable about this study is its simplicity and disproportionate impact. From a gaming perspective, there was no immediate feedback or scoring, no dynamic personalization of content, and no group social interaction, save for 6-month assessments—and yet significant health improvements occurred. The opportunity moving forward is to not only repeat these outcomes in different locations with other disease states, but to deliver even better results by applying additional gaming elements to the text message protocol.
Let’s examine some of the game approaches that could be deployed in order to drive the number needed to treat down into the single digits. Rajat Paharia, gamification expert and author of Loyalty 3.0, describes ten elements that help to drive a successful gaming experience. How might each of these elements be applied to text messaging campaigns such as the one conducted in India so that even fewer patients develop type 2 diabetes?
- Fast feedback—I get immediate feedback or response to actions
New text messaging technologies such as Trext (http://trext.me/) allow for individualized branching responses to be sent to users, delivering custom answers based on specific feedback. These systems are cost-effective and easy to program, creating personalized text messaging feedback with a minimal investment of time or programming resources.
- Transparency—I can see where everyone (including me) stands, quickly and easily
If you have ever seen the list of “high score” initials at the start of a game, you understand the impact of transparency. Leader boards, social networks, and even the simple sharing of the “Top 10” achievers have all proven powerful gaming mechanisms for raising the performance of entire teams. In terms of health interventions, it is easy to imagine participants receiving text messages with monthly or semi-annual lists of the exemplary achievers in their cohort.
- Goals—I have short and long-term goals to achieve
As part of his extensive research on behavior modification, Stanford professor BJ Fogg has created a deceptively powerful course called “Tiny Habits,” in which players commit to achieving three “tiny” goals. (http://tinyhabits.com/) The participants simply e-mail in their goals, and then receive short feedback messages each day, prompting them about their progress. For more aggressive interventions, this type of daily feedback can drive even higher health outcomes.
- Badges—I can display evidence of my accomplishments
2013 will mark a technology milestone, with more than 1 billion smartphones being sold around the world. As these graphically rich phones proliferate to even the poorest of the poor, it will become cost-effective for any patient in the world to receive health-related badges that can be unlocked as they achieve various prevention milestones.
- Leveling up—I can achieve status within my community
Achievements and levels are well-known aspects of game play that have yet to be applied to healthcare prevention in any meaningful way—with perhaps the exception of cholesterol numbers and blood sugar levels. Creating more sophisticated and engaging levels of achievement in diabetes prevention, and providing means to “level up” via text messaging will drive patient engagement and outcomes.
- Onboarding—I can learn in a compelling and engaging way
Clear instructions at the beginning of any game are critical so that everyone understands the rules, and knows what it takes to succeed. Visual storytelling, delivered via smartphones, can quickly convey important health information at the beginning of the intervention, ensuring that the goals, objectives, and most importantly—the benefits of “winning” are clearly understood.
- Competition—I can see how I’m doing against others
Competition is not a word typically used to describe interactions between patients, but—like transparency, it can be harnessed in a powerful way to help increase compliance and consistency, while helping spark ongoing motivation. Creating small teams of patients who progress through the text messaging intervention together can help to create this supportive, but powerful level of competition.
- Collaboration—I can work with others to accomplish goals
Competition and collaboration are not mutually exclusive. Small teams progressing through the text campaign together can collaborate with one another, sending their own text messages of encouragement, and offering tips and suggestions about how to level up.
- Community—I can see what the community is doing—the community can see me
Linking the text messaging campaign to a social network like Facebook can provide a powerful mirror for all participants to see their progress in a larger context of achievement and accomplishment. This type of community experience can also become a forum for recognition and celebration of healthy accomplishments.
- Points—I can see tangible, measurable evidence of my accomplishments
Many health interventions can lend themselves to a range of activities and physical tests that can be translated into game points. BMI, weight, blood pressure, daily activity, and caloric intake—all of these can be measured and translated into an ongoing scorecard for each patient. Although capturing these measurements is currently either self-reported or requires a trip to the doctor, in the near future, the wide proliferation of health-monitoring devices such as the FitBit will allow for scoring to be recorded and sent to the cloud automatically.
It’s important to note that applying these 10 principles across health interventions is both a science and an art. Brian Burke, research vice president at Gartner, notes that “the challenge facing project managers and sponsors responsible for gamification initiatives is the lack of game design talent to apply to gamification projects. Poor game design is one of the key failings of many gamified applications today.”
“The focus is on the obvious game mechanics, such as points, badges and leader boards, rather than the more subtle and more important game design elements, such as balancing competition and collaboration, or defining a meaningful game economy,” Mr. Burke said. “As a result, in many cases, organizations are simply counting points, slapping meaningless badges on activities and creating gamified applications that are simply not engaging for the target audience. Some organizations are already beginning to cast off poorly designed gamified applications.”
Despite these challenges, the promise of engaging and powerful text-based health interventions delivered directly to patients on their phones is enormous. We are just in the early stages of what will be a dramatic leap forward in public health education and prevention.
Focusing on diabetes alone, the potential for cost savings is substantial. Researchers at the CDC, publishing in the Journal of Preventative Medicine, have projected that the US diabetes population—currently at 26 million—will approximately double in the next 20 years, with costs tripling to $336 billion. Each new case of diabetes incurs an average cost of $85,000 on the healthcare system. (http://www.ajpmonline.org/webfiles/images/journals/amepre/AMEPRE_3842-stamped-080613.pdf)
Furthermore, diabetes hits seniors particularly hard. Senator Susan Collins (R-Maine), co-chair of the Senate diabetes caucus, notes that approximately 1 in 3 Medicare dollars is spent on diabetes. “When you look at the impact of diabetes, it’s evident that if we could prevent the disease and come up with better treatments, we could have a major impact not only on people’s lives, but also on the Medicare and Medicaid budgets.”
So, based on the results seen in the Indian diabetic research, what are the potential costs savings of a text-based diabetes intervention in the US?
- First, let’s estimate the projected new diabetes cases over the next 20 years in the US at 25 million.
- At a current lifetime expense of $85,000 per case, the epidemic of new diabetes cases is going to cost an additional $2.1 trillion (25 million cases x $85,000/case, assuming costs per case don’t rise).
- Moving forward, if just 5 million people who are pre-diabetic enroll in a text message-based intervention similar to the one in India, with a number needed to treat of 11, we can estimate that approximately 500,000 cases of diabetes could be prevented.
- Preventing 500,000 cases at $85,000 a case will result in a cost savings of $42.5 billion.
- However, if gaming technology can improve the impact of text-based interventions to a number needed to treat of 1 case prevented for every 5 enrolled, (aggressive but not impossible to imagine) and 12 million pre-diabetic patients participate over 20 years, then approximately 2.5 million cases would be prevented at a cost savings of $212 billion (2.5 million x $85,000).
- Assuming that in any given year, 4 million patients are enrolled in the program and administration costs are estimated at $2.00/patient/week, (actual costs would likely be lower) total implementation costs per year would be $400 million—or $8 billion over 20 years. In other words, for every $1 dollar spent developing and delivering text message-based diabetes intervention, there will be a potential savings of more than $25. (Invest $8 billion with the potential to save $212 billion in additional healthcare costs).
From a health policy perspective, it beyond the scope of this article to determine where the investment in text-based preventive programs will originate from, but—given the significant potential costs of the diabetes epidemic—funding to develop advanced text-based health interventions will most likely come from a combination of federal, insurer, and employer-based organizations.