The Tipping Point for Digital Health

If a patient falls ill in their home, and no doctors are around to see it, did it really happen?

The answer to this question may seem straightforward until we consider its larger contextual and philosophical underpinnings. From the patient’s perspective, the answer is certainly “yes” as they experienced first-hand the symptoms of their illness and its impact on their daily life. From a health system perspective, however, the answer becomes less straightforward. Did the patient visit their general practitioner (GP) or receive a prescription for their illness? Did their illness worsen to the point where they ended up visiting the emergency department? If the answer to these questions is “no”, then it is unlikely that any relevant information has been captured on the patient’s current health state. From the perspective of the health care system, the patient may as well have never been sick.

Make no mistake, the capture of this data is powerful. Real world data (RWD), or data gathered outside of clinical trials, can be used to shape clinical decision making, inform epidemiological research and our understanding of disease burden, develop clinical guidelines, monitor post-market safety, shape regulatory decisions, support treatment coverage, or even reassess and disinvest in current treatments. Despite the wide array of RWD applications, the sources of RWD used to inform these critical decisions are contrastingly narrow.

Across the globe, researchers, healthcare providers, and industry are limited by data silos and fragmentation. Layered onto the issues of access and infrastructure are the pervasive challenges of data privacy, governance, and quality. Even in regions that have overcome many of these obstacles and championed systems of robust and accessible data, the predominant type of RWD captured is readily available or routinely collected administrative health data (e.g. claims data, prescriptions, hospitalizations), which on their own present a narrow view of the patient experience.

DIGITAL HEALTH TECHNOLOGY: PANDEMIC-DRIVEN DISRUPTION

In the last decade, the world has seen a bourgeoning culture of health digitization and increasingly, patients (or consumers) have been able to collect vast amounts of information on their mobile and wearable devices. The excitement surrounding personal wellness tracking and innovative care delivery has also been met with a healthy skepticism about the inclusiveness of these technologies and the potential to create a digital divide (described below). Despite the promise of artificial intelligence (AI) and predictive analytics, the adoption of digital health technologies has lagged innovation in other sectors. Since the widespread adoption of mobile banking, visiting one’s financial institution has become a rare occurrence, while the same cannot be said for healthcare institutions with their heavy reliance on face-to-face interaction.

Then along came COVID-19. Years of research and technology advancement could not compare to the catalytic effect of the coronavirus pandemic on digital health adoption. With widespread lockdowns and infection control measures, there has been a rapid shift away from the traditional face-to-face care model and as a result, a rapid deployment of digital health solutions. As the level of urgency warranting in-person care has been elevated, patients with non-urgent needs in many parts of the world now have increased access to care and medication provision within the comfort of their own home. Perhaps the most notable shift is in the adoption of telehealth, which has seen substantial scale-up of offerings that we must work to sustain post-pandemic.

Beyond telehealth, COVID-19 has dramatically accelerated the adoption of other digital health technologies. From remote monitoring of patients with COVID-19 or other conditions, to the emergence of mobile apps to verify vaccination credentials, it is clear that these tools have a critical role in pandemic response as well as providing longer-term solutions to pervasive health system challenges. Electronic Health Records (EHRs), for instance, have supported public health responses to COVID-19; yet are far from fulfilling their full potential as standardized and interoperable systems accessible to both clinicians and patients. The pandemic has highlighted gaps in health data ecosystems, particularly related to the timely collection of clinically-relevant patient information outside of traditional clinical settings. Addressing these gaps through digital health technologies and artificial intelligence (AI)-driven analytics has evolved from a “nice-to-have" to a necessity. With this renewed interest in digital health from all system stakeholders, it is critical that our ultimate aim for any disruptive innovation is an equity-enhancing health system.

DIGITAL DIVIDE: CLOSING THE GAP

The digital divide refers to the gap between demographics and regions that have access to and use of modern information and communications technology and those that do not. As use of digital health technologies becomes more widespread, it is important to consider who has access to these tools and their health benefits and, conversely, what populations may be disproportionately excluded. Factors contributing to the digital divide can include lack of access to the Internet or digital devices (e.g. smartphones, laptop), low literacy or digital literacy, and even distrust in technology or the people who govern it. As these factors are deeply rooted in the social determinants of health, it is not hard to imagine how a seemingly helpful new digital health app could exacerbate health inequities in underserved populations.

However, traditional models of paternalistic, clinic-focused care are arguably even greater threats to health equity in certain contexts. A poignant example of this reality is seen with Indigenous populations, for whom digital health technologies may provide a necessary, but not sufficient tool to tackle more deeply rooted marginalization. From issues of access (e.g. distance to clinic) to experience (e.g. culturally insensitive care), the absence of digital health technologies creates insurmountable barriers for many patients. Consider hard-to-reach populations experiencing mental illness, such as depression or schizophrenia, whose data footprint may reveal no clinical care visits despite experiencing major deterioration in their health status. Another population to consider is those living in resource-limited settings. As digital technologies have become increasingly accessible around the world, low- and middle-income countries have demonstrated tremendous success in the adoption and scale-up of digital care models. Across these population groups, there is a fundamental need to collect more (and better quality) health data from patients in real time and in the real world. Helping to deal with the volume and velocity of data are AI applications spanning detection, diagnosis, treatment, and prediction of health outcomes. Propelled by the COVID-19 pandemic, AI-driven prediction—from patient-level illness up to population-level outbreaks—is one of the greatest opportunities afforded by patient-generated digital health data.

Now is the moment to build on the momentum ignited by the pandemic to advance digital health technologies, to embrace new digital healthcare delivery models, to invest in digital health research and partnerships, and to relentlessly innovate for patient-centered care. Now is the moment for digital health.

About the author:

Evelyn Pyper is an Evidence Based Healthcare (EBHC) DPhil Student at the University of Oxford, supervised by Dr. Jamie Hartmann-Boyce and Prof John Powell.

Outside of her studies, Evelyn works for Johnson & Johnson’s Global Public Health team. The views expressed here are her own and do not represent the views or opinions of her employer.