Time to Keep a Smart Watch on Precision Health

A Science-Based Vision for the Wearable Revolution

It is 2022, Juan, 45 years old, is thrilled to purchase a new smartwatch he can afford because it is offered at a discount through a special program with his employer. Juan plans to track his physical activity, heart rate, diet, and sleep. Under the program, Juan also receives a set of small biomonitoring sensors. Juan eagerly syncs and tries on the new devices – but anticipation turns to alarm when Juan’s watch begins to warn: “Blood pressure high. 191/142. Crisis level. Seek medical attention now.”

Juan has no obvious symptoms and thinks his new gadget defective, but his wife takes him to the emergency center just to be safe. He is shocked when the doctor tells him that, without immediate treatment and a prescription, he faces an imminent risk of heart attack or stroke. With a new perspective on life, Juan seeks follow-up care and learns that he also has prediabetes and borderline-high cholesterol levels – especially concerning given the history of early heart disease in his family. Juan then learns how to continuously monitor his blood glucose and periodically his cholesterol using his bio-monitoring sensors. He learns that his levels could improve with changes in his diet and physical activity – steps which might prevent him from ever developing type 2 diabetes or requiring statins to control his cholesterol. With continuous feedback at his wrist, Juan begins to eat better and is more physically active; now he feels great and has more energy. As a result of his experience, Juan is so enthusiastic about mobile health technology that he signs up for a long-term research program using sensors to monitor environmental exposures in people with seasonal asthma.

If you think the fictional story above is far-fetched, consider recent leaps in the field. In 2017 a leading technology company partnered with Stanford University to conduct a study of 419,297 participants to determine if a certain smartwatch could be used to identify people with atrial fibrillation (AFib). AFib is a condition in which the heart beats in a very irregular fashion increasing the risk of serious complications including stroke. The results, released last March, revealed that .5% of participants were notified that they might have AFib. When a subset of this number took follow-up testing with electrocardiogram (ECG) equipment, notifications matched between watch and ECG 84% of the time (positive predictive value of .84). In April 2018, the same company received Food and Drug Administration clearance to market a newer version of the smartwatch – now with a built-in ECG to measure heart rhythm by optical sensor. Early data indicates that the potential of the new ECG smartwatches might be even more accurate. The same device also includes a feature which can notify emergency contacts in the event that the wearer falls – a feature that could prove especially important to disabled or elderly people who live or work alone and might not otherwise receive timely help. Also of note, a different manufacturer received FDA approval in 2017 to market a device for blood pressure monitoring which was later incorporated in to a smart watch.

Consider also the enormous advances made in continuous glucose monitoring (CGM). By wearing a small, wireless sensor with a probe under the skin, the latest CGM technology enables people with diabetes to track their blood glucose levels and trends at any time of day or night with their smartwatch or other devices. When first introduced in 2000, CGM measurement error was more than 20% but by 2016 had been reduced by half and continues to improve.

Vision for Real Benefits

We see then that the story of Juan above is actually not as futuristic as it might have first appeared. In order to make such stories a common reality, however, we need more science in the development, evaluation, implementation, and equitable access to this emerging type of precision health technology. The following critical elements are needed to realize the full potential benefits of the wearable health revolution.

Critical Evaluation Research and Data Privacy Issues

Mobile health applications must be ready before they can be recommended for practice. For example, in the large smartwatch study described above which did not use the current ECG version cleared for marketing by the FDA, 16% of the wearers that received notification for AFib did not actually have the condition (i.e., they were false positives). In a large population this could mean that many thousands of people would be referred for unnecessary testing and possibly treatment, which is not without risks and unnecessary costs. Mobile health devices do not have to be as accurate as medical grade equipment, but, in our view, they should meet a threshold of validity where benefits exceed harms. If we insist on evidence before accepted practice, there is every reason to think the technology will continue to improve. We need to know more about the cost-effectiveness of mobile health technology, which, if it results in earlier diagnosis and interventions with better outcomes, might also save resources.

We also need to understand how patient-generated data can be best leveraged to improve both clinical practice and lifestyle decisions. While some research has found that health-related apps can be effective at changing people’s behavior, concern remains about bias and data quality. A 2018 systematic review of mobile health related apps concluded by recommending that effectiveness testing should be conducted prior to release and that greater controls were needed for bias, because without adequate evidence, digital medicine might “stall in its infancy for some time to come.”

Next generation bio-monitoring sensors can be developed and used to improve health in a number of growing applications through partnerships which build upon the unique strengths of industry, health care, public health, and academia. The technology, which could be combined with geographic information, may also create research opportunities to understand the critical role of long term environmental exposures for disease causation in ways never before possible. In addition, new cloud-based data platform approaches can be used to conduct groundbreaking secondary analyses of this burgeoning data in a secure, anonymous, and rigorous manner. Just as all personal health information must be protected and managed appropriately, data generated from mobile devices can only improve health if used in a manner in which privacy is ensured and consistent with best ethical, legal, and social issues practices.

Implementation Research

Health care and public health must be prepared to adapt to and understand the benefits and limitations of emerging mobile health technology. Public health could provide an important service to clinical care and the public by providing evidence-based information about these applications. In addition, secure cloud-based platforms, for patients to share bio-metric data with their health care providers and receive feedback, will likely be optimized in the future and might improve access and outcomes. Implementation research can provide a scientific foundation to address the barriers and facilitators for implementing new technologies in different communities and in real world settings. It can also help assess affordability, acceptability, appropriateness of use, and long-term sustainability.

Everyone Should Benefit

An important role for public health is to work to ensure that the mobile health revolution can benefit everyone, not just a few. A 2018 Pew survey found that 33% of Americans did not own a smartphone and by another estimate, more than 90% did not own a smartwatch. Also, while currently available CGM has been shown t o be cost-effective, many living with diabetes today cannot afford it. Although health gadgets are now viewed as luxury goods, they might eventually prove to be great economic equalizers. For example, through research we might find that, if equipped with a cellular smartwatch and a care network, some elderly or disabled people might be able to choose to live independently longer, with a higher quality of life, and at a fraction of the cost of a 24 hour care center.

We are entering a new era of precision health beyond genomic medicine. If implemented based on science, mobile health technology can greatly empower people to be engaged participants in their care and more responsible for their own health while improving preventive medicine, ultimately saving lives. It is an exciting time – not just to watch – but to ensure that we do it right.

As always, we welcome your comments and input.