Can Wearable Devices Help Reduce Health Disparities and Add Value?

John Watkins, PharmD, MPH, BCPS,
Premera Blue Cross, Mountlake Terrace, WA, USA

COVID-19 lockdowns changed our lives overnight, eliminating the divide between people who had ready access to healthcare and those who didn’t. Suddenly, everyone had barriers. Clinics closed and mobility evaporated. For a few weeks in the spring of 2020, all of us experienced frustrations that are daily life for many of America’s poor. Through the pandemic, we learned what it is like to live without that access. We accepted workarounds including telehealth visits and other digital health technologies (DHT) as substitutes for in-person care. Often, it was our first exposure to telehealth and the technology that goes with it. Having seen a real-world demonstration of how these technologies can bridge access barriers, how will we use that knowledge to address disparities and improve access to care for all?

As one participant in the Innovation and Value Initiative Health Equity Initiative noted, “Equity is about removing barriers and obstacles to having just opportunity for health. If you have not worked to understand the social, cultural, and community drivers that affect people, then you are not assessing value.”¹ DHT is transforming healthcare, reducing the negative impact of physical and geographical barriers.

As clinics reopened following the disruption of pandemic lockdowns, there was a shortage of physicians, nurses, and other health professionals due to burnout and early retirement. Not unexpectedly, payers saw a cost-trend rate increase driven by treatment of acute and long COVID, and by patients with delayed diagnosis of treatable conditions due to suspension of routine screenings caused by the pandemic. The global societal cost of long COVID alone has been estimated to be $2.6 trillion² and the direct medical cost $163 billion.³ Much of this burden has fallen on the United States, making it more urgent to increase efficiency. Consumers demand transparency, and health systems are improving patient portals to provide more access to their electronic health records. These portals provide a ready connection point for home digital devices. With the decreasing cost and increasing power of hardware and software, we can expect a technology explosion.

Although COVID-19 has spread everywhere, statistics show substantially worse outcomes in lower income Black, Hispanic, and Native American communities, where vaccination rates were lower and crowded living conditions facilitated transmission. People in remote rural areas that already had difficulty reaching providers saw that access further limited. These barriers can be subtle or obvious. They are an integral part of social structures, including healthcare, and they will not be breached easily. Is digital technology up to the challenge? This article explores some of these barriers and suggests ways in which it could help address them. In addition to improving patient access to care, monitoring devices are expected to see expanded use in clinical trials, enabling direct measurement of endpoints that were previously unavailable to researchers and could be tracked only through patient and caregiver diaries. Clinicians and researchers must consciously work to include the voices of target patient communities in the design, implementation, and evaluation of projects, ensuring that the work aligns with their needs and priorities.

Barriers Separate and Isolate
Geography limits access. People in remote areas must drive long distances to see specialists. In rural Alaska, even basic primary care access may be unavailable in some communities at some times. Distance is not the only barrier. Many low-income urban residents do not own cars. Their trips to clinics or hospitals on buses and trains are often short distances that take a long time due to multiple transfers. This is challenging enough when one is healthy; it is more difficult for the chronically ill. Furthermore, low-wage workers may have difficulty getting time off for provider visits and may not get paid for missed work hours, adding financial stress to their health concerns. Less education and poor health literacy correlate with chronic disease,⁴ and these individuals are subject to depression and reduced employment that add to their underlying medical problems.

Language and culture are common barriers in immigrant communities, where translators that can interpret cultural nuances and expectations may not be available. Even when language is understood, clinicians’ advice is likely to be ignored if it conflicts with traditional health beliefs. For immigrants, allopathic medicine may be their last resort after familiar remedies have failed. For example, working in Nepal, I learned that traditional health beliefs based on Ayurvedic medicine classify diseases as either “hot” or “cold.” Our patients wanted to know which foods they should eat while taking the drugs we gave them, so the pharmacy staff would add, “Don’t eat hot (or cold) foods while taking this medicine,” to the usual prescription counseling. This advice was medically meaningless and the choice of hot or cold random, but we hoped it would improve credibility and adherence.

Historical abuses of Black Americans by the healthcare system have created reasonable suspicions that impact willingness to seek care and impair trust in medical advice. A symposium speaker⁵ recently described a documentary on gene therapy for sickle cell disease, in which the narrator had casually mentioned without further explanation that a lentiviral vector, which is a modified HIV virus, was used to deliver the gene. Recalling the infamous Tuskegee syphilis study,⁶ a logical reaction from a Black person would be, “Great! First, you gave us syphilis—now you want to give us HIV!” There is a long history of such medical abuses and “separate and unequal” care. Cultural memories last for generations, and unintentional ignorance adds to justified distrust.

Ethnic and genetic characteristics mix to create heterogeneous populations, for which standard racial classifications used in medical records are insufficiently granular to guide clinicians and researchers. A person classified as “Hispanic” could be White, Black, Native American, or any combination. Cultural beliefs and practices in Caribbean countries show African influence, while people living along the Andes inherit the beliefs of their ancestral native cultures. Geography and climate are radically different across Latin America, so it is unlikely that 2 cultures in different regions would share all the same health beliefs. Two individuals classified in this overly broad category may share no racial ancestry and have only the Spanish language, colored with local vernacular and pronunciation, and vestiges of Spanish colonial culture in common. If we are to seriously consider these patients’ perspectives, more precise information is needed.

Home-Based Monitoring Can Improve Access
As the population ages, efficient secondary prevention for chronic conditions is needed. Home monitoring technologies, including in-home sensors and wearable digital devices, offer increasingly detailed and sophisticated continuous monitoring of patients. Artificial intelligence can interpret the results, disaggregate inputs, and filter noise. Continuous measurement gives a fuller picture than discreet data points collected at intervals during clinic visits, allowing the development of individualized strategies to manage patients’ conditions, maintaining health and functionality, and potentially improving outcomes. Continuous glucose monitoring in diabetic patients is an early example of a well-developed mature technology that helps patients reduce hemoglobin A1c levels and avoid acute hypoglycemic episodes. Remote monitoring could help patients that have difficulty accessing clinics by reducing the need for in-person visits with their providers.

Physical activity is of critical importance to complex internal medicine patients and the elderly in general. Reduced mobility decreases overall health. Objective measurement can give clinicians a more realistic picture of the patient’s daily patterns of movement and alert them when activity levels decrease, as is often the case after changes in medications or surgical procedures. This technology can also help diagnose and follow neurodegenerative diseases in the elderly, monitoring fall risk and the need for in-home assistance. Specialties that could benefit from applying digital monitoring include oncology, cardiology, immunology, endocrinology, pulmonology, neurology, psychiatry, geriatric medicine, and rheumatology.

Patients with Parkinson’s disease would be prime candidates for this type of assistance since they suffer complex movement disorders, often accompanied by depression, cognitive decline, and sometimes psychosis. Patients can be monitored remotely for changes to functional status, response to changes in medication, and needs for in-home care. Pharmacotherapy for patients with late-stage Parkinson’s disease involves a delicate balance of multiple medications, including those used to treat comorbidities. Home monitoring can quickly identify providers when a medication change has not improved things or has caused unwanted side effects.

We are a data-driven society. Increased computing power and memory, miniaturization, and artificial intelligence will expand the range of potential applications. As patients acquire “smart home” technology, it becomes easier and less costly to combine multiple devices, improving accuracy and sophistication of measurement. Payers must develop appropriate coverage criteria, and that will require new evaluation methods. Low-income individuals may need financial assistance to upgrade home infrastructure to support the technology.

Monitoring Can Improve Usefulness of Trial Outcomes
Potential applications of in-home digital monitoring in pharmaceutical research were explored in a recent ISPOR webinar series.^7,8 The safety and efficacy of many drugs depend on how they impact patients’ functioning in a real-world setting, which is difficult to reproduce with in-clinic monitoring. Patient-reported outcomes are subject to reporting errors and may be colored by subjective experience. Combining objectives in-home measurements with patient-reported outcomes may provide a fuller picture by combining subjective and objective inputs. For example, Alzheimer’s disease, a major target for drug development, produces changes in daily behavior and sleep patterns that are measured by patient and caregiver diaries. Patients with early stage dementia, although still capable of living independently, may forget to report things but appear alert and oriented in clinic visits. Direct measurement could give a clearer picture of how a drug regimen affects the patient. Current trial evidence for Alzheimer’s disease drugs is frustratingly inadequate; it is hoped that digital monitoring will improve our understanding of their true effectiveness. 

"Clinicians and researchers must consciously work to include the voices of target patient communities in the design, implementation, and evaluation of projects, ensuring that the work aligns with their needs and priorities."

Regulators must approve the endpoints in registration trials. This begins with agreement that the endpoint is an appropriate measure of the proposed clinical outcome. The accuracy of measurement of the monitoring device must be demonstrated, and its validity in the patient population and setting(s) of interest must be shown. For example, daily movement patterns of a person in a home setting might be different from those of the same individual in an assisted living facility. The device(s) must be acceptable to patients to wear long-term, and a device capable of multiple measurements would be preferable to multiple devices. The algorithms that analyze raw data must be validated (analytical validity—does it measure what we think we are measuring). Artificial intelligence will play an important role in refining this. Then, the developers must confirm that the measurement correlates with a clinical outcome of interest (clinical validity). These steps are required of any diagnostic and are relatively easy.

The final step, of greatest interest to payers and health technology assessors, is demonstrating clinical utility, which means that the use of the intervention in a population of interest produces overall net health benefit. Because clinical utility is not required in safety and efficacy trials, developers may have limited incentive to generate this evidence, which requires long-term, real-world use. However, if the specific measurement comes to be used for routine monitoring outside of trials, the manufacturer could collect and analyze data from large databases to produce the required real-world evidence.

The Voices of Lived Experience Are Essential
Patient centricity is supported by the Affordable Care Act, which created the Patient-Centered Outcomes Research Institute. The US Food and Drug Administration has emphasized including patients at earlier stages in the clinical drug development process. Researchers and technology assessors understand that value assessments must incorporate patient perspectives. Historically, research was based on what sponsors, investigators, payers, public health experts, and regulators thought was important. Recognizing patients was a crucial first step in the right direction, but we must move beyond thinking of patients as a homogenous group. Another Innovation and Value Initiative Health Equity Initiative participant observed, “If you don’t see how race, income, gender, and other patient characteristics inherently drive value, then you are not assessing true value in healthcare.”⁹

Individuals’ and patient subgroups’ perspectives may vary, and a collective approach usually underrepresents the perspectives and concerns of minorities. Information asymmetry negatively affects patient empowerment in these interactions, a problem that is magnified by the lower educational levels and power imbalance in minority communities. This is a problem we must address to achieve fair treatment for these groups.

Organized efforts to educate patient representatives can reduce the asymmetry. For example, the European Patients’ Academy on Therapeutic Innovation (EUPATI) is a collaborative nonprofit organization that “provides education and training to increase the capacity and capability of patients and patient representatives to understand and meaningfully contribute to medicines research and development, and to improve the availability of medical information for patients and other stakeholders.”¹⁰ EUPATI graduates have impressive knowledge, understanding, and ability to engage in peer-to-peer conversations regarding their needs and concerns with health professionals, policy experts, and others. These conversations are enlightening and often expose erroneous presuppositions held by professionals.

Researchers must work in partnership with patients to incorporate lived experience from patients, caregivers, and communities, including groups that have been left out of the conversation. This is necessary to develop clinical trial designs and endpoints that more realistically reflect what matters to each patient. Are we asking the right research questions? Will the trial designs, populations, and endpoints produce the data needed to support minority patients’ choice of the best treatment for them? The addition of objective endpoints collected by wearable devices and other in-home monitoring to clinical trial designs can provide a fuller picture of how an intervention actually affects patients if we choose the right endpoints and interpret them correctly. Adequate representation of minorities in trial populations will help us determine whether the intervention works for them as well as for the majority.

This sounds simple, but there is no agreement as to how to make it happen. People are incredibly diverse in so many ways that the one-dimensional data points in a clinical trial cannot be expected to do them justice. Whose voices should we listen to? How will we know when we have sampled enough? These are tough questions, but their difficulty does not excuse ignoring them.

Setting Priorities
To return to the original question, how can digital technologies help level the playing field of access and reduce health disparities? The Innovation and Value Initiative’s Health Equity project is examining the relationship between health equity and value, with the goal of “elevating the national discussion” on this important issue from a societal perspective. The group’s steering committee explains that “Health technology assessment advances health equity when it reduces health disparities by aligning access and affordability of healthcare technologies and services with the differing needs and values of diverse patient populations, especially those who are most marginalized.”¹

The conversation about value in healthcare has been dominated by payers, providers, health economists, and policy experts, but the patient is the ultimate judge of whether a healthcare intervention has value. All of us will agree that certain outcomes (eg, freedom from pain, adequate nutrition, mobility, etc) are important, but beyond these, priorities vary among and within different populations. When the population in question is a disadvantaged minority, it is especially important to listen with open minds and hearts, connecting and establishing credibility, and facilitating the necessary education and support to empower their participation as equals in the process. Viewing value assessment from the perspective of lived experience is complex when you consider ethnic, racial, cultural, and genetic differences among patients that share a common medical condition, but we must make the effort.

References
1. Innovation and Value Initiative. Value Blueprints: No Value Without Equity: Action Opportunities Emerging from the IVI Health Equity Initiative. Published January 2023. Accessed January 30, 2023. https://thevalueinitiative.org/wp-content/uploads/2023/01/Value-Brief_No-Value-Without-Equity_FINAL.pdf

2. Cutler DM. The costs of long COVID. JAMA Health Forum. 2022;3(5): e221809. Accessed January 28, 2023. doi:10.1001/jamahealthforum.2022.1809

3. Richards F, Kodjamanova P, Chen X, et al. Economic burden of COVID-19: a systematic review. Clinicoecon Outcomes Res. 2022;14: 293–307.

4. Oates GR, Jackson BE, Partridge EE, et al. Sociodemographic patterns of chronic disease: how the mid-south region compares to the rest of the country Am J Prev Med. 2017 Jan; 52(1 suppl 1): S31–S39.

5. Gene Therapy for Sickle Cell Disease. PGTME Working Group 3rd Annual Lunchtime Lecture Series, November 30, 2022. Accessed January 28, 2023. https://www.eventbrite.com/cc/3rd-annual-pgtme-lunchtime-lecture-series-1278259

6. CDC. The USPHS Syphilis Study at Tuskegee. Reviewed November 3, 2022. Accessed January 28, 2023. https://www.cdc.gov/tuskegee/index.html

7. Grant S, Gua C, Craddock I, Campbell M. Part I: Achieving fit for purpose data from wearables for age-related diseases. January 13, 2023. Accessed January 25, 2022. https://www.ispor.org/conferences-education/education-training/webinars/webinar/achieving-fit-for-purpose-data-from-wearables-for-age-related-diseases

8. Eustaquio T, Ritchie A, Kolasa K, et al. Part II: Digital endpoint adoption: the how, what and why. February 9, 2023. Accessed January 30, 2023. https://www.ispor.org/conferences-education/education-training/webinars/webinar/digital-endpoint-adoption-the-how-what-and-why.

9. European Patients’ Academy on Therapeutic Innovation (EUPATI). Accessed January 30, 2023. https://eupati.eu/