Highlighted DH

DISCUSSION PAPER
The Promise of Digital Health: Then, Now, and

the Future
Amy Abernethy, MD, PhD, Verily; Laura Adams, MS, National Academy of Medicine;
Meredith Barrett, PhD, ResMed; Christine Bechtel, MA, X4 Health; Patricia Brennan,
RN, PhD, FAAN, National Library of Medicine; Atul Butte, MD, PhD, University of California,
San Francisco; Judith Faulkner, MS, Epic Systems; Elaine Fontaine, BS, National Academy
of Medicine; Stephen Friedhoff, MD, Anthem, Inc.; John Halamka, MD, Mayo Clinic;
Michael Howell, MD, MPH, Google Health; Kevin Johnson, MD, University of Pennsylvania;
Peter Long, PhD, Blue Shield of California; Deven McGraw, JD, MPH, Ciitizen Corporation;
Redonda Miller, MD, MBA, Johns Hopkins Hospital; Peter Lee, PhD, Microsoft Corporation;
Jonathan Perlin, MD, PhD, MSHA, The Joint Commission; Donald Rucker, MD, 1upHealth;
Lew Sandy, MD, MBA, UnitedHealth Group; Lucia Savage, JD, Omada Health, Inc.; Lisa
Stump, MS, Yale New Haven Health System and Yale School of Medicine; Paul Tang, MD, MS,
Stanford University; Eric Topol, MD, The Scripps Research Institute; Reed Tuckson, MD, FACP,
Tuckson Health Connections, LLC; and Kristen Valdes, b.well Connected Health
June 27, 2022

Digital Health in the 21st Century biomedical science. These developments promise to drive
earlier diagnoses and interventions, improve outcomes, and
Over the past several decades, the development and ac-
support more engaged patients (McGinnis et al., 2021).
celerated advancement of digital technology has prompted
In the mid-20th century, the newly established World
change across virtually all aspects of human endeavor. The
Health Organization (WHO) defined the concept of health
positive and negative effects of these changes have been
as “a state of complete physical, mental and social well-
and will remain the focus of active speculation, including the
being and not merely the absence of disease or infirmity”
implications for human health. Application of mechanical
(WHO, 2006). As an integrative concept, this definition is
and digital recording and capture of physical status, experi-
a vision for the planet that is at once bold and elusive, even
ences, and narratives have set the stage for revolutionary
for the United States as the world’s wealthiest nation. The
progress in individual health and medical management,
WHO definition is clear that health derives from much more
population-wide health strategies, and integrated real-time
than medical care. Since WHO’s founding, much has been
generation of new knowledge and insights. Together, these
learned about how different factors, including but extending
developing digitally mediated capacities are termed digital
far beyond medical care, interact to shape health prospects.
health.
Indeed, research indicates that social and behavioral fac-
Digital health has evolved as a broad term encompass-
tors both outweigh medical care in determining health sta-
ing electronically captured data, along with technical and
tus and modulate the contributions of genetics and physical
communications infrastructure and applications in the health
environments (Kottke et al., 2016; McGovern et al., 2014;
care ecosystem. Revolutionary advances in digital health
Schroeder, 2007; McGinnis et al., 2002).
are transforming health, medicine, and biomedical sci-
Unfortunately, U.S. health policies and health system in-
ence, and redefining and re-engineering the tools needed
vestments remain misaligned with these insights. In the U.S.,
to create a healthier future. Developments such as cloud
approximately 90% of all health expenses go to disease
computing, artificial intelligence, machine learning, block-
and injury treatment rather than to addressing the predis-
chain, digitally mediated diagnostics and treatment, tele-
posing factors of these illnesses and injuries. By 2020, U.S.
health, and consumer-facing mobile health applications are
health expenditures had grown to $4.1 trillion. Spending in
now routinely used in self-management, health care, and
the health sector is projected to increase to over $6 trillion
Perspectives | Expert Voices in Health & Health Care

DISCUSSION PAPER
annually and encompass 20% of the nation’s gross domes- ditions for long-term progress; and
tic product by 2028 (Keehan et al., 2020; CMS, 2019). The • identify critical priorities for cooperation and col-
U.S. is falling far short of the WHO vision, despite spend- laboration between policy makers, practitioners,
ing nearly twice as much as other high-income countries. and industry leaders to propel the development and
The U.S. currently has a lower life expectancy, higher rate application of best-in-class digital health tools.
of death by suicide, higher chronic disease burden, higher
rates of preventable hospitalizations, higher use of unneces- This paper aims to provide a comprehensive review of
sary expensive testing and procedures, and lower use of digital health tools and their promise and to identify critical
primary care than its peer countries (Tikkanen and Abrams, priorities for cooperation and collaboration among policy
2020). makers and industry leaders. The challenge is addressing
Despite important gains in the last two decades, made both the breadth and depth of the issues, which are multi-
possible by significant investment by payers, providers, and factorial and overlapping.
the federal government in electronic health records (EHRs), It is important to note that the narrative and suggestions
progress toward interoperable systems, and advanced tech- here represent the views of the individual authors, not nec-
nology to coordinate care and manage disease, the prom- essarily those of the National Academy of Medicine or the
ise of digital health remains illusory. The ability to use in- organizations with which the authors are affiliated. In de-
teroperable digital technology to improve the effectiveness, veloping the text, the authors have been informed by their
efficiency, equity, and continuity of care remains substan- respective roles and responsibilities in those organizations.
tially conceptual. For example, digital interfaces in inpatient These include various efforts in contending with the digital
care systems are often clumsy; volumes of health data are health challenges and opportunities of the COVID-19 pan-
mostly sequestered, inaccessible, and difficult to aggregate demic. The discussion paper Digital Health COVID-19 Im-
in a meaningful and actionable way, in part due to the on- pact Assessment: Lessons Learned and Compelling Needs
going need for evolving data standards. In addition, digital was produced in parallel to and in coordination with this
tools and data are relatively ineffective in assisting clinicians work and serves as a use case of the key concepts present-
in better understanding patient and family preferences and ed here (Lee et al., 2022). In addition, the development of
circumstances that facilitate health progress outside of the this paper was informed by the National Academy of Medi-
clinic. The notion of digital tools that can be applied in wide- cine Leadership Consortium’s Digital Health Action Collab-
spread fashion to coordinate health care organizations and orative (DHAC) and DHAC’s prior work stewarding devel-
public health efforts to identify and engage those at particu- opment of the international statement on Digital Health and
lar risk from behavioral, social, and environmental public the Learning Health System, issued collectively in 2020 by
health risks remains rudimentary at best. The expansive vi- national academies of science and medicine of 14 countries
sion of real-time generation of evidence in a learning health (NASEM, 2020).
system that links datasets and analyzes them using artificial
intelligence and machine learning is nascent and limited to Digital Innovation and Medical Care
a few pilots. Digital technology has now been developed and applied to
Ongoing and accelerated progress must be made to fully every aspect of health and health care. Figure 1 groups the
realize the vision of a learning health system. In the digital various digital health tools into a dozen application arenas,
age, regardless of the specific barrier to the creation and but the individual applications number in the thousands.
support of individual and population health (e.g., COV- The authors see the potential in digital innovation in health
ID-19, staff burnout, challenging financial outlook, equity, care delivery in the following areas: advancing diagnosis
etc.), digital health can and should act as a “force multi- and treatment, ensuring care continuity, facilitating off-site
plier” of the interventions to combat these challenges. As patient management through telemedicine, partnering with
active participants in advancing prospects and practices in individuals to support self-management, and reducing error
digital health, the authors of this paper hope to: and waste in the delivery system.
• highlight the compelling possibilities and unresolved
challenges for advancing trustworthy digital technol- Advancing Diagnosis and Treatment
ogy for the benefit of all people at every stage of Research shows that a significant proportion of health
their lives; spending is attributed to chronic diseases, with individuals
• underscore the importance of ensuring that the ben- experiencing multiple comorbidities accounting for a dis-
efits are equally shared across society; proportionate share of expenditures (Buttorff et al., 2017).
• identify the structural, technical, and policy precon- Although additional research is necessary, a recent review
Page 2 Published June 27, 2022

The Promise of Digital Health: Then, Now, and the Future
FIGURE 1 | Evolving Applications of Digital Technology in Health and Health Care

SOURCE: National Academy of Medicine. 2019. Digital Health Action Collaborative, NAM Leadership Consortium: Collaboration
for a Value & Science-Driven Health System.
concluded that self-management as part of a treatment images, supporting potentially more accurate and timely
program for patients with chronic conditions has small-to- diagnosis and individualized treatment plans for various
moderate impacts on health behaviors, health outcomes, cancers and renal disease (Barisoni et al., 2020). Drug re-
and service utilization and should be an ongoing priority searchers and manufacturers are also leveraging various
in promoting population health (Allegrante et al., 2019). forms of AI for patient recruitment, virtual engagement, and
Thus, the market appetite and the necessity exist to facili- literature review, and using the technologies to assist in de-
tate diagnosis, reduce disease burden, and improve care tecting and refining pharmaceutical targets (Lamberti et al.,
for those who experience chronic disease. To address these 2019).
problems, innovators, software vendors, payers, and gov- Treatment decisions can be augmented by clinical deci-
ernment regulators are investing heavily in digital health so- sion support (CDS) systems and enriched with advanced
lutions for diagnosis and treatment, with particular attention analytics. An editorial in the New England Journal of Medi-
to high-need, high-cost populations (The Commonwealth cine succinctly summarized the challenge: “The complexity
Fund, 2016). of medicine now exceeds the capacity of the human mind”
An example of a diagnostic tool enhanced by digital (Eddy, 1982). While AI-based systems are currently unable
health includes smartphone-based photoplethysmography to discern a grimace, notice sweating, or hear a tremor in
(using a smartphone camera to capture video from the sub- a patient’s voice—skills at which humans excel—these sys-
ject’s index fingertip), combined with a deep neural net- tems offer the unique opportunity to augment clinician per-
work, a form of artificial intelligence (AI), to detect diabetes formance by creating order and transforming vast amounts
(Avram et al., 2020). While not widely adopted, such tools of mostly unstructured data into clinically actionable infor-
could be used for self-administered, low-cost, widespread mation to support optimal care. This field, although nascent,
screening. AI is also used in radiology and pathology to is rapidly advancing. For example, AI has been used to im-
augment human interpretation of diagnostic (e.g., ocular, prove the speed of prediction and diagnosis of sepsis (Goh
x-ray, or magnetic resonance imaging) and pathology slide et al., 2021). Integrated with the care delivery workflow,
NAM.edu/Perspectives Page 3
DISCUSSION PAPER
these technologies could identify patterns, form linkages and the health system writ large. To facilitate data interop-
between disparate data sources, and suggest treatment op- erability, the U.S. health system must expand embedded,
tions for clinicians to review. In addition, AI-powered CDS open-source interoperability beyond nationally regulated
systems might offer opportunities for improving efficiency technologies like EHRs.
and mitigating clinician burnout, another potential down- Research has documented the potential for Health Infor-
stream benefit. mation Exchange (HIE) and interoperability to improve care
coordination and reduce costs (Walker et al., 2005), and
Ensuring Care Continuity will likely also benefit public health reporting. An example
Even the most sophisticated digital diagnostics will have of HIE use to support care coordination is the delivery of
little impact on clinical outcomes if they are implemented in near real-time dashboards to primary care and substance
a fragmented health care ecosystem. Regulations promul- use disorder providers about inpatient and emergency de-
gated by the 21st Century Cures Act Final Rule (Cures Act) partment admissions and discharges for their patient panels,
have the potential to address this shortcoming by promoting supporting post-discharge care coordination (HealthIT.gov,
seamless interoperability and supporting increased control 2017). Still, patients and providers will struggle to realize
for the individual regarding their health data (HealthIT.gov, these benefits at scale as the existing reimbursement system
2020). The Cures Act addresses foundational standards, in- continues to disincentivize care coordination, which results
cluding technical, syntactic, and semantic issues surrounding in duplicative service utilization. The ongoing transition to
health data interoperability and prioritizes ensuring patients value-based payment can support the realignment of finan-
have choices when managing their own health data. Further cial incentives and serve as a significant driver for expand-
complexities associated with a robust trust framework, data ing interoperability (Biel et al., 2019).
accuracy, identity matching, and privacy protections of in- In this regard, banking, which provides ubiquitous, near
dividual data managed by noncovered entities will likewise real-time, standardized access to account information glob-
be critical to confront. ally, provides lessons about industry-wide information ex-
Significant progress on interoperability has occurred over change that might be adopted in health care. The Society for
the past decade with the implementation of foundational Worldwide Interbank Financial Communications (SWIFT)
data standards such as Health Level 7 Fast Healthcare In- established a financial transaction messaging system in the
teroperability Resources (FHIR) (HL7 International, n.d.), 1970s with a focus on essential transactions, a strong busi-
SNOMED (SNOMED International, n.d.), RxNorm (NLM, ness case for participation, and an industry-supported over-
2022), and the United States Core Data for Interoperability sight organization (Glaser, 2019). The Office of the National
(USCDI) (HealthIT.gov, n.d.). Still, the broad interoperability Coordinator for Health IT (ONC), through the Trusted Ex-
of health care data platforms is incomplete in many settings change Framework and Common Agreement (TEFCA), has
due to incomplete record availability, lack of terminology made inroads toward this vision with the formal recognition
standards, and concern about bidirectional incorporation of of an industry-supported oversight organization through the
data between health systems using different EHRs. In a 2019 Recognized Coordinating Entity (RCE), which was awarded
study of primary care physicians in high-income countries, to the Sequoia Project in 2019 (HealthIT.gov, 2022).
the Commonwealth Fund (2019) found that just over 50%
of American primary care physicians surveyed were able to Facilitating Off-Site Patient Management through
electronically exchange data with physicians outside of their Telemedicine
practice. In addition, since health is not primarily produced Digital tools that collect data and support interventions out-
by health care, interoperability with data outside of EHRs side the clinical setting offer meaningful opportunities to
may add to a holistic picture of an individual and support identify risks and engage patients. Consumer-facing apps
continuity of care. However, this level of interoperability is and clinical monitors that actively or passively collect data
nascent, as demonstrated by a recent review of data ex- can also serve as an early warning system for prevention
change capacity of wearables, which found limited ability and disease management. During the COVID-19 pandem-
to transfer data from mobile monitoring systems into medi- ic, digital contact tracing apps provided patients with no-
cal records (Muzney et al., 2019). The power of EHR sys- tifications about potential exposure to COVID-19. Beyond
tems to capture and organize clinical data allows for rapid COVID-19, some tools generate warnings to individuals or
cycle learning and organizational agility, but barriers—both caregivers regarding changes in environmental risks, such
technical and economic—to transmitting non-native data as pollen or air pollution alerts, while other platforms gen-
into the EHR limit the comprehensive view of individuals erate alerts to patients, families, and providers in the event
and populations needed to transform health care delivery of disease exacerbation. Additionally, while not widely ac-

ceptable or accessible by all populations, use of remote pa- force individuals to choose between needed health care
tient monitoring (RPM) tools increased during the COVID-19 and medication and other household expenses, can result
pandemic. RPM enables clinicians to assess symptoms for in not taking medicines as prescribed, including pill splitting
patients at home with mild cases of COVID-19 and observe and dose skipping (Kearny et al., 2021). These barriers of-
non-COVID-19-related health outcomes in the context of ten lead to clinical inertia and are amplified by structural
daily living for patients with chronic conditions (e.g., Blue- racism, furthering health disparities among underresourced
tooth scales for patients with congestive heart failure, con- communities.
nected blood pressure cuffs for patients with hypertension). By applying digital tools successfully used in other indus-
Digital tools have also expanded care delivery for pro- tries, such as consumer-directed, preference-based sched-
viders beyond the hospital or exam room. A 2020 analysis uling; personalized recommendations; and regular text
found that virtual urgent care visits could reduce the need communications, the health care system may be positioned
for emergency room care by approximately 20%, and to develop a more robust partnership between individuals,
20% of all office care, outpatient, and home health services families, and providers. Data and digital health tools serve
could be delivered virtually or near-virtually (Bestsennyy, et as a bonding agent in their shared understanding of the in-
al., 2020; Cigna Newsroom, n.d.). Non-acute care visits dividual’s state of health and a shared health management
for many conditions were implemented virtually during the plan. Individuals and families have grown accustomed to
COVID-19 pandemic to reduce risk of exposure for patients mobile and online tools in other aspects of their lives, such
and providers. Even with the sharp decline in telehealth in as airline booking, car services, and banking. Developing
2021—after the steep rise associated with COVID-19 in a robust partnership between individuals, families, and pro-
2020—a review by a large payer in 2022 supported the viders requires further adoption of systems that function the
value of virtual care (Cigna, 2022). same way that these other tools do, offering patient-centric,
Even acute care can be delivered outside the health care easy, and secure two-way communication for appointment
delivery setting, as witnessed during the COVID-19 pan- booking, self-check-in, and feedback surveys. Such tools
demic when severely ill patients occupied many hospital can and should be seamlessly interoperable within health
beds (Heller et al., 2020). Virtual intensive care units can systems workflows. While patient portals support many of
deliver remote 24/7 monitoring of patients by intensivists these functions, adoption among adults in the U.S. is below
who can manage patients in multiple locations, allowing 50% (HINTS, 2018). Strategies should acknowledge user
patients to get intensive care unit-level care in community comfort with technology and offer multiple communication
hospitals. modes, including text messaging, audio, and video, de-
pending on the user preference (Zachrison et al., 2021).
Partnering with Individuals to Support Self-Man- These approaches also need to consider form and frequen-
agement cy of communication to ensure maximum engagement and
Given that most chronic disease management occurs out- understanding.
side of the traditional health care setting, partnering with
individuals so that they can fully engage in their own care Reducing Error and Waste in the Delivery System
and meeting people where they are physically and mentally Extensive research indicates that health care resources are
is essential to achieving better health outcomes, improving inappropriately allocated within the current system. Waste
quality of life, and reducing health care spending (Allegran- has been shown to carry consequences for quality outcomes
te et al., 2019). However, meeting individuals on their own and patient safety (e.g., medical errors and delays) and
terms may present multiple challenges to both individuals economic efficiency (e.g., unnecessary spending) (Shrank
and the delivery system. Basic knowledge gaps about anat- et al., 2019). In the context of safety, since the Institute of
omy and physiology are worsened by issues of language Medicine’s (IOM) report titled To Err Is Human: Building a
fluency, health and reading literacy, numeracy, conflicting Safer Health System was published in 2000, health care
cultural beliefs, and limitations in cognitive capacity. These providers have made progress in reducing harm in hospi-
same challenges may be further exacerbated by poor medi- tal settings, but that progress varies widely (IOM, 2000).
cation tolerance and complex clinical care plans, including Equally troubling is the inability to accurately measure the
polytherapy and polypharmacy (Settineri et al., 2019). Ac- harm associated with the lack of timely, standardized, and
cess issues, including distance from the delivery system for accurate information movement across systems (Bates and
rural residents, lack of transportation, and difficulty taking Singh, 2018). As identified in a 2010 report from the IOM ti-
time away from work, all affect attendance at provider visits tled The Healthcare Imperative: Lowering Costs and Improv-
and can result in delays in seeking care. Financial barriers ing Outcomes, disruptive innovation has been foundational
DISCUSSION PAPER
across sectors to reduce waste and increase efficiency, and bulatory care and inpatient settings. Barcoding has been
its use as a strategy to address these issues in health care is widely used in hospital pharmacies for over a decade, re-
essential (IOM, 2010). sulting in a reduction in adverse drug events (Boyde and
The digitization of health data has long been considered Chaffee, 2019). Repetitive tasks such as scheduling, billing,
the foundation for patient safety, operational efficiency, capacity coordination, and asset management are amena-
and quality of care. It was also a driving force behind the ble to automation, optimizing use of system resources and
Health Information Technology for Economic and Clinical creating a frictionless experience for patients. For example,
Health (HITECH) Act, which incentivized the adoption of health care organizations can emulate the airline industry
EHRs (IOM, 2004; HealthIT.gov, 2009). By 2017, 80% in maximizing automation and self-service functionality in
of office-based physicians and 96% of non-federal acute scheduling while addressing customer demand, service
care hospitals had adopted certified EHRs (Health IT Dash- supply, and equipment needs (ONC, 2020).
board, 2016). Multiple studies have documented improve-
ments in care quality (Atasoy et al., 2019; Buntin et al., Digital Innovation and Population Health
2011). However, in a recent survey of over 5,000 physi- Figure 1 also identifies various tools applicable to improving
cians across specialties, perceived EHR usability was poor. population health and drivers of health that are upstream
Results showed a “dose-response relationship between EHR from medical care—e.g., geospatial and environmental
usability and physician burnout”, which is negatively asso- sensors, personal health devices, and knowledge genera-
ciated with patient safety (Melnick et al., 2020; Panagioti et tors and integrators. The importance of using digital tools in
al., 2018). However, patient safety is improved regardless helping to integrate critical social services into care delivery
of physician experience (Tanner et al., 2015). In addition, has been clearly demonstrated by the nation’s experience
ongoing opportunities to better integrate clinical and ad- with COVID-19 and the disproportionate impacts on com-
ministrative functions, streamline documentation (e.g., via munities of color and other economically disadvantaged
voice technologies), automate quality metrics reporting, and underresourced populations (Isasi et al., 2021; Health
and embed AI and advanced CDS systems represent mean- IT Dashboard, 2016). Innovations in digital health hold the
ingful advancements that EHR vendors are pursuing as these potential to help identify and address many of the barriers to
platforms mature—either as new functionality within their achieving the vision of a healthy society. When thoughtfully
platforms or by connecting to external third-party vendors, designed, equitably deployed, and effectively used, digital
creating a “both/and” approach to maximizing efficiency. health tools have the potential to improve the identification,
When it is clinically appropriate to address health con- measurement, and modification of the root sources of illness,
cerns without an exam or with good quality video or still health, and well-being. Without the precise analytic infor-
images, telehealth can reduce delays for specialty consulta- mation possible through a robust digital infrastructure, the
tions and primary care, as the constraints of a shared physi- nation will not be able to accelerate the identification and
cal space for an exam are eliminated. Advanced analytics engagement of the causes and consequences of structural
can also reduce waste by helping health care professionals racism, which plays such a perverse and pervasive role in
work at the top of their licenses. Advanced analytics can the health disparities of far too many Americans.
improve clinical risk stratification, allowing less skilled care As digital health tools become increasingly sophisti-
team members to address the needs of patients who require cated and capable of capturing social, behavioral, and
minimal care. Higher skilled care team members are freed environmental determinants of health, clinicians and care-
up to spend additional time with patients with complex med- givers can learn more about the individual in the context
ical needs, resulting in the delivery of the right care, to the of their daily lives, including individual preferences, values,
right patients, at the right time, in the right place, by the right interactions, and exposures, to deliver targeted preventive
clinical team members. Machine learning (ML) and natural and acute care and to restore health after illness. This digi-
language processing (NLP) algorithms have outperformed tally enabled health ecosystem has the potential to create
nursing staff and provided comparable levels of accuracy long-term partnerships between individuals and their care
to skilled physicians in assessing acuity risk in emergency teams that support healthy behaviors. Similarly, if thought-
departments (Ivanov et al., 2021). However, caution must fully designed, equitably deployed, and effectively used,
be used because algorithms can inadvertently perpetuate such digital health applications have the potential to help
significant bias (Tanner et al., 2015). prevent, mitigate, and reduce disparities in access and care
Process automation is another area of opportunity to use (Craig et al., 2020). In such a system, health information
digital health technology to improve efficiency in both am- flows freely within a trust-enabled and robust security and

privacy framework across both the health care industry and use, an improvement in symptom-free days for individuals,
nontraditional commercial entrants into the market. and a reduction in health care resource utilization (Mer-
chant et al., 2018; Barrett et al., 2013). Furthermore, “ag-
Digital Innovation and the Social Determinants of gregated data on inhaler use, combined with environmental
Health data, led to policy recommendations”, a community asthma
Kaiser Family Foundation defines the social determinants of notification system, community-wide improvements in asth-
health (SDoH) as “the conditions in which people are born, ma symptoms, and reductions in asthma-related emergency
grow, live, work and age that shape health,” with these department use (Barrett et al., 2018; Barrett et al., 2013).
conditions including “socioeconomic status, education, Consumer-facing tools also can provide smartphone alerts
neighborhood and physical environment, employment, and for heat or air pollution data at the neighborhood level,
social support networks, as well as access to health care” making public health efforts more efficient. The use of tele-
(Artiga and Hinton, 2018). Although approximately 15% of health and HIE can also support coordinated patient care
premature deaths are attributed to SDoH, these upstream during natural disasters. Of course, none of these tools ad-
drivers of health have largely been considered out of scope dress the root causes of these environmental problems—for
and not yet routinely addressed by providers or health care example, a person might be able to know that their drinking
systems (McGinnis et al., 2002). water contains lead, but the tool cannot assist in solving the
For digital technology to have a meaningful effect on underlying drinking water problem. These SDoH must be
SDoH, information about nonmedical factors and services addressed at the root level to realize improved health and
must be better collected and integrated into mobile apps well-being for all.
and standardized, aggregated, and integrated into EHRs to
promote trust and ensure secure and private management. Digital Innovation and Health Behavior
Digital tools could play a role in screening and identifying Digital health technologies are also developing new use
SDoH factors that impact a patient, alerting the provider to cases to address various environmental factors, including air
discuss them with the patient at the next visit, and connecting pollution and climate change. Digital inhaler sensors have
the patient with relevant community services. While existing been used to monitor when and where patients with asthma
digital health tools are already capable of supporting the used medications and needed adjustments to treatment
collection, exchange, and integration of SDoH to support plans and are associated with a reduction in rescue inhaler
risk stratification and shared care planning, the benefits of use, an improvement in symptom-free days for individuals,
these tools have been limited by inconsistent use across care and a reduction in health care resource utilization (Mer-
delivery settings and the significant risk of algorithmic bias chant et al., 2018; Barrett et al., 2013). Furthermore, “ag-
(Meyer et al., 2020; Lindau, 2019). For example, schedul- gregated data on inhaler use, combined with environmental
ing algorithms designed to identify patients who frequently data, led to policy recommendations”, a community asthma
miss appointments may both stigmatize people of lower so- notification system, community-wide improvements in asth-
cioeconomic status and distort the real issues. Many of these ma symptoms, and reductions in asthma-related emergency
“no shows” cannot afford childcare or to leave work for a department use (Barrett et al., 2018; Barrett et al., 2013).
medical appointment, or they may have health problems Consumer-facing tools also can provide smartphone alerts
that cause disability or reduced cognitive function, caus- for heat or air pollution data at the neighborhood level,
ing them to miss appointments (Murray et al., 2020). Un- making public health efforts more efficient. The use of tele-
derstanding and intervening on SDoH and systems factors health and HIE can also support coordinated patient care
could reduce missed appointments, helping patients to get during natural disasters. Of course, none of these tools ad-
needed care and reducing lost care capacity for the system. dress the root causes of these environmental problems—for
These issues highlight the need for transparency in data col- example, a person might be able to know that their drinking
lection and encoding and the criticality of proactive action water contains lead, but the tool cannot assist in solving the
to mitigate unintended consequences and biases when de- underlying drinking water problem. These SDoH must be
veloping algorithms. addressed at the root level to realize improved health and
Digital health technologies are also developing new use well-being for all.
cases to address various environmental factors, including air Although consumer demand for interventions that support
pollution and climate change. Digital inhaler sensors have behavior change is high, and successes have been evident
been used to monitor when and where patients with asthma in areas such as tobacco use and the consumption of foods
used medications and needed adjustments to treatment high in saturated fat, the complexity of behavioral interven-
plans and are associated with a reduction in rescue inhaler tions can be vexing. Consider the case of weight manage-
DISCUSSION PAPER
ment programs. The overall weight loss market in the U.S. in Digital Innovation, Genomics, and Precision
2020 was estimated at $71 billion, yet many programs elicit Health
only marginal and temporary changes in weight, with par- Digital technologies are accelerating the “genomics revolu-
ticipants often experiencing weight regain (LaRosa, 2020; tion”—advances in understanding the health implications of
Hall and Kahan, 2018). As such, interest in digital and virtual structural and functional variations in the human genome.
weight loss programs is mounting as an alternative (LaRosa, These are often discussed in terms of augmented abilities to
2020). However, while several well-controlled studies have target individual medical interventions more precisely. While
demonstrated improved clinical outcomes when incorporat- this is certainly an important likelihood, broader scale ben-
ing digital tools relative to usual care, most applications in efits in terms of reduced mortality and morbidity are likely
the consumer marketplace are not supported by evidence, to result from “precision public health”—the ability to better
nor are they produced by subject matter experts in health identify populations at greater risk from certain character-
behavior change (Gordon et al., 2020; Pagoto and Ben- istics or exposures and implement protective interventions.
nett, 2020; Steinmetz et al., 2020). Whole genome sequencing and digitally enabled risk
This example illustrates some of the broader challenges scores generated by such sequencing will help identify in-
and opportunities for digital tools to support self-manage- dividuals and groups at risk for common health conditions
ment of individual health behaviors. In their ideal form, in their earliest stages. These data can be used to support
evidence-based digital health tools that focus on health mitigation strategies such as behavior change, medication
behavior can improve self-awareness, provide on-demand use, or early screening to decrease the risk of sequelae from
health information and education, support improved self- a genetic disease or gene variants. Examples of existing
efficacy, and promote accountability with social support consumer-facing mobile health apps today draw from sev-
networks, health coaches, and providers. The resulting data eral data sources and partnerships, including self-reported
can also be analyzed to identify behavioral risk factors that family history data, laboratory results from personal genet-
contribute to chronic disease, resulting in real-time, person- ics companies, and collaboration with providers, payers, or
alized feedback and messaging to support health behavior employers (Tung et al., 2018). The ongoing integration of
change in a way that is more compelling than traditional genetic or genomic data and clinical histories, accelerated
patient education (Shegog et al., 2020; Barrett et al., 2013). by emerging AI and ML technologies, increases the feasi-
Similarly, these data can be aggregated at the community bility of leveraging precision medicine into clinical practice
level to more accurately measure the health behaviors and (Luchini et al., 2022). For example, AI is currently used in
activities of populations, supporting resource allocation and oncology, including FDA-approved AI used in support of
data-driven public health decision making at the local level care for breast, lung, and prostate cancers (Luchini et al.,
(Barrett et al., 2013). 2022). Advanced computational analytics used on such
Digital health tools designed to support adherence to datasets could ultimately be employed to deliver near real-
treatment plans also present an important opportunity. Con- time feedback to individuals to promote health using a voice
nected self-monitoring tools (e.g., glucometers), wearables, assistant, much like a digital health coach (Topol, 2019).
digital inhaler sensors, and SMS messages and reminder
systems have shown promise in patients with a variety of Digital Innovation and the Learning Health
conditions, including epilepsy, asthma, chronic obstructive System
pulmonary disease (COPD), diabetes, depression, and hy- The application of digital technologies at scale serves as
pertension (De Keyser et al., 2020; Kaye et al., 2020; An- the nervous system for the continuously learning health care
derson et al., 2019; Shan et al., 2019; Patel et al., 2013). For system: “one in which science, informatics, incentives, and
example, objective, passive data about adherence to asth- culture are aligned for continuous improvement, innovation,
ma medication treatment plans identified issues with medi- and equity—with best practices and discovery seamlessly
cation-taking technique errors and presented an opportu- embedded in the delivery process, individuals and fami-
nity for intervention and education (Anderson et al., 2019). lies active participants in all elements, and new knowledge
Interestingly, patients with asthma and COPD who received generated as an integral by-product of the delivery experi-
digital support (reminders for missed medication doses and ence” (NAM, 2020). Digital health will serve a critical role,
education) increased their medication adherence during the and its promise must be fully leveraged. Effectively applied,
early months of COVID-19 (Kaye et al., 2020). digital health tools have the potential to catalyze progress
on each of the key principles for a digitally facilitated learn-
ing health system, presented below in Box 1.

BOX 1 | Core Principles for Stewards of the Digital Health Infrastructure and Data
Personal: Discretion on control and use of personal data resides with the individual or their designee.
Safe: Data stewardship protocols safeguard against use resulting in personal harm.
Effective: Data are collected and maintained according to validated stewardship protocols.
Equitable: Data systems are designed to identify and counter bias or disparities.
Efficient: Every digital equipment acquisition or service license enhances health system interoperability.
Accessible: Data are available when and where needed for decision-making.
Measurable: Digital health performance is continuously monitored for accuracy and interoperability.
Transparent: Personal data sources and uses are clearly indicated, including timing and context.
Adaptive: Data strategies are regularly calibrated to ensure continuity, currency, and utility.
Secure: Data sharing protocols are considered secure by users.
SOURCE: National Academy of Medicine Leadership Consortium: Collaboration for a Learning Health System. n.d.
Digital Health Action Collaborative Strategic Framework.
Leveraging Big Data for Knowledge Generation and maximize research output while protecting individual
Much of the data collected in clinical care or recorded in agency and privacy (Baker et al., 2016; Ideas for Change,
consumer apps are available for further research and learn- 2016; MiDATA, n.d.). If successful, this digitally enhanced
ing. Currently, the broader application of available health approach to research could allow multiple stakeholders, in-
data is more likely to be used in service of product devel- cluding professional societies, health care providers, patient
opment rather than for learning, discovery, or continuous advocacy groups, individuals, families, legal experts, medi-
improvement of the health of individuals, families, or popu- cal administrators, the private sector, and governments, to
lations. There is an unrealized opportunity to share, aggre- share data, experiences, and research priorities.
gate, and analyze that data in alignment with the goals of a
learning health system while also protecting and tightening Leveraging Big Data for Population-Level and
the processes and procedures for unwarranted access to Public Health Insights
and use of personal data and inadvertent sharing of sen- Fully realizing the benefit of vast datasets with informa-
sitive data, including medical records, via third-party con- tion collected in near real time across the health continuum
sumer apps. promises to improve population and public health. Some
The investment, innovation, and amassing of data pres- noteworthy examples of these public datasets include the
ent important opportunities to affect not just health and the National Patient-Centered Clinical Research Network
health care delivery system but also knowledge develop- (PCORnet) (PCORNet, n.d.), the Research Data Assistance
ment in a learning health system. If appropriately managed Center (ResDAC) for CMS data (ResDAC, 2022), the Ob-
and analyzed, datasets that incorporate structured and servational Health Data Sciences and Informatics program
unstructured clinical data, SDoH information, genomics, (OHDSI) (OHDSI, 2022). This promise includes the active
digital phenotype data collected from wearables, and other and passive collection of real-time data from patients’ daily
data can make it possible to change baseline understand- living activities, gathered in clinical systems and payer sys-
ings of health and disease (Engelhard et al., 2020; Jain et tems and the analysis of that data to make well-reasoned
al., 2015). Statistical tools and techniques, including AI and decisions using standard analytics and AI/ML (Singhal et
ML, can be used to develop dataset assessment tools and al., 2020; Bughin et al., 2017). To apply analytics tools to
to support evolving research designs that meld traditional health care will require significant investment; fortunately,
randomized controlled trials (RCTs) with observational stud- the Cures Act authorized $1.5 billion over 10 years to
ies. Similarly, analytical models can be applied across at- support the NIH’s All of Us Research Program, which is
risk populations to ensure equity in opportunities to create designed to build and make available to researchers a se-
health and treat disease. Development of virtual health data cure and expansive database, including EHR, survey, and
trusts, with shared governance and individuals controlling biometrics data of one million people to support medical
and contributing their data to support scientific discovery, discovery (NIH, 2020). While NIH did not explicitly create
present an important opportunity to distribute the costs the All of Us program for AI/ML, as a by-product of the
DISCUSSION PAPER
FIGURE 2 | Infrastructure Requirements for Progress in Digital Health

SOURCE: National Academy of Medicine. 2019. Digital Health Action Collaborative, NAM Leadership Consortium: Collaboration
for a Value & Science-Driven Health System.
program, researchers will have access to new datasets and learning health system and accelerates the identification
platforms upon which they can train their models. and elimination of wide-scale disparities in individual, lo-
As health systems, payers, and community organizations cal, regional, and global health care. As individuals gain
collaborate and share data to serve specific populations, more access to their health data via application program-
public health agencies are positioned to seamlessly collect ming interfaces (APIs), and as providers use these data for
data and apply advanced analytics for health surveillance critical clinical decision making using AI/ML, it is essential
and community intervention. Interoperability links health to consider several foundational infrastructure requirements.
systems, community agencies, geographical information Figure 2 presents the essential infrastructure requirements for
systems, and public health agencies to address medical, progress in digital health. While there has been some prog-
environmental, and SDoH (Buckeridge, 2020). Interoper- ress, opportunities remain in each interrelated component.
ability can also create opportunities, via big data and preci- Each area must be carefully reviewed and addressed to
sion public health, to tailor interventions to subpopulations, fully establish the framework to allow the benefits of digital
which will help ensure equity (Buckeridge, 2020). During health to be fully realized. Of particular interest for priority
COVID-19, the public health sector is experiencing an op- action are individual access and engagement, equity and
portunity to test a variety of new precision public health ethics, privacy and identifier protocols, cybersecurity, data
tools, including the use of cell phone location data, activ- quality and reliability, data storage, sharing, and steward-
ity trackers, and sewage data to intervene early to identify ship, interoperability, AI/ML, and workforce.
outbreaks and to limit morbidity and mortality (Rasmussen
et al., 2020). Individual Access and Engagement and Equity
and Ethics
Requirements for the Digital Health Infra- To ensure digitally facilitated health for all, access to digital
structure health writ large, supported by widespread broadband in-
Digital technology serves as the nervous system for the ternet access, is essential across all economic strata and all

regions of the U.S. Unfortunately, while COVID-19 resulted actors. Some apps serve as a core communication device
in the practical and essential application of telehealth, key between individuals and their physicians and sit squarely
gaps in consumer access to such technologies—“the digital within HIPAA. In contrast, other tools and vendors are un-
divide”—were also exposed. regulated by HIPAA, creating uneven protection and confu-
Equity in available broadband access will spur growing sion for consumers. The expansion of HIPAA to redefine and
consumerism and engagement in health and health care. protect health information outside of covered entities could
The public has routine exposure to digitally facilitated con- mitigate risks to individuals.
venience, agency, transparency, and privacy based on Another critical area requiring progress in support of
their experience with other industries and now expects the digitally facilitated health is accurately matching individu-
same from the health care ecosystem (Accenture, 2019). As als across systems (The Pew Charitable Trusts, 2018). A
the understanding of what creates health and well-being unique national patient identifier was envisioned as a foun-
grows, it is imperative to engage patients, families, and dational element of HIPAA, but privacy and security con-
communities in the design of new structures, processes, and cerns prevented the enactment of necessary regulatory ac-
solutions to support health and well-being. It is also essential tion. Promulgating such regulations remains a valuable aim
to address systemic racism and institutional health inequities to support efficient, accurate matching. In 2021, the ONC
and disparities within the U.S. when designing these new advanced efforts to accurately match patient data across
structures, processes, and solutions (Feagin and Bennefield, systems with Project US@, which was established “to de-
2014). These steps are necessary to mitigate the risk that velop a unified, cross-standards, health care industry-wide
new technologies will deepen the existing digital divide or specification for representing patient addresses to improve
perpetuate historical mistrust in the health system. patient matching” (HHS, 2021). Correctly matching an in-
In addition, it will be important to translate what is learned dividual’s data across organizations (with sufficient gold
through the collection of digital health data writ large (e.g., standard matches that allow for appropriate algorithm de-
better insight into environmental determinants of health, Bar- velopment) remains an essential component for the learning
rett et al., 2013) into local and national policies to make health system to support the right care for the right person
these learnings applicable at the individual and population today and to support the use of AI and research to ensure
level. These improved policies should, in turn, translate into the best outcomes for tomorrow.
community improvements (e.g., urban planning decisions
about not placing schools next to freeways, informing na- Cybersecurity
tional air quality standards with research into the associa- Cybersecurity and privacy concerns are major obstacles to
tion between air pollutants and respiratory symptoms) to im- digital health adoption, continue to erode patient trust, and
prove the health of those who live in that community. reinforce health systems’ reluctance to share data. Psycho-
logical resistance, the risk of ubiquity of data, consequences
Privacy and Identifier Protocols of a breach, and patchwork of local and national privacy
The opportunity to share, aggregate, and analyze health protections—or lack thereof—have created barriers to the
data to improve individual health and to advance the learn- use of pioneering, forward-looking digital health tools, and
ing health system is significant, as is the risk of loss of privacy as such, cybersecurity must not be an afterthought.
for individuals sharing their most sensitive data via third-par- These critical challenges require technologic, governance,
ty consumer apps. Consumers have a limited but growing and legal protocols. A public-private partnership is neces-
understanding of the risks (including loss of privacy) and sary to develop a superstructure framework to ensure the
benefits of sharing their health data and express a range safety, security, and privacy of digital health architecture.
of views about sharing health information. For example, a As noted earlier, the cybersecurity framework produced by
2019 focus group on consumers’ perception of interoper- the National Institute of Standards and Technology (NIST)
ability found that “participants overwhelmingly supported provides such guidance (NIST, 2018). Transparency and
greater access to data both for health care providers and consent for consumers and patients regarding data shar-
for themselves” (The Pew Charitable Trusts, 2020). ing, agency, and privacy within and across platforms and
In the intervening decades since the enactment of the stakeholders—including those not covered by HIPAA—must
Health Insurance Portability and Accountability Act of 1996 be simplified and standardized, including understandable
(HIPAA), health data systems have grown exponentially. A consent forms and the extension of HIPAA protections to
new industry of health-related applications was launched, currently noncovered entities like third-party app vendors.
giving individuals the ability to readily share their most pri- In addition, privacy and security risks with big data and AI
vate data with a variety of health sector and commercial require special attention to avoid intentional corruption of
DISCUSSION PAPER
AI/ML training datasets (training data poisoning), use of AI and meaningful adoption of new consumer-facing digital
by attackers, or anti-privacy designs in digital health (Hartz- health tools.
og, 2018). Finally, decisions will need to be made about how data
are stored in cloud-based systems to advance the com-
Data Quality and Reliability, Storage, and Stew- mon good. Virtual data repositories must be structured and
ardship controlled to protect the integrity and privacy of the data
Foundational to digital health, the standards and curation through all aspects of data management - acquisition, stor-
protocols for data and information (e.g., Findability, Acces- age, access, maintenance and release. (NASEM, 2020) Si-
sibility, Interoperability, and Reusability [FAIR] principles), multaneously, computing power should migrate to the cloud
while best-practice, are not required by regulation. How- to support this future vision, as the cloud has both sophisti-
ever, such standards and protocols are required to achieve cated security and economies of scale. Cloud-based com-
uniform value between and among stakeholders. Data stan- puting will require a paradigm shift for organizations with
dards and stewardship guidelines and national cooperation on-premises systems.
are critical, while simultaneous attention must be paid to
“economic, legal, philosophical, and practical issues” relat- Interoperability
ing to health data (NASEM, 2020). In principle, the indi- Through the work of ONC, data and interoperability stan-
vidual, the source of health data, controls access to and use dards have grown increasingly sophisticated over the past
of the data derived from their health care and interaction decade. While more work is needed, early progress with
with digital platforms. In practice, the organization collecting HIEs, APIs, and EHR integration has yielded improvements in
and managing the data has differing custody and control of care coordination, and recent efforts during the COVID-19
the data, depending on the nature of the individual’s data pandemic demonstrated the capacity of HIEs to deliver
and regulations to which the data’s collector and custodian value by generating public health reporting (Dixon et al.,
is subjected. Differences among organizations concerning 2021).
“data access, control, and monetization” limit the potential In addition, interoperability standards need to extend be-
of digital health, and expanding structures for cooperation yond the current focus on EHRs. Existing interoperability of
and exchange are essential (NASEM, 2020). health care data systems neither adequately supports op-
The availability of patient portals in most EHR systems and timal longitudinal care delivery nor advances the nation’s
consumer-facing digital health tools and the data associated health needs. The COVID-19 pandemic illuminated the
with these applications represent a meaningful opportunity needs and opportunities for digital health and transforma-
to improve patient care. However, significant challenges, tive preparedness and response capacity. The rapid pace
including the digital divide, issues of systemic racism, data of the pandemic’s spread emphasized the need for a rapid
curation, integration into the care setting, and data sharing learning system that relies on capturing, organizing, sharing,
for research, impede progress toward realizing improved and analyzing large amounts of data digitally across public
patient care. health, research, and clinical systems. An effective response
Further, the strong drive to innovate and rapidly market to public health crises is highly dependent on interoperable
mHealth tools has led to product development outpacing data, without which there is an inability to understand what
the capacity of regulators to establish standards and com- is needed in terms of resources and capacity and to under-
municate clear guidance to various stakeholders, including stand the impact of interventions. While data was critical for
consumers and payers. These unclear standards and lack of forecasting and coordination, its collection, sharing, and ag-
regulatory guidance and oversight have created a market- gregation were, at times, chaotic and burdensome for clini-
place where promising digital health solutions that provide cians and administrators.
superior quality, impact, and value are difficult to distinguish The post-COVID-19 era can help ensure the interoper-
from poor quality innovations and work to the disadvantage ability of all mediums of digital recordkeeping used to sup-
of rigorously studied digital health products . For example, port health and deliver health care services, including labs,
emerging scientific evidence indicates that some RPM de- certified EHRs, home-grown EHRs, digital devices, consumer
vices can predict five-year mortality in adults between 50 electronics with health features, and databases to support
and 85 years and empower patients to better manage their research and public health. Before the pandemic, the Cen-
health and participate in health care (Halamka and Cerrato, ters for Disease Control and Prevention (CDC) had launched
2020). Clear standards and widespread rigorous review of a data modernization initiative to undergird disease surveil-
innovations, including the evaluation of technical design, lance systems. The Coronavirus Aid, Relief, and Economic
clinical value, and usability, could increase confidence in Security (CARES) Act allocated $500 million to the CDC

to implement a “modern, interoperable, and real-time pub- Workforce

lic health data and surveillance systems that will protect the To support digitally enabled health in a learning health sys-
American public” (CDC, 2020). tem, the workforce of the future will require a comprehen-
sive set of skills that are currently rarely seen. Besides basic
Artificial Intelligence and Machine Learning competency on core organizational applications (e.g., EHR
As the U.S. moves to value-based payment models, trans- functionality), clinicians, health system staff and manage-
parent and advanced analytics are needed to calculate ment, and vendors/innovators will all require at least basic
population risk, the foundation upon which medical budgets or conceptual knowledge of data management (collection,
are established in contracts between payers and providers. storing, normalizing), interoperability, basic statistics and
AI-driven predictive modeling and other sophisticated statis- data science, data governance and collaboration, ethics,
tical techniques can be used to identify subpopulations for process improvement, and implementation science. Finally,
intense care management to prevent inappropriate emer- diversity training is critical to all engaged in supporting digi-
gency room use or early intervention for an acute worsen- tally facilitated health in the learning health system and must
ing event to reduce hospital admissions. For example, in the mitigate disparities and build awareness among all parties—
inpatient environment, AI has been used to identify patients especially those individuals producing AI algorithms—to the
at risk of decompensation using data collected in the back- consequences of bias for vulnerable populations.
ground during clinical care (Lin et al., 2019). A recent lit- The technical workforce of the future will also need exper-
erature review of AI algorithms for sepsis models found the tise in user-centered design, which seeks to involve end users
models to be highly predictive but noted several issues with throughout the product development life cycle. The earliest
algorithmic standards (Deng et al., 2021). As digital health digital health care applications did not incorporate these
tools incorporate increasingly disparate data into predictive principles, and as such, use cases were limited to the auto-
models using various AI techniques, standard outcome and mation of paper processes rather than the reimagination of
data definitions, bias in training datasets and final models, care delivery and payment. This issue remains a problem to-
and frequently updated algorithms must be considered. day, as evidenced by burnout and frustration among provid-
Harnessing AI will depend on coherent data architecture ers using EHRs (Melnick et al., 2020). In seeking to achieve
and diverse training datasets, which are large, sampled better health, better care quality, lower costs, and greater
adequately, and represent subgroups adequately (e.g., by satisfaction among individuals and providers, user-centered
gender, race, age, socioeconomic status). The Food and design will be an essential ingredient of any infrastructure
Drug Administration (FDA) has released guiding principles strategy. Particular attention to culturally appropriate design
for “Good Machine Learning Practice for Medical Device and addressing the needs of historically underrepresented
Development,” which are practical and should be consid- populations has shown early positive effects when deliver-
ered when embarking upon model development (FDA, n.d.). ing interventions to populations in need and is another criti-
The regulatory framework for AI as a medical device is na- cal issue when ensuring that unintentional bias does not fur-
scent and must address certification of constantly changing ther the digital divide (Schueller et al., 2019).
algorithms and maintenance of accountability of vendors
to ensure reliable and valid processes. There are alterna- Stewarding Digital Innovation for Our Health
tive ways to regulate AI, including principles and standards Futures
developed by multi-stakeholder collaboration that can cre- To achieve the full potential of digital health, the health care
ate adaptable standards and guidelines. Components of industry and governmental leaders must collaborate, coop-
the European Union’s proposed rules governing AI might erate, and develop shared governance, creating a unified
be considered in the U.S. (EC, 2021). Additional standards digital health system architecture from independently func-
to consider include the International Medical Device Regu- tioning infrastructure building blocks.
lators Forum “Software as a Medical Device (SaMD): Ap- Key priorities must be identified and pursued within both
plication of Quality Management System,” FDA Center for the environmental and the technical contexts to achieve
Devices and Radiological Health “Software as a Medical the full potential of digital health. The key priorities in the
Device (SaMD): Clinical Evaluation Guidance for Industry,” environmental context include focusing on the individual,
and “Artificial Intelligence/Machine Learning (AI/ML)- embedding equity and transparency as first principles, re-
Based Software as a Medical Device (SaMD) Action Plan” forming health system payments in support of outcomes and
(IMDRF, 2015; IMDRF, 2017; FDA, 2021). These strategies value, and nurturing a learning health system ethos. From
could work in tandem with regulations updated for rapidly the technical perspective, the priorities include establishing
changing capacities.
DISCUSSION PAPER
seamless system interoperability, ensuring cybersecurity, benefits of digital health, racial bias in AI, and misuse of
and expanding algorithm validation and real-world testing. personal information in discriminatory practices. For digi-
tal health to improve health and well-being, a data-centric
Focusing on the Individual and patient-centric approach to developing and deploy-
Fully engaging individuals in their health and well-being ing these tools is essential, and data must reflect the diverse
through digital health, responding to public demand for communities and populations across the U.S. Here again,
participation in the growing digital health ecosystem, and the health system, researchers, and commercial ventures
balancing demand for consistent, transparent protections must address issues of mistrust with transparent, account-
for health data within and outside of the health care system able, and unbiased protections so that the benefits of digital
is a priority in achieving a fully realized future for digital health are shared equally across society.
health. Health data are intensely personal, and uninten-
tional or nefarious exposure of that data has the potential Reforming Health System Payments in Support of
to upend an individual’s life. Capturing the full potential of Outcomes and Value
digital health will require broad confidence in health sys- COVID-19 has provided a further reminder of the systemic
tems and commercial ventures to protect the individual from shortcomings of fee-for-service reimbursement, renew-
negative outcomes. ing the impetus for restructuring health care financing in
Transparent stewardship standards are needed to ensure America. Given the tremendous uptake of platforms such as
individual agency in using their data. A critical first step in telehealth and RPM during the COVID-19 pandemic, forth-
building trust in health data governance is a public dialogue coming payment reforms must account for the role of digital
about digital health—bringing together stakeholders into the health writ large in driving delivery system transformation.
policy process to address individual rights regarding data Policy makers will also need to address concerns that ex-
sharing, issues of consent, transparency, secondary uses of tending digital technologies will increase costs and the risk
data, common patient identifiers, consideration of health of fraud and abuse or otherwise negatively impact quality
data as a public good, and regulation of AI/ML. These dia- or provider-patient engagement.
logues will build comfort levels and demands for expanded Furthermore, the infrastructure improvements required to
applications while also maintaining safeguards against advance the digital functions of a learning health system
abuse and unintended consequences. (e.g., population health management, data and analytics
Central to the critical priorities for fully actualized digital for risk stratification) are often unfunded activities that would
health is the need to promote a sector-wide culture of trans- benefit from additional incentives and investments such as
parency and truthfulness without fear of retribution. Similar those that accompanied HITECH.
to how To Err Is Human called upon the health care indus- The financial benefits of payer and provider organizations
try to acknowledge where their practices were worsening must align with the health benefits of digital tools. This align-
health, a critical next step in advancing digital health is to ment will require data sharing from industry, evaluations
take definitive action to ensure that people feel comfortable from academia and regulators, and collaboration across
reporting errors without fear of punitive actions (Shrank et sectors to develop progressive payment structures across
al., 2019). payers that allow flexibility for innovation. The path forward
Beyond individual agency over health data, engaging for value-based payment will therefore require a renewed
consumers in their own health and health care via digital commitment to building trust and collaboration and aligning
platforms will require both systems developers and health incentives to balance the drive to innovate with stewardship
system leaders to include the customer’s voice in the devel- of cost, quality, outcomes, and safety.
opment, execution and evaluation of digital health tools and
platforms. A model for patient and family engagement in Nurturing a Learning Health System Ethos
digital health initiatives is in development in Canada and The vision of digitally facilitated health depends on a con-
could serve as a starting point for advancing a model in the tinuously learning health system and a dramatically short-
U.S. (Shen et al., 2021). ened interval between evidence generation, deployment to
the field, and incorporation into standard practice. There is
Embedding Equity and Transparency as First Prin- also a need to use real-world data (from wearables to am-
ciples bulatory care to robotics) to generate real-world evidence
The rapid development and application of digital health is that complements the results of randomized controlled tri-
also accompanied by the need for vigilance on equity and als, which often suffer from limited racial or socioeconomic
equality issues that include availability and access to the diversity in patient recruitment. Rapid cycle learning must

also be employed, as it will enable the necessary organiza- Ensuring Cybersecurity

tional agility to respond to an accelerated rate and nature The rapidly evolving landscape of cyberattacks highlights
of change that has become the norm. the urgent need for collaboration across the government,
Digital health tools must be well integrated into the health health organizations, and consumer-facing vendors to de-
care delivery system to enable the continuously learning velop consensus on security protocols and upgrade secu-
health system. With expanded data assets and improved in- rity infrastructure. Existing approaches such as multi-factor
teroperability, the delivery system has an opportunity to rei- authentication, intrusion detection monitoring, etc., must be
magine and recreate a care system that is culturally attuned, employed as we explore more advanced strategies, such
personalized, holistic, and comprehensive—one unlike as adopting blockchain technologies to share immutable re-
our current system, which consists of specialty, sector, and cords of transactions among network participants. Places to
system silos. New care models can be developed with an start could be expansion of HIPAA, national application of
understanding of disease and digital phenotypes and en- the California Consumer Protection Act, and a comprehen-
virotypes that will each have different treatment responses. sive privacy regime similar to the European Union’s Gen-
Advanced analytics are needed to create cohorts of similar eral Data Protection Regulation to protect all types of data
patients for more effective population health management deemed essential for health improvement.
to address the high prevalence of chronic disease and cre-
ate a feedback loop regarding outcomes and evidence- Expanding Algorithm Validation and Real-World
based treatment in the care delivery system. Testing
There is a clear need to invest in the capacity and coopera-
Establishing Seamless System Interoperability tion necessary to advance data science and AI. AI/ML and
Seamless connectivity and communication among health deep learning that apply transparent algorithms and deci-
care-related devices are essential prerequisites for promot- sion rule architecture to large, diverse databases present the
ing optimal health. Incompatible interfaces, corrupted data opportunity to develop increasingly precise insights for indi-
written between systems, or mismatched patient data have viduals and populations. Critical issues include explicit and
the potential to have dire consequences, requiring collec- implicit bias in the development and application of mod-
tive action to ensure adherence to standards to protect data eling, visualization, explainability, validity, and regulation
integrity. Technological advancement and national policies (The Lancet Digital Health, 2019; Buolamwini and Gebru,
have made possible the vision for a digital infrastructure 2018). A regulatory framework must address certification of
that can facilitate seamless interfaces and real-time interop- constantly changing algorithms and must hold vendors ac-
erability of devices and data streams. Released in March countable for valid and reliable processes and must include
2020, the Cures Act final rules set forth penalties for infor- codes of conduct and the development of “data science
mation blocking and expanded the access of individuals to tools, …pathways, agreements, and protocols for estab-
their health records by leveraging the FHIR specifications. lishing curated virtual health data trusts” (NASEM, 2020).
Such standards allow information to be shared and pro- The FDA’s AI/ML-based Software as a Medical Device
cessed consistently. In addition, there are several industry- (SaMD) Action Plan proposes such a framework and shares
led initiatives, such as the Integrating Healthcare Enterprise, valuable stakeholder feedback (FDA, 2021).
Argonaut Project, and others, aimed at promoting seamless The capacity to advance data science and AI is depen-
data exchange (IHE International, 2021; USF Morsani Col- dent on a highly skilled digital health workforce, and “the
lege of Medicine, 2021). As a promising indicator, many training challenge for leveraging digital health is vast—in
health systems have aligned organizational priorities to- health care, public health, and biomedical science” (NAS-
ward interoperability objectives. EM, 2020). In addition, as AI/ML is applied to CDS tools, it
Nonetheless, a great deal of work remains to achieve full is essential to address unintended bias in algorithm creation.
system interoperability, as semantic interoperability is lim- Tools designed for the clinical system and providers can
ited. Progress is uneven across the industry, with some health be evaluated on their impacts on health outcomes and costs,
systems being pioneers in real-time data sharing while oth- as well as their impact on both patient and provider satisfac-
ers are lagging. Moreover, interoperability continues to be tion. Real-world testing across unique health systems is re-
stunted by the systemic misalignment of incentives, competi- quired to understand impacts on usability, clinical workflow,
tive forces, and lack of coordination. provider burden, and staff time requirements that benefit
providers and patients. While time consuming, these pilots
are useful and must be tied to scaling opportunities if suc-
DISCUSSION PAPER
cessful. When relevant, testing of devices and AI-supported Below is a sampling of specific, actionable items for consid-
CDS must achieve FDA clearance. eration within this national blueprint, with specific reference
While the availability of digital health tools and associat- to the key priorities identified above.
ed data sharing has better positioned America to face CO- • A multi-stakeholder panel should be convened to
VID-19 and harness opportunities for long-term prepared- develop recommendations to meaningfully engage
ness and system resiliency, limitations such as the ability to the diverse individual consumers of health care in all
aggregate data have emerged. The full potential benefits health care sectors. This panel should follow the ad-
of these tools has not been realized, and the adoption and age “nothing about me without me” to ensure the pri-
application of digital health remains uneven and subject to orities of focusing on the individual and embedding
significant structural, technical, social, geographic, political, equity and transparency as a first principle.
and economic impediments, limiting the nation’s ability to be • A multi-stakeholder panel should be convened to
as nimble as needed in such crisis. establish use cases and support the development of
guidelines for applications laboratories to advance
Priority Near-Term Actions the learning health system ethos and expand algo-
The progress of digital technologies writ large is undisputed rithm validation and real-world testing.
and can be observed in the millions of enthusiastic viewers • Congress should promulgate rational, right-sized,
who use streaming video services; the countless customers risk-based regulation, standards, and frameworks
who shop online; and the growing number of consumers, to enable the seamless flow of data while protect-
patients, and clinicians who are embracing mobile health ing privacy and ensuring transparency and account-
apps, AI-enabled diagnostic aids, and many other CDS ability to advance system interoperability and cyber-
tools. However, while the predictive analytics used to sug- security, as well as focusing on the individual and
gest a person’s next favorite movie may be similar to the expanding ethical and effective algorithm develop-
analytics used to suggest a medical diagnosis or treatment ment, validation and real-world testing.
option, one key difference remains: when a streaming ser- • ONC should develop and implement a governance
vice recommends a new movie, viewers may find it helpful, infrastructure and policy framework regarding data,
annoying, or even amusing—not life threatening. When al- virtual health data trusts, privacy, and regulations to
gorithms are used to assist in the diagnosis of diabetic reti- advance focus on the individual, seamless system
nopathy or the recommendation of a therapeutic approach interoperability, and cybersecurity, working collab-
to sepsis, the stakes are much higher (Lin et al., 2019). To oratively with industry to ensure broad coverage of
fully realize the goal of health and well-being for every indi- these principles.
vidual, these concerns must be considered as all stakehold- • CMS should lead the effort to ensure sustainable
ers in the health care ecosystem make intense and sustained payment coverage to ensure equal access to digital
efforts to improve the capabilities of the health care deliv- health tools for all individuals and providers, regard-
ery system, impact SDoH, ensure equal benefit from digital less of private versus public payer source. In addi-
health, and establish an overarching architecture and gov- tion, CMS should significantly accelerate the move to
ernance framework that engages the public. value-based payments to support outcomes, innova-
ONC has made significant inroads toward an overarch- tion, and aligned incentives.
ing digital health blueprint for fully enabling digital health. • ONC should ensure the timely, full implementation of
Augmented by broader authority, continued progress on in- standards of structure, coding, security, and common
teragency collaboration, and robust public-private partner- APIs, as these standards are foundational for most
ships, this progress will ensure a digital health superstructure progress on digital health.
that:
• ensures equitable and ethical use of data; Envisioning and achieving a seamless, healthier future
• supports the collection, storage, protection, and through digital innovation will require a deeper investment
seamless sharing of accurate datasets and gener- in evidence-based research, more clinical and field studies,
ated insights in near-real time; and commitment from diverse stakeholders. But the poten-
• ensures the curation of that data into actionable intel- tial for rewards is enormous. Validated information, curated
ligence; and across the health data continuum and easily shared, can de-
• enables transformative advances in medical care liver insight at the point of care, easing provider burden and
and patient safety based on the actionable intelli- augmenting clinical reasoning skills. An “Internet of Things”
gence generated. in health care serves the public’s need for accurate health

advice, and a digital health ecosystem that provides high- With Technology, Crowdsourcing, Cross-Sector Col-
quality, personalized, equitable care to all who need it is laboration, And Policy. Health Affairs 37(4):525-534.
achievable and worthy of our best individual and collective https://doi.org/10.1377/hlthaff.2017.1315.
efforts. 10. Barrett, M. A., O. Humblet, R. A. Hiatt, and N. E. Adler.
2013. Big data and disease prevention: From quanti-
References fied self to quantified communities. Big Data 1(3):168-
1. Accenture. 2019. Today’s consumers reveal the future 175. http://doi.org/10.1089/big.2013.0027.
of healthcare. Available at: https://www.accenture. 11. Bates, D. W., and H. Singh. 2018. Two Decades Since To
com/us-en/insights/health/todays-consumers-re- Err Is Human: An Assessment of Progress And Emerging
veal-future-healthcare (accessed January 26, 2022). Priorities In Patient Safety. Health Affairs 37(11):1736-
2. Allegrante, J. P., M. T. Wells, and J. C. Peterson. 2019. 1743. https://doi.org/10.1377/hlthaff.2018.0738.
Interventions to Support Behavioral Self-management 12. Bestsennyy, O., G. Gilbert, and A. Harris. 2020. Tele-
of Chronic Diseases. Annual Review of Public Health health: A quarter-trillion-dollar post-COVID-19 real-
40:127-146. https://doi.org/10.1146/annurev-publ- ity? Available at: https://www.mckinsey.com/indus-
health-040218-044008. tries/healthcare-systems-and-services/our-insights/
3. Anderson, W. C., R. Gondalia, H. E. Hoch, L. Kaye, telehealth-a-quarter-trillion-dollar-post-covid-19-reali-
S. J. Szefler, and D. A. Stempel. 2019. Screening for ty (accessed October 5, 2020).
inhalation technique errors with electronic medica- 13. Biel, D., M. DeLone, W. Gerhardt, and C. Chang. 2019.
tion monitors. Journal of Allergy and Clinical immu- Radical Interoperability. Available at: https://www2.
nology: In Practice 7(6):2065-2067. https://doi. deloitte.com/us/en/insights/industry/health-care/
org/10.1016/j.jaip.2019.02.006. radical-data-interoperability-survey.html (accessed
4. Artiga, S., and E. Hinton. 2018. Beyond Health Care: October 5, 2020).
The Role of Social Determinants in Promoting Health 14. Boyde, A. M. and B. W. Chaffee. 2019. Critical Evalu-
and Health Equity. Available at: https://www.kff.org/ ation of Pharmacy Automation and Robotic Systems: A
racial-equity-and-health-policy/issue-brief/beyond- Call to Action. Hospital Pharmacy 54(1):4-11. https://
health-care-the-role-of-social-determinants-in-pro- doi.org/10.1177/0018578718786942.
moting-health-and-health-equity/ (accessed October 15. Buckeridge, D. L. 2020. Precision, Equity, and Pub-
5, 2020). lic Health and Epidemiology Informatics—A Scoping
5. Atasoy, H., B. N. Greenwood, and J. S. McCullough. Review. Yearbook of Medical Informatics 29(1):226-
2019. The Digitization of Patient Care: A Review of the 330. https://doi.org/10.1055/s-0040-1701989.
Effects of Electronic Health Records on Health Care 16. Bughin, J., E. Hazan, S. Ramaswamy, M. Chu, T. Allas,
Quality and Utilization. Annual Review of Public Health P. Dahlstrom, N. Henke, and M. Trench. 2017. Artifi-
40:487-500. https://doi.org/10.1146/annurev- cial Intelligence: The Next Digital Frontier? McKinsey
publhealth-040218-04420. Global Institute. Available at: https://www.mckinsey.
6. Avram, R., J. E. Olgin, P. Kuhar, J. W. Hughes, G. M. com/~/media/mckinsey/industries/advanced%20
Marcus. M. J. Pletcher, K. Aschbacher, and G. H. Ti- electronics/our%20insights/how%20artificial%20in-
son. 2020. A digital biomarker of diabetes from telligence%20can%20deliver%20real%20value%20
smartphone-based vascular signals. Nature Medicine to%20companies/mgi-artificial-intelligence-discus-
26:1576-1582. https://doi.org/10.1038/s41591- sion-paper.ashx (accessed February 24, 2022).
020-1010-5. 17. Buntin, M. B., M. F. Burke, M. C. Hoaglin, and D. Blu-
7. Baker, D. B., J. Kaye, and S. F. Terry. 2016. Privacy, menthal. 2011. The benefits of health information tech-
Fairness, and Respect for Individuals. eGEMS 4(2):7. nology: a review of the recent literature shows predom-
https://doi.org/10.13063/2327-9214.1207. inantly positive results. Health Affairs 30(3):464-471.
8. Barisoni, L., K. J. Lafata, S. M. Hewitt, A. Madabhushi, https://doi.org/10.1377/hlthaff.2011.0178.
and U. G. J. Balis. 2020. Digital pathology and com- 18. Buolamwini, J., and T. Gebru. 2018. Gender Shades:
putational image analysis in nephropathology. Na- Intersectional Accuracy Disparities in Commercial
ture Reviews Nephorology 16:669-685. https://doi. Gender Classification. Proceedings of Machine Learn-
org/10.1038/s41581-020-0321-6. ing Research 81:1-15. Available at: http://proceed-
9. Barrett, M., V. Combs, J. G. Su, K. Henderson, and ings.mlr.press/v81/buolamwini18a/buolamwini18a.
M. Tuffli. 2018. AIR Louisville: Addressing Asthma pdf (accessed February 14, 2022).
19. Buttorff, C., T. Ruder, and M. Bauman. 2017. Multiple
DISCUSSION PAPER
Chronic Conditions in the United States. Santa Monica: hhs-releases-project-us-draft-technical-specification-

RAND Corporation. Available at: https://www.rand. for-comment.html (accessed January 26, 2022).
org/pubs/tools/TL221.html (accessed February 14, 28. Dixon, B. E., S. J. Grannis, C. McAndrews, A. A. Bro-
2022). yles, W. Mikels-Carrasco, A. Wiensch, J. L. Williams,
20. Byrne, D. W., L. A. Rolando, M. H. Aliyu, P. W. Mc- U. Tachinardi, and P. J. Embi. 2021. Leveraging data
Gown, L. R., Connor, B. M. Awalt, M. C. Holmes, L. visualization and a statewide health information ex-
Wang, and M. I. Yarbrough. 2016. Modifiable Healthy change to support COVID-19 surveillance and re-
Lifestyle Behaviors: 10-Year Health Outcomes From sponse: Application of public health informatics. Jour-
a Health Promotion Program. American Journal of nal of the American Medical Informatics Association
Preventive Medicine 51(6):1029-1037. https://doi. 28(7):1363-1373. https://doi.org/10.1093/jamia/
org/10.1016/j.amepre.2016.09.012. ocab004.
21. Centers for Disease Control and Prevention (CDC). 29. Eddy, D. M. 1982. Clinical Policies and the Qual-
2020. Public Health Surveillance and Data. Available ity of Clinical Practice. New England Journal of
at: https://www.cdc.gov/surveillance/surveillance- Medicine 307:343-347. https://doi.org/10.1056/
data-strategies/dmi-investments.html (accessed Janu- NEJM198208053070604.
ary 26, 2022). 30. Engelhard, M. M., J. A. Oliver, and F. J. McClernon.
22. Centers for Medicare & Medicaid Services (CMS). 2020. Digital envirotyping: quantifying environmental
2019. Historical. Available at: https://www.cms. determinants of health and behavior. npj Digital Medi-
gov/Research-Statistics-Data-and-Systems/Statistics- cine 3(36):1-3. https://doi.org/10.1038/s41746-
Trends-and-Reports/NationalHealthExpendData/ 020-0245-3.
NationalHealthAccountsHistorical (accessed October, 31. European Commission (EC). 2021. Proposal for
5, 2020). a regulation of the European parliament and of
23. Cigna Newsroom. n.d. Convenient, Cost-Effective, and the council laying down harmonised rules on ar-
High-Quality Virtual Care Is Here to Stay. Available tificial intelligence (artificial intelligence act) and
at: https://newsroom.cigna.com/convenient-cost- amending certain union legislative acts. Available
effective-and-high-quality-virtual-care-is-here-to-stay at: https://eur-lex.europa.eu/legal-content/EN/
(accessed May 18, 2022). TXT/?uri=CELEX%3A52021PC0206 (accessed Janu-
24. Craig, K. J. T., N. Fusco, K. Lindsley, J. L. Snowdon, V. ary 26, 2022).
C. Willis, Y. E. Arriaga, and I. Dankwa-Mullan. 2020. 32. Food and Drug Administration (FDA). n.d. Good Ma-
Rapid review: Identification of digital health interven- chine Learning Practice for Medical Device Develop-
tions in atherosclerotic-related cardiovascular disease ment: Guiding Principles. Available at: https://www.
populations to address racial, ethnic, and socioeco- fda.gov/medical-devices/software-medical-device-
nomic health disparities. Cardiovascular Digital Health samd/good-machine-learning-practice-medical-de-
Journal 1(3):139-148. https://doi.org/10.1016/j. vice-development-guiding-principles (accessed Janu-
cvdhj.2020.11.001. ary 26, 2022).
25. De Keyser, H. E. H., L. Kaye, W. C. Anderson, R. 33. FDA. 2021. Artificial Intelligence and Machine Learn-
Gondalia, B. Theye, S. J. Szefler, and D. A. Stempel. ing in Software as a Medical Device. Available at:
2020. Electronic medication monitors help determine https://www.fda.gov/medical-devices/software-
adherence subgroups in asthma. Respiratory Medicine medical-device-samd/artificial-intelligence-and-ma-
164. https://doi.org/10.1016/j.rmed.2020.105914. chine-learning-software-medical-device (accessed
26. Deng, H. F., M-.W. Sun, Y. Wang, J. Zeng, T. Yuan, T. February 14, 2022).
Li, D-. H. Li, W. Chen, P. Zhou, Q. Wang, and H. Jiang. 34. Feagin, J., and Z. Bennefield. 2014. Systemic racism and
2021. Evaluating machine learning models for sep- U.S. health care. Social Science & Medicine 103:7-14.
sis prediction: A systematic review of methodologies. https://doi.org/10.1016/j.socscimed.2013.09.006.
iScience 25(1):103651. https://doi.org/10.1016/j. 35. Glaser, J. 2019. What Banking Can Teach Health Care
isci.2021.103651. About Handling Customer Data. Available at: https://
27. Department of Health and Human Services (HHS). hbr.org/2019/10/what-banking-can-teach-health-
2021. HHS Releases Project US@ Draft Technical care-about-handling-customer-data (accessed Octo-
Specification Version 1.0 for Comment. Available at: ber 5, 2020).
https://www.hhs.gov/about/news/2021/06/16/ 36. Goh, K. H., L. Wang, A. Y. K. Yeow, H. Poh, K. Li, J. J. L.

Yeow, and G. Y. H. Tan. 2021. Artificial intelligence in 48. Health Information National Trends Survey (HINTS).
sepsis early prediction and diagnosis using unstructured 2021. HINTS Briefs Number 45. Available at: https://
data in healthcare. Nature Communications 12:711. hints.cancer.gov/docs/Briefs/HINTS_Brief_45.pdf
https://doi.org/10.1038/s41467-021-20910-4. (accessed May 18, 2022).
37. Gordon, W. J., A. Landman, H. Zhang, and D. W. 49. HL7 International. n.d. FHIR® (HL7 Fast Healthcare
Bates. 2020. Beyond validation: getting health apps Interoperability Resources). Available at: https://
into clinical practice. Npj Digital Medicine 3(14):1-14. www.hl7.org/implement/standards/product_brief.
https://doi.org/10.1038/s41746-019-0212-z. cfm?product_id=491 (accessed May 18, 2022).
38. Halamka, J., and P. Cerrato. 2020. The Digital Recon- 50. Hruby, A., H. R. Lieberman, and T. J. Smith. 2018. Self-
struction of Health Care. NEJM Catalyst Innovations in reported health behaviors, including sleep, correlate
Care Delivery 1(6):1-12. https://doi.org/10.1056/ with doctor-informed medical conditions: data from the
CAT.20.0082. 2011 Health Related Behaviors Survey of U.S. Active
39. Hall, K. D., and S. Kahan. 2018. Maintenance of Lost Duty Military Personnel. BMC Public Health 18:1-16.
Weight and Long-Term Management of Obesity. Med- https://doi.org/10.1186/s12889-018-5781-2.
ical Clinics of North America 102(1)183-197. https:// 51. Ideas for Change. 2016. SalusCoop Report. Avail-
doi.org/10.1016/j.mcna.2017.08.012. able at: https://www.ideasforchange.com/en/tools/
40. Hartzog, W. 2018. Privacy’s Blueprint: The Battle to saluscoop-report (accessed February 14, 2022).
Control the Design of New Technologies. Boston: Har- 52. IHE International. 2021. Integrating the Healthcare En-
vard University Press. terprise (IHE). Available at: https://www.ihe.net/ (ac-
41. HealthIT.gov. 2022. Trusted Exchange Framework and cessed March 3, 2021).
Common Agreement (TEFCA). Available at: https:// 53. International Medical Device Regulators Forum Soft-
www.healthit.gov/topic/interoperability/trusted- ware as a Medical Device Working Group (IMDRF).
exchange-framework-and-common-agreement-tefca 2015. Software as a Medical Device (SaMD): Appli-
(accessed January 26, 2022). cation of Quality Management System. Available at:
42. HealthIT.gov. 2020. About ONC’s Cures Act Final https://www.imdrf.org/sites/default/files/docs/
Rule. Available at: healthit.gov/curesrule/overview/ imdrf/final/technical/imdrf-tech-151002-samd-qms.
about-oncs-cures-act-final-rule (accessed February pdf (accessed February 14, 2022).
14, 2022). 54. IMDRF. 2017. Clinical Evaluation. Available at: https://
43. HealthIT.gov. 2017. Rhode Island Quality Institute Tran- www.imdrf.org/sites/default/files/docs/imdrf/fi-
sitions of Care. Available at: https://www.healthit. nal/technical/imdrf-tech-170921-samd-n41-clinical-
gov/success-story/rhode-island-quality-institute-tran- evaluation_1.pdf (accessed February 14, 2022).
sitions-care (accessed January 26, 2022). 55. Institute of Medicine (IOM). 2010. The Healthcare
44. HealthIT.gov. 2009. Index for Excerpts from the Ameri- Imperative: Lowering Costs and Improving Out-
can Recovery and Reinvestment Act of 2009 (ARRA). comes: Workshop Series Summary. Washington,
Available at: https://www.healthit.gov/sites/default/ DC: The National Academies Press. https://doi.
files/hitech_act_excerpt_from_arra_with_index.pdf org/10.17226/12750.
(accessed February 14, 2022). 56. IOM. 2004. Patient Safety: Achieving a New Standard
45. HealthIT.gov. n.d. United States Core Data for Interop- for Care. Washington, DC: The National Academies
erability (USCDI). Available at: https://www.healthit. Press. https://doi.org/10.17226/10863.
gov/isa/united-states-core-data-interoperability-usc- 57. IOM. 2000. To Err Is Human: Building a Safer Health
di (accessed May 18, 2022). System. Washington, DC: The National Academies
46. Health IT Dashboard. 2016. Quick Stats. Available at: Press. https://doi.org/10.17226/9728.
https://dashboard.healthit.gov/quickstats/quickstats. 58. Isasi, F., M. D. Naylor, D. Skorton, D. C. Grabowski,
php (accessed September 17, 2020). S. Hernández, and V. Montgomery Rice. 2021. Pa-
47. Heller, D. J., K. A. Ornstein, L. V. DeCherrie, P. Saenger, tients, Families, and Communities COVID-19 Impact
F. C. Ko, C.-P. Rousseau, and A. Siu. 2020. Adapting Assessment: Lessons Learned and Compelling Needs.
a Hospital‐at‐Home Care Model to Respond to New NAM Perspectives. Discussion Paper, National Acad-
York City’s COVID‐19 Crisis. Journal of the American emy of Medicine, Washington, DC. https://doi.
Geriatrics Society 68(9):1915-1916. https://doi. org/10.31478/202111c.
org/10.1111/jgs.16725. 59. Ivanov, O., L. Wolf, D. Brecher, E. Lewis, K. Masek,
DISCUSSION PAPER
K. Montgomery, Y. Andrieiev, M. McLaughlin, S. Liu, Academy of Medicine, Washington, DC. https://doi.

R. Dunne, K. Klauer, and C. Reilly. 2020. Improving org/10.31478/202201c.
ED Emergency Severity Index Acuity Assignment Us- 68. Lin, S. Y., M. R. Mahoney, C. A. Sinky. 2019. Ten Ways
ing Machine Learning and Clinical Natural Language Artificial Intelligence Will Transform Primary Care. Jour-
Processing. Journal of Emergency Nursing 47(2):265- nal of General Internal Medicine 34(8):1620-1630.
278. https://doi.org/10.1016/j.jen.2020.11.001. https://doi.org/10.1007/s11606-019-05035-1.
60. Jain, J. S., B. W. Powers, J. B. Hawkins, and J. S. Brown- 69. Lindau, S. T. 2019. CommunityRx, an E-Prescribing
stein. 2015. The digital phenotype. Nature Biotech- System Connecting People to Community Resources.
nology 33(5):462-463. https://doi.org/10.1038/ American Journal of Public Health 109(4):546-547.
nbt.3223. https://doi.org/10.2105/AJPH.2019.304986.
61. Kaye, L., B. Theye, I. Smeenk, R. Gondalia, M. A. Bar- 70. Luchini, C., A. Pea, and A. Scarpa. Artificial intelligence
rett, and D. A. Stempel. 2020. Changes in medication in oncology: current applications and future perspec-
adherence among patients with asthma and COPD tives. British Journal of Cancer 126:4–9. https://doi.
during the COVID-19 pandemic. The Journal of Aller- org/10.1038/s41416-021-01633-1.
gy and Clinical Immunology: In Practice 8(7): 2384- 71. McGinnis, J. M., H. V. Fineberg, and V. J. Dzau.
2385. https://doi.org/10.1016/j.jaip.2020.04.053. 2021. Advancing the Learning Health System. New
62. Kearney, A, L. Hamel, M. Stokes, and M. Broadie. England Journal of Medicine 385:1-5. https://doi.
2021. Americans’ Challenges with Health Care Costs. org/10.1056/NEJMp2103872.
Kaiser Family Foundation, December 14. Available at 72. McGinnis, M. J., P. Williams-Russo, and J. R. Knickman.
https://www.kff.org/health-costs/issue-brief/amer- 2002. The Case For More Active Policy Attention To
icans-challenges-with-health-care-costs/ (accessed Health Promotion. Health Affairs 21(2):78-93. https://
May 18, 2022). doi.org/10.1377/hlthaff.21.2.78.
63. Keehan, S. P., G. A. Cuckler, J. A. Poisal, A. M. Sisko, S. 73. McGovern, L., G. Miller, and P. Hughes-Cromwick.
D. Smith, A. J. Madison, K. E. Rennie, J. A. Fiore, and J. 2014. The Relative Contribution of Multiple Determi-
C. Hardesty. 2020. National Health Expenditure Pro- nants to Health Outcomes. Available at: https://www.
jections, 2019–28: Expected Rebound In Prices Drives rwjf.org/en/library/research/2014/08/the-rela-
Rising Spending Growth. Health Affairs 39(4):704- tive-contribution-of-multiple-determinants-to-health-
714. https://doi.org/10.1377/hlthaff.2020.00094. out.html (accessed October 5, 2020).
64. Kottke, T. E., M. Stiefel, and N. P. Pronk. 2016. 74. Melnick, E. R., L. N. Dyrbye, C. A. Synsky, M. Trockel, C.
“Well-Being in All Policies”: Promoting Cross-Sec- P. West, L. Nedelec, M. A. Tutty, and T. Shanafelt. 2020.
toral Collaboration to Improve People’s Lives. NAM The Association Between Perceived Electronic Health
Perspectives. Discussion Paper, National Acad- Record Usability and Professional Burnout Among US
emy of Medicine, Washington, DC. https://doi. Physicians. Mayo Clinic Proceedings 95(3):476-487.
org/10.31478/201604c. https://doi.org/10.1016/j.mayocp.2019.09.024.
65. Lamberti, M. J., M. Wilkinson, B. A. Donzanti, G. E. 75. Merchant, R., S. J. Szefler, B. G. Bender, M. Tuffli, M.
Wohlhieter, S. Parikh, R. G. Wilkins, and K. Getz. 2019. A. Barrett, R. Gondalia, L. Kaye, D. V. Sickle, and D.
A Study on the Application and Use of Artificial Intel- A. Stempel. 2018. Impact of a digital health interven-
ligence to Support Drug Development. Clinical Thera- tion on asthma resource utilization. World Allergy Or-
peutics 41(8):1414-1426. https://doi.org/10.1016/j. ganization Journal 11(28). https://doi.org/10.1186/
clinthera.2019.05.018. s40413-018-0209-0.
66. LaRosa, J. 2020. $71 Billion U.S. Weight Loss Indus- 76. Meyer, D., E. Lerner, A. Phillips, and K. Zumwalt. 2020.
try Pivots to Survive Pandemic. Available at: https:// Universal Screening of Social Determinants of Health
blog.marketresearch.com/71-billion-u.s.-weight-loss- at a Large US Academic Medical Center. American
market-pivots-to-survive-pandemic (accessed January Journal of Public Health 110(S2):S219-S221. https://
26, 2022). doi.org/10.2105/AJPH.2020.305747.
67. Lee, P., A. Abernethy, D. Shaywitz, A. V. Gundlapalli, 77. MiDATA. n.d. My Data—Our Health. Available at:
J. Weinstein, P. M. Doraiswamy, K. Schulman, and https://www.midata.coop/en/home/ (accessed
S. Madhavan. 2022. Digital Health COVID-19 Im- October 5, 2020).
pact Assessment: Lessons Learned and Compelling 78. Murray, S. G., R. M. Wachter, and R. J. Cucina. 2020.
Needs. NAM Perspectives. Discussion Paper, National Discrimination By Artificial Intelligence In A Commer-

cial Electronic Health Record—A Case Study. Avail- org/10.1007/s13142-013-0234-z.

able at: https://www.healthaffairs.org/do/10.1377/ 88. Panagioti, M., K. Geraghty, J. Johnson, A. Zhou, E.
hblog20200128.626576/full/ (accessed March 3, Panagopoulou, C. Chew-Graham, A. Hodkinson, R.
2021). Riley, and A. Esmail. 2018. Association Between Physi-
79. Muzney, M., A. Henriksen, A. Giordanengo, J. Muzik, cian Burnout and Patient Safety, Professionalism, and
A. Grøttland, H. Blixgård, G. Hartvigsen, and E. År- Patient Satisfaction: A Systematic Review and Meta-
sand. 2020. Wearable sensors with possibilities for analysis. JAMA Internal Medicine 178(10):1317-1331.
data exchange: Analyzing status and needs of differ- https://doi.org/10.1001/jamainternmed.2018.3713.
ent actors in mobile health monitoring systems. Interna- 89. Patel, S., L. Jacobus-Kantor, L. Marshall, C. Ritchie, M.
tional Journal of Medical Informatics 133:1-8. https:// Kaplinski, P. S. Khurana, and R. J. Katz. 2013. Mobi-
doi.org/10.1016/j.ijmedinf.2019.104017. lizing Your Medications: An Automated Medication
80. National Academy of Medicine (NAM). 2020. NAM Reminder Application for Mobile Phones and Hyper-
Leadership Consortium: Collaboration for a Value & tension Medication Adherence in a High-Risk Urban
Science-Driven Health System. Available at: https:// Population. Journal of Diabetes Science and Technol-
nam.edu/programs/value-science-driven-health- ogy 7(3):630-639. https://doi.org/10.1177/19322
care/ (accessed October 5, 2020). 9681300700307.
81. National Academies of Sciences, Engineering, and 90. The National Patient-Centered Clinical Research Net-
Medicine (NASEM). 2020. Joint Science Academies work (PCORNet). n.d. Home. Available at: https://
Statements on Global Issues (G-Science). Available pcornet.org/ (accessed May 18, 2022).
at: https://www.nationalacademies.org/our-work/ 91. Rasmussen, S. A., M. K. Khoury, and C. Del Rio. 2020.
joint-science -academies-statements-on-global- Precision Public Health as a Key Tool in the COVID-19
issues#sectionProjectScope (accessed January 26, Response. JAMA 324(10):933-934. https://doi.
2022). org/10.1001/jama.2020.14992.
82. National Institute of Standards and Technology 92. Research Data Assistance Center (ResDAC). 2022.
(NIST). 2018. Framework for Improving Critical In- Home. Available at: https://resdac.org/ (accessed
frastructure Cybersecurity, Version 1.1. Available at May 18, 2022).
https://nvlpubs.nist.gov/nistpubs/CSWP/NIST. 93. Schroeder, S. A. 2007. We Can Do Better—Improv-
CSWP.04162018.pdf (accessed May 13, 2022). ing the Health of the American People. New England
83. National Institutes of Health (NIH). 2020. All of Journal of Medicine 357(12):1221-1228. https://doi.
Us Research Program Backgrounder. Available at: org/10.1056/NEJMsa073350.
https://allofus.nih.gov/news-events/press-kit/all-us- 94. Schueller, S. M., J. F. Hunter, C. Figueroa, and A. Agu-
research-program-backgrounder (accessed October ilera. 2019. Use of Digital Mental Health for Margin-
26, 2020). alized and Underserved Populations. Current Treat-
84. National Library of Medicine (NLM). 2022. RxNorm ment Options in Psychiatry 6:243-255. https://doi.
Overview. Available at: https://www.nlm.nih.gov/ org/10.1007/s40501-019-00181-z.
research/umls/rxnorm/overview.html (accessed May 95. Settineri, S., F. Frisone, E. M. Merlo, D. Geraci, and G.
18, 2022). Martino. 2019. Compliance, adherence, concordance,
85. Observational Health Data Sciences and Informatics empowerment, and self-management: five words to
(OHDSI). 2022. Home. Available at: https://www. manifest a relational maladjustment in diabetes. Journal
ohdsi.org/ (accessed May 18, 2022). of Multidisciplinary Healthcare 12:299-314. https://
86. Office of the National Coordinator for Health Informa- doi.org/10.2147/JMDH.S193752.
tion Technology (ONC). 2020. Health IT Workflow 96. Shan, R., S. Sarker, and S. S. Martin. 2019. Digital
Automation: Background Report, Industry Lessons for health technology and mobile devices for the manage-
Health Care. Available at: https://www.healthit.gov/ ment of diabetes mellitus: state of the art. Diabetologia
sites/default/files/topiclanding/2021-07/Work- 62:877-887. https://doi.org/10.1007/s00125-019-
flow_Automation_Background_Report_FINAL.pdf 4864-7.
(accessed May 18, 2022). 97. Shegog, R., L. Braverman, and J. D. Hixson. 2020.
87. Pagoto, S., and G. G. Bennett. 2020. How behav- Digital and technological opportunities in epilepsy:
ioral science can advance digital health. Translation- toward a digital ecosystem for enhanced epilepsy
al Behavioral Medicine 3(3):271-276. https://doi. management. Epilepsy & Behavior 102. https://doi.
DISCUSSION PAPER
org/10.1016/j.yebeh.2019.106663. their-care-providers (accessed January 26, 2022).

98. Shrank, W. H., T. L. Rogstad, and N. Parekh. 2019. 108. The Pew Charitable Trusts. 2018. Enhanced Pa-
Waste in the US Health Care System: Estimated Costs tient Matching Is Critical to Achieving Full Prom-
and Potential for Savings. JAMA 322(15):1501-1509. ise of Digital Health Records. Available at: https://
https://doi.org/10.1001/jama.2019.13978. www.pewtrusts.org/en/research-and-analysis/re-
99. Shen, N., D. Jankowicz, and G. Strudwick. 2021. Pa- ports/2018/10/02/enhanced-patient-matching-crit-
tient and Family Engagement Approaches for Digi- ical-to-achieving-full-promise-of-digital-health-records
tal Health Initiatives: Protocol for a Case Study. JMIR (accessed January 26, 2022).
Research Protocols 10(7):e24274. https://doi. 109. Tikkanen, R., and M. K. Abrams. 2020. U.S.
org/10.2196/24274. Health Care from a Global Perspective, 2019: Higher
100. Singhal, S., B. Kayyali, R. Levin, and Z. Greenberg. Spending, Worse Outcomes? Available at: https://
2020. The next wave of healthcare innovation: The www.commonwealthfund.org/publications/issue-
evolution of ecosystems. Available at: https://www. briefs/2020/jan/us-health-care-global-perspec-
mckinsey.com/industries/healthcare-systems-and-ser- tive-2019 (accessed October 5, 2020).
vices/our-insights/the-next-wave-of-healthcare-inno- 110. Topol, E. J. 2019. A decade of digital medicine inno-
vation-the-evolution-of-ecosystems (accessed October vation. Science Translational Medicine 11(498):1-3.
5, 2020). https://doi.org/10.1126/scitranslmed.aaw7610.
101. SNOMED International. n.d. Who We Are. Available 111. Tung, J. Y., R. J. Shaw, J. M. Hagenkord, M. Hack-
at: https://www.snomed.org/snomed-international/ mann, M. Muller, S. H. Beachy, V. M. Pratt, S. F. Terry,
who-we-are (accessed May 18, 2022). A. K. Cashion, and G. S. Ginsburg. 2018. Accelerating
102. Steinmetz, M., C. Rammos, T. Rassaf, and J. Lortz. precision health by applying the lessons learned from
2020. Digital interventions in the treatment of car- direct-to-consumer genomics to digital health technolo-
diovascular risk factors and atherosclerotic vascular gies. NAM Perspectives. Discussion Paper, National
disease. International Journal of Cardiology Heart & Academy of Medicine, Washington, DC. https://doi.
Vasculature 26:1-8. https://doi.org/10.1016/j.ij- org/10.31478/201803c.
cha.2020.100470. 112. USF Morsani College of Medicine. 2021. What is
103. Tanner, C., D. Gans, J. White, R. Nath, and J. Pohl. the Argonaut Project? Available at: https://www.
2015. Electronic Health Records and Patient Safety. usfhealthonline.com/resources/health-informatics/
Applied Clinical Informatics 6:136-147. https://doi. what-is-the-argonaut-project/ (accessed January 26,
org/10.4338/ACI-2014-11-RA-0099. 2022).
104. The Commonwealth Fund. 2019. 2019 Common- 113. Walker, J., E. Pan, D. Johnston, J. Adler-Milstein, D. W.
wealth Fund International Health Policy Survey of Pri- Bates, and B. Middleton. 2005. The Value of Health
mary Care Physicians. Available at: https://www. Care Information Exchange and Interoperability.
commonwealthfund.org/publications/surveys/2019/ Health Affairs 24(S1):W5-10-W5-18. https://doi.
dec/2019-commonwealth-fund-international-health- org/10.1377/hlthaff.w5.10.
policy-survey-primary (accessed March 3, 2021). 114. World Health Organization (WHO). 2006. Constitu-
105. The Commonwealth Fund. 2016. High-Need, tion of the World Health Organization. Available at:
High-Cost Patients: Who Are They and How Do They https://www.who.int/governance/eb/who_constitu-
Use Health Care? Available at: https://www.com- tion_en.pdf (accessed February 13, 2022).
monwealthfund.org/publications/issue-briefs/2016/ 115. Zachrison, K. S., Z. Yan, T. Sequist, A. Licurse, A. Tan-
aug/high-need-high-cost-patients-who-are-they-and- McGrory, A. Erskine, and L. H. Schwamm. 2021. Pa-
how-do-they-use (accessed January 26, 2022). tient characteristics associated with the successful
106. The Lancet Digital Health. 2019. There is no such transition to virtual care: Lessons learned from the first
thing as race in health-care algorithms. The Lancet million patients. Journal of Telemedicine. https://doi.
Digital Health 1(8):e375. https://doi.org/10.1016/ org/10.1177/1357633X211015547.
S2589-7500(19)30201-8.
107. The Pew Charitable Trusts. 2020. Patients Seek Bet- DOI
ter Exchange of Health Data Among Their Care Pro- https://doi.org/10.3147/202206e
viders. Available at: https://www.pewtrusts.org/
en/research-and-analysis/issue-briefs/2020/03/
patients-seek-better-exchange-of-health-data-among-

Suggested Citation Kristen Valdes is Founder and CEO at b.well Connected

Health.
Abernethy, A., L. Adams, M. Barrett, C. Bechtel, P. Brennan,
A. Butte, J. Faulkner, E. Fontaine, S. Friedhoff, J. Halamka, M.
Drs. Brennan, Butte, Halamka, Johnson, Miller, Lee, Perlin,
Howell, K. Johnson, P. Lee, P. Long, D. McGraw, R. Miller, J.
Tang, Topol, and Tuckson are members of the National
Perlin, D. Rucker, L. Sandy, L. Savage, L. Stump, P. Tang, E.
Academy of Medicine.
Topol, R. Tuckson, and K. Valdes. 2022. The Promise of Digi-
tal Health: Then, Now, and the Future. NAM Perspectives. Acknowledgments
Discussion Paper, National Academy of Medicine, Wash-
ington, DC. https://doi.org/10.31478/202206e. This paper benefited from the thoughtful input of John Gla-
ser, Siemens Healthcare; Clement McDonald, National
Author Information Library of Medicine; and Marc Overhage, Anthem.
Amy Abernethy, MD, PhD, is President, Clinical Research
The authors would like to recognize Paul Cerrato with the
Platforms, at Verily. Laura Adams, MS, is Senior Advisor
Mayo Clinic for contributing to earlier drafts of this publica-
at the National Academy of Medicine. Meredith Barrett,
tion and to thank Mahnoor Ahmed and Asia Williams
PhD, is Vice President, Population Health Research, at
with the National Academy of Medicine Leadership Con-
ResMed. Christine Bechtel, MA, is Co-Founder at X4
sortium for assistance with research and fact-checking.
Health. Patricia Brennan, RN, PhD, FAAN, is Director
at the National Library of Medicine. Atul Butte, MD, Conflict-of-Interest Disclosures
PhD, is Priscilla Chan and Mark Zuckerberg Distinguished
Professor and inaugural Director, Bakar Computational Amy Abernethy discloses employment by Verily, an Al-
Health Sciences Institute at University of California, San phabet Company; employment by the U.S. FDA while this
Francisco. Judith Faulkner, MS, is Founder and Chief paper was being drafted; and serving as member of the
Executive Officer at Epic Systems. Elaine Fontaine, Board of Directors or EQRx. Laura Adams discloses em-
BS, is Consultant at the National Academy of Medicine. ployment as Senior Advisor to the National Academy of
Stephen Friedhoff, MD, is Senior Vice President, Clinical Medicine. Meredith Barrett discloses receiving personal
Strategy and Programs, at Anthem, Inc. John Halamka, fees from ResMed and Propeller Health. Atul Butte dis-
MD, is President, Mayo Clinic Platform, at the Mayo closes receiving grants and non-financial support from Pro-
Clinic. Michael Howell, MD, MPH, is Chief Clinical genity; personal fees and other support from NuMedii, Per-
Officer at Google Health. Kevin Johnson, MD, is Penn sonalis, Assay Depot, GNS Healthcare, uBiome, and Nuna
Integrates Knowledge University Professor at the University Health; grants and personal fees from NIH and Genentech;
of Pennsylvania. Peter Lee, PhD, is Corporate Vice grants from L’Oreal and Samsung; personal fees and non-
President at the Microsoft Corporation. Peter Long, PhD, financial support from Merck, Lilly, Geisinger Health, and
is Executive Vice President, Strategy and Health Solutions at Roche; and serving as consultant to Wilson Sonsini Goorich
Blue Shield of California. Deven McGraw, JD, MPH, is & Rosati, Orrick Herrington & Sutcliffe, Verinata, 10x Ge-
Chief Regulatory Officer at Ciitizen Corporation. Redonda nomics, Pathway Genomics, Guardant Health, and Gerson
Miller, MD, MBA, is President of Johns Hopkins Hospital. Lehrman Group. Elaine Fontaine discloses employment
Jonathan Perlin, MD, PhD, MSHA, is President of The as a consultant to the National Academy of Medicine.
Joint Commission. Donald Rucker, MD, is Chief Strategy Stephen Friedhoff discloses employment by Anthem
Officer at 1upHealth. Lewis Sandy, MD, MBA, is Senior BC while this paper was being drafted; current employ-
Vice President, Clinical Advancement, at UnitedHealth ment by BCNC; serving as board member to Agape Care
Group. Lucia Savage, JD, is Chief Privacy and Regulatory and Medical Review Institute of America; and advisor to
Officer at Omada Health, Inc. Lisa Stump, MS, is Senior Rialtic. Michael Howell discloses employment and eq-
Vice President and Chief Information Officer at Yale New uity in Google, an Alphabet company. Kevin Johnson
Haven Health System and Yale School of Medicine. Paul discloses employment by University of Pennsylvania. Peter
Tang, MD, MS, is adjunct professor, Clinical Excellence Lee discloses employment by Microsoft. Deven McGraw
Research Center at Stanford University. Eric Topol, MD, discloses employment by Invitae Corporation and Ciitizen
is Executive Vice President and Professor at The Scripps Corporation; personal fees from Datavant and All of Us Re-
Research Institute. Reed Tuckson, MD, FACP, is search Program; and serving as board member for CARIN
Managing Director at Tuckson Health Connections, LLC. Alliance and Manifest Medex. Redonda Miller discloses
DISCUSSION PAPER
employment by the Johns Hopkins University. Jonathan

Perlin discloses former employment by HCA Healthcare.
Donald Rucker discloses former employment by the Na-
tional Coordinator for Health IT; current employment by 1up
Health; and personal fees from Cync Health. Paul Tang
discloses employment by IBM Watson Health. Eric Topol
discloses receiving personal fees from Illumnia and serving
on the board of directors at Dexcom. Kristen Valdes dis-
closes employment by b.well Connected Health.
Correspondence
Questions and comments about this paper should be direct-
ed to Laura Adams at LAdams@nas.edu. Additional inqui-
ries should be sent to NAMedicine@nas.edu.
Disclaimer
The views expressed in this paper are those of the authors
and not necessarily of the authors’ organizations, the Na-
tional Academy of Medicine (NAM), or the National
Academies of Sciences, Engineering, and Medicine (the
National Academies). The paper is intended to help inform
and stimulate discussion. It is not a report of the NAM or the
National Academies. Copyright by the National Academy
of Sciences. All rights reserved.

Highlighted DH

Uploaded by

Copyright:

Available Formats

Highlighted DH

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Highlighted DH

Uploaded by

Copyright:

Available Formats

DISCUSSION PAPER

The Promise of Digital Health: Then, Now, and

June 27, 2022

Perspectives | Expert Voices in Health & Health Care

Page 2 Published June 27, 2022

FIGURE 1 | Evolving Applications of Digital Technology in Health and Health Care

Page 4 Published June 27, 2022

Page 6 Published June 27, 2022

Page 8 Published June 27, 2022

FIGURE 2 | Infrastructure Requirements for Progress in Digital Health

Page 10 Published June 27, 2022

Page 12 Published June 27, 2022

to implement a “modern, interoperable, and real-time pub- Workforce

Page 14 Published June 27, 2022

also be employed, as it will enable the necessary organiza- Ensuring Cybersecurity

Page 16 Published June 27, 2022

Chronic Conditions in the United States. Santa Monica: hhs-releases-project-us-draft-technical-specification-

Page 18 Published June 27, 2022

K. Montgomery, Y. Andrieiev, M. McLaughlin, S. Liu, Academy of Medicine, Washington, DC. https://doi.

Page 20 Published June 27, 2022

cial Electronic Health Record—A Case Study. Avail- org/10.1007/s13142-013-0234-z.

org/10.1016/j.yebeh.2019.106663. their-care-providers (accessed January 26, 2022).

Page 22 Published June 27, 2022

Suggested Citation Kristen Valdes is Founder and CEO at b.well Connected

employment by the Johns Hopkins University. Jonathan

Page 24 Published June 27, 2022

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.