February 3, 2014 | 4
“So…do I have diabetes or not, doc?” my patient asked me, one eyebrow raised skeptically. Juan was a mechanic with a calm gaze; we had first met in the emergency room a few months earlier, when I treated him for high blood pressure. Now I explained how his blood tests indicated that he had “prediabetes,” a condition that put him at higher risk for developing diabetes and its complications, including heart problems, kidney disease, and stroke.
“Does that mean I have to take more pills?” was Juan’s next question. I hesitated. The answer wasn’t straightforward, but the reality was that he was likely to require more medication at some point in the near future. Until then, all I had to offer were my best efforts at persuading him to exercise and improve his diet.
Seventy-nine million Americans are estimated to have prediabetes. More are at risk for other chronic diseases, such as cancer, cardiovascular disease, and lung disease. Many of these illnesses stem from four common risk factors: tobacco use, poor diet, lack of physical activity, and alcohol use. Yet too often, as with Juan, a doctor’s diagnosis is the first time a patient realizes they are even at risk for chronic disease. To change this, we need better preventive medicine.
A new preventive medicine would capitalize on three scientific convergences: the ability to detect predisease states earlier; more patient-centered health care models; and advances in risk factor interventions.
Detecting predisease states is not a novel concept. Pap smears, first introduced in 1928, essentially screen for precancerous tissue in the cervix; colonoscopy can serve the same purpose for intestinal cancer. Beyond oncology, predisease states include prehypertension (for high blood pressure) and osteopenia (for osteoporosis). But more sophisticated risk prediction, sometimes incorporating genetic information, may help push the detection of predisease states further upstream. For example, the U.S. Preventive Services Task Force recently recommended that women with a family history of breast, ovarian, and certain other cancers undergo genetic risk assessment. Depending on the results of this assessment, women should receive genetic counseling and, if indicated, testing for mutations in the BRCA1 and BRCA2 cancer susceptibility genes. In another example, technology to better identify predisease states could be developed around the urine metabolome, a database of detailed information on over 3,000 chemical compounds found in human urine.
Patients like Juan who are at risk for chronic diseases may also benefit from the burgeoning “consumer biometrics” movement. Startups like mc10 and Scanadu seek to enable personal monitoring of physiologic parameters like blood pressure, heart rate, temperature, and blood oxygenation. Tracking this information provides a richer set of data for clinicians and scientists, and individuals are empowered to understand their own data, potentially leading to greater engagement with their health issues. For instance, mc10, building on research by John Rogers at the University of Illinois at Urbana-Champaign, is prototyping ultrathin, flexible, skin-adherent devices akin to smart Band-Aids. One potential application is continuously monitoring blood sugar—without needles.
Along with consumer biometrics, more convenient care models promise to expand the locus of health care beyond hospitals and clinics. New patient-centered care options such as retail clinics have expanded rapidly in recent years; several major pharmacies including CVS, Caremark and Walgreens, and retailers such as Wal-Mart either have clinics or plan to launch them. Retail clinic visits were estimated to account for almost 6 million annual visits in 2012. While most visits are for simple health issues like sore throats or routine immunizations, the distributed infrastructure of retail clinics may be just as beneficial for addressing complicated, chronic diseases like diabetes. A recent partnership between Walgreens and Theranos, a company offering cheap, complete blood analysis with just a fingerstick, demonstrates how this might advance prevention. Integrating real-time laboratory capacity with medical expertise at retail clinics—and including lifestyle coaching like nutritional counseling for prediabetics—offers a convenient, comprehensive preventive medicine package.
Ultimately, earlier illness diagnosis is fruitless without effective interventions to stave off disease progression. For my patient, Juan, the most effective option is the Diabetes Prevention Program (DPP), a lifestyle intervention focused on behavioral self-management strategies for weight loss and physical activity. But scaling up the Diabetes Prevention Program to reach the 79 million patients who might benefit from it has proven challenging. An innovative partnership with the YMCA has spread the program to locations in 36 states—but only a total of about 6,000 people have enrolled. The right technology may help: a startup called Omada Health has tried to redesign the DPP to make it more user-friendly with digital tracking tools, personalized coaching and social support.
A more fundamental reinvention may be needed to achieve broad scale and address other risk factors for chronic disease. Kevin Volpp, David Asch and colleagues at the University of Pennsylvania have applied insights from behavioral economics to health, successfully structuring financial incentives to engender smoking cessation and weight loss. They are now testing whether data gathered by wireless technologies, such as smart phones, can be harnessed to infer behavioral patterns and instantaneously transmit small incentives to reward healthy behaviors and penalize unhealthy ones. For example, each day a person walks 10,000 steps could enter them into a lottery where they have a 1/100 chance of winning $100.
An important critique of the “new preventive medicine” cautions that too much of a focus on the individual draws attention away from population-based approaches to prevention, such as cigarette taxes or bans on trans fats. In this view, first posited by the epidemiologist Geoffrey Rose, “a large number of people at small risk may give rise to more cases of disease than a small number of people at high risk.” One answer to this critique is that individualized prevention is not mutually exclusive with population-based prevention. Indeed both strategies reinforce each other—treating prehypertension and reducing population-wide sodium consumption would save more lives together than either alone.
Anthony Viera at the University of North Carolina has laid out an important framework for thinking about predisease states. One question predominates: Do the benefits of intervening in the predisease stage outweigh the harms in the population? Fulfilling this condition is particularly important to avoid overdiagnosis and overtreatment. For example, an imperfect diagnostic test for breast ‘precancer’ may cause more harm than good if it subjects women to unnecessary biopsies and surgeries. As Dartmouth’s H. Gilbert Welch has pointed out, the flip side of early diagnosis is disease-related anxiety and complications from needless medications and procedures.
Juan ended up redoubling his commitment to exercise and lost about five pounds. I referred him to a local YMCA Diabetes Prevention Program, but the timing didn’t dovetail with his work schedule. He’s tried a pedometer and we’ve discussed a home blood pressure cuff, but they are a tough sell (he’s still happy with his flip phone). All in all, I won’t be surprised if Juan’s numbers keep creeping up and I make the diagnosis of diabetes sometime in the next five years. And I can’t help but think that we should be able to do better—for Juan, and for so many others like him.