Remote Digital Technologies for Improving the Care of People with Respiratory Disorders

Unstructured Summary

Respiratory diseases are a leading cause of morbidity and mortality globally. However, existing systems of care, built around predefined episodic interactions, are not well designed to support the needs of people with chronic or acute respiratory conditions that can change rapidly and unexpectedly. Home-based and personal digital health technologies (DHTs) allow implementation of new models of care, built around the unique needs of individuals. It is the individual nature of respiratory triggers and their unique response to them that requires a personalized solution to impact health. The real-world, repetitive monitoring capabilities of DHTs enable identification of the “normal operating characteristics” for each individual, and therefore, recognition of the earliest deviations from that state. However, despite this potential, the number of clinical validation studies of DHTs is quite small, and there is an urgent need to evaluate clinical effectiveness in improving health quality in real-world settings.

Respiratory conditions are a tremendous burden on the individuals experiencing them, their family, the healthcare providers caring for them, and society. Digital health technologies (DHTs) could lessen this burden, but there is a lack of consensus about what to measure to reflect relevant clinical status and changes in status for people with respiratory conditions. In this Viewpoint article, we provide two examples of how the application of remote sensor data and other personal DHTs might be incorporated into patient care to 1) monitor people with chronic respiratory conditions remotely and 2) allow early detection of acute respiratory infection. These examples allow us to describe both the potential benefits and remaining critical gaps regarding DHT use in these and other cases to improve the care of people with respiratory conditions.

Lynnae is a 37-year-old woman with a ^~20-year history of difficult-to-control asthma with recurrent exacerbations severe enough, in the past, to require hospitalization 3–4 times a year. She lives in rural Alaska and receives most of her medical care through an advanced practitioner in her hometown, supplemented by primarily virtual follow-up by a regional pulmonologist who is otherwise reachable only by plane flight. Between routine episodes of care, her pulmonologist monitors her rescue inhaler use through her connected, digital inhaler. Shortly after starting to use the smart inhaler ^~2 years ago, she and her pulmonologist recognized that her rescue inhaler use significantly increased during a period of poor air quality (Figure 1), which led to Lynnae installing high-efficiency particulate air (HEPA) filters in her home and initiating tracking her indoor air quality with a monitor.

Figure 1 -.

Multiple sensor data streams, such as automated tracking of inhaler use frequency, with indoor and/or outdoor air quality, can be combined to better inform the user and their care provider of personal exacerbating factors.

This first case highlights both the challenges that individuals with significant chronic respiratory diseases face and some of the potential advantages of incorporating remote digital health technologies into their care. For many people with chronic respiratory conditions, the substantial burden of self-management and constant hypervigilance for environmental triggers and potential infectious risks negatively impacts quality of life. Too often, patients with respiratory illness can feel they bear the responsibility for the episodic deteriorations in their condition.¹

Existing systems of care, built around somewhat arbitrarily timed routine follow-up visits, with the emergency room often as a sole contingency option, are not well designed to support the needs of people like Lynnae whose condition can change rapidly and unexpectedly. She, and all people at risk for respiratory decompensation, have limited options for rapid evaluation when they begin experiencing the early symptoms of allergies or infections, leading to delays in diagnosis, isolation, and treatment.

Fortunately, many existing gaps in the care continuum are now well suited to be solved through the implementation of redesigned systems of care built around a continuously increasing variety of novel home-based and personal DHTs. DHTs are still quite novel and rapidly changing, and therefore can be a confusing term. For this viewpoint, we define DHTs based on previously published criteria that include technologies that: 1. Collect clinical or health-related data, 2. Contain both a physiologic sensor and embedded software, 3. Are portable (not restricted to facility use), 4. Connect to the internet or other internet-connected technologies to transmit data, and 5. Are designed for patient-facing use (limited clinician involvement).²

Burden of Respiratory Conditions

The lungs are the only internal organs continuously exposed to our environment, including its toxins, allergens, particulate matter, and infectious agents. Considering the diversity and constancy of insults to the airways, it is not surprising that respiratory diseases are a leading cause of death. While there are numerous chronic respiratory conditions, including cystic fibrosis, interstitial lung disease, sarcoidosis, and post-lung transplantation, the two most common chronic respiratory diseases are asthma and COPD, which will be the primary focus of this viewpoint. On an annual basis, chronic obstructive pulmonary disease (COPD) is the third leading cause of death globally.³ Most years, lower respiratory tract infections are already a major killer globally, especially for those at the extremes of age.⁴ However, the COVID-19 pandemic has eclipsed those usual numbers with ^~7 million deaths at the time of this writing, most due to respiratory complications.⁵

The true burden of respiratory diseases extends well beyond excess mortality and morbidity, exacting a significant emotional and financial cost to individuals, their families, and societies. Asthma is a leading cause of missed school days for children, affecting not only their educational attainment but also increasing the stress and lowering the quality of life of their family caregivers.⁶ The majority of people with COPD find their symptom burden to be a significant challenge to their routine daily activities, and with that, their quality of life.⁷ While the limited duration of symptoms for most people experiencing an acute respiratory infection creates a minimal individual burden, due to their high incidence rate, acute respiratory infections create an enormous societal economic burden. As staggering as the estimated $11.2 billion average cost of annual influenza is in the U.S., that number was transcended by the $16 trillion estimated cost of the COVID-19 pandemic in 2020 alone.^8,9

The Problems to be Solved

All people are affected by respiratory illness. For most, addressing the sporadic and unpredictable nature of acute infection may be all that is needed. But for over half a billion people with chronic respiratory conditions, like Lynnae, and those close to them, the challenges to be addressed are more ubiquitous and complex.¹⁰ In addition, as populations age worldwide, the incidence of chronic pulmonary conditions will only continue to increase.¹¹ This will require already struggling health systems to adapt to better meet the needs of populations experiencing chronic disability for a longer time of their lives, necessitating reorganization of their systems of care to enable self-management. DHTs, when implemented in an innovative and person-centric manner, can provide the flexibility of design and scale needed to better address the unique needs of each individual. They also might reduce inequities in access to healthcare for underserved populations, as well as lessen the need for unnecessary travel and other resource use.

While the unmet needs are numerous, the overarching goal of any solution is to provide the individual patient with a personalized management strategy imposing the lowest burden possible. What works best for the very unique needs of Lynnae would not necessarily be helpful for others. A successful approach must enable the user to maximize their functional capacity while minimizing life-disrupting management activities and, most importantly, acute exacerbations and their consequences.

A distinctive capability of personal sensor technologies that helps make this possible is the ability to identify the “normal operating characteristics” for each individual, and early deviations from that. In Lynnae’s case, that was a change in her rescue inhaler use, but in future scenarios, such as described later, this could include unexpected changes in heart or respiratory rate during routine activities.

It is the individual nature of respiratory triggers and a person’s unique response to them that mandates a personalized solution to meaningfully improve health. As with Lynnae, environmental triggers can be learned through the right sensors and data, and once identified, can be used to proactively minimize exacerbations. Similarly, public health data, along with physiologic data, can inform avoidance or early diagnosis and treatment strategies. As neighborhoods exposed to higher concentrations of air pollutants are too often lower-income communities, the combination of personal and public data could be especially valuable in beginning to address this aspect of health inequities.¹²

Addressing Chronic Respiratory Illnesses

The selection of novel DHTs available to address respiratory care is vast and growing, as shown for the U.S. in Figure 2 and Table 1. However, irrespective of the inventiveness of a technology, it is rarely a standalone solution in patient care. Rather, it needs to be considered as a novel “tool” around which inventive approaches to care can be built—approaches that, if done correctly, can motivate and encourage users to feel comfortable engaging in their own care.

Figure 2 -.

Personal sensor technologies currently available that have been, or could be, utilized within redesigned systems of care built to better identify early signs of decompensation while lessening the burden of care for patients.

Table 1:

Personal Physiologic Monitoring Technologies Available to Address Challenges in Respiratory Care

Digital technology	Parameters Measured#	Continuous or Intermittent	Potential Clinical Use Cases	Device FDA Regulatory Clearance*?	User’s Role**	Clinician’s/Health System’s Role**
Smart watches, fitness trackers, smart rings, smart clothing	Activity Pulse rate & irregular rhythm detection Heart rate variability Respiratory rate Oxygen saturation Electrodermal activity Skin temperature Sleep duration Blood pressure ECG	Besides activity, most parameters that require photoplethysmogra phy (PPG) are measured only among immobile patients or at rest for ambulatory patients, especially while sleeping.	Early detection of physiologic response to decompensation.	No clearance for most devices and parameters. Exceptions for ECG for multiple manufacturers plus irregular rhythm detection. One device with clearance for BP, pulse, respiratory rate, temperature and oxygen saturation.	Wear device as frequently as possible. Charge the device as needed. Typically requires Bluetooth connection to a smartphone or other internet-connected device. Be aware of limitations of non-validated data streams.	Ability to track continuously, when necessary, or ingest summary statistics periodically. Recognize and react to individual deviations from expected normals. Seamless incorporation of pertinent data into electronic health record. Ability to respond to user questions/concerns.
Patch sensors	Activity ECG Heart rate variability Respiratory rate Oxygen saturation Skin & derived core temperature Sleep duration/quality	For patches with continuous ECG, all vital signs derived from it, including respiration, are available continuously. For non-ECG patches, vital sign data are available only for immobile patients or at rest for ambulatory patients.	Early detection of physiologic response to decompensation.	Majority with clearance for all measured vital signs as well as some individually derived digital biomarkers.	Wear device continuously as long as indicated. Maintain Bluetooth connection to a smartphone or other internet-connected device as frequently as needed. Charge or change the device as needed.	Ability to track continuously. Recognize and react to individual deviations from expected normals. Seamless incorporation of pertinent data into electronic health record. Ability to respond to user questions/concerns.
Finger pulse oximeters	Oxygen saturation Pulse rate	Primarily recurring spot checks, but continuous available.	Specific monitoring for hypoxia in settings of infections or exacerbations.	Majority with clearance.	Use device as indicated or as needed. Maintain Bluetooth connection to a smartphone or other internet-connected device as frequently as needed. Charge as needed.	Ability to capture spotcheck data. Recognize and react to individual deviations from expected normals. Seamless incorporation of pertinent data into electronic health record. Ability to respond to user questions/concerns.
Thermometers	Body temperature	Primarily spot checks, but also available as a continuous patch sensor.	Identification of fever or concerning temperature trajectories suggesting infection.	Although considered a Class II device, clearance is currently not required.	Use device as indicated or as needed. Maintain Bluetooth connection to a smartphone or other internet-connected device as frequently as needed. Charge as needed.	Ability to capture spotcheck data. Recognize and react to individual deviations from expected normals. Seamless incorporation of pertinent data into electronic health record. Ability to respond to user questions/concerns.
Home spirometry	Peak expiratory flow Forced expiratory volume, 1 sec (FEV1) Forced vital capacity	Intermittent spot checks.	Tracking lung function over time in relation to therapeutic or environment changes.	Majority of currently available devices in U.S. have clearance.	Use device as indicated or as needed Maintain Bluetooth connection to a smartphone or other internet-connected device as frequently as needed. Charge as needed.	Ability to capture spotcheck data. Recognize and react to individual deviations from expected normals. Seamless incorporation of patient data into electronic health record. Ability to respond to user questions/concerns.
Cough frequency detectors	Cough frequency	Continuous	Potential for early detection of decompensation and track clinical course.	Most available technologies, primarily apps, are not cleared. One patch sensor has approval for cough detection.	Wear approved device continuously as long as indicated. Be aware of limitations of non-validated sensor data. Maintain Bluetooth connection to a smartphone or other internet-connected device as frequently as needed. Charge or change the device as needed.	Ability to track cough frequency. Recognize and react to individual deviations from expected normals. Seamless incorporation of data into electronic health record. Ability to respond to user questions/concerns.
Breath analysis	Fractional exhaled nitric oxide Exploratory breath biomarkers	Intermittent spot checks.	Identifying and potentially monitoring airway inflammation in asthma.	Multiple cleared devices for fractional exhaled nitric oxide. Other devices for research only.	Use device as indicated or as needed. Maintain Bluetooth connection to a smartphone or other internet-connected device as frequently as needed. Charge as needed.	Ability to capture spotcheck data. Recognize and react to individual deviations from expected normals. Seamless incorporation of pertinent data into electronic health record. Ability to respond to user questions/concerns.
Lung acoustics	Lung sounds	Patches for continuous monitoring and digital personal devices for monitoring lung sounds.	Identifying presence and tracking changes in pathologic lung sounds.	One currently cleared patch device and multiple cleared digital selfuse stethoscopes to transmit breath sounds.	Use device as indicated or as needed. Maintain Bluetooth connection to a smartphone or other internet-connected device as frequently as needed. Charge as needed.	Ability to capture spotcheck data. Recognize and react to individual deviations from expected normals. Seamless incorporation of pertinent data into electronic health record. Ability to respond to user questions/concerns.
Passive sleep-based monitors	Respiratory rate Pulse rate Sleep duration/quality	In-bed sensors and near-bed radar systems; continuous while in bed.	Early detection of physiologic response to decompensation and overall impact of condition on sleep.	None currently FDA-cleared.	Use device as frequently as possible. Maintain Bluetooth connection to a smartphone or other internet-connected device. Be aware of limitations of non-validated data streams.	Ability to track continuously, when necessary, or ingest summary statistics periodically. Recognize and react to individual deviations from expected normals. Seamless incorporation of pertinent data into electronic health record. Ability to respond to user questions/concerns.

^#There is variability among device types listed regarding the level of validation data supporting their accuracy in measuring the listed parameters.

^*FDA Clearance refers to FDA allowing a device to market through the 510(k) process based on substantial equivalence to a legally marketed predicate device. No respiratory digital devices have been granted De Novo designation or have been approved via the Pre-Market Assessment (PMA) pathway.

^**As noted in the text, clinicians and health systems will need to educate patients to recognize ‘alarm’ conditions that require a response; in turn, clinicians and health systems will need to respond to increased patient queries regarding alarm conditions.

Personal physiologic sensors, such as wrist-worn wearables, are a key technology for the development of solutions that can improve chronic condition management, as they enable care determinations to be based on personal, rather than population, ranges of normal values. While the majority of wrist-worn wearables currently available are consumer devices with minimal to no regulatory review of the physiologic and behavioral metrics they track, newer devices such as Corsano (https://corsano.com; Switzerland), Empatica (https://www.empatica.com; U.S.) and Biobeat (https://www.bio-beat.com; Israel) are examples of multiparametric wrist-worn wearables that have, or are seeking, regulatory approval in the U.S., Europe, and beyond for all of their physiologic measurements. All of these wearables continuously track a range of vital signs and activity and have been found to allow for the early detection of respiratory infections.^13,14 For individuals with chronic respiratory conditions, daily use of these sensors can allow for recognition of changes in their vital parameters, such as increases in their usual nocturnal respiratory rate, which can potentially herald exacerbations.¹⁵ Looking at the larger picture, aggregated continuous data might increase our understanding of variabilities in respiratory diseases and contribute to more refined phenotypes.

Beyond wearable physiologic sensors, multiple additional technologies can play an important role in the creation of digital health-enabled solutions for people with respiratory illness. “Smart” inhalers are digital monitoring systems connected to inhalers that can instantaneously transmit an array of inhaler usage information to a connected device in a user-friendly manner. Considering that an estimated ^~75% of people who use metered-dose inhalers use them incorrectly and/or inconsistently, smart inhalers have the potential to significantly improve care.¹⁶ Current capabilities of smart inhaler systems, beyond recording exactly when an inhaler was used, include next-dose reminders and feedback on technique. That information can also be securely shared with the healthcare provider. Studies of smart inhalers have shown they can improve adherence, although primarily only when linked with feedback and reminders.¹⁷

Higher levels of outdoor air pollution, even when well below regulatory levels, are associated with increased risk of asthma and COPD exacerbations.^18,19 While information from stationary outdoor units can provide a general index of local air pollution, they do not capture individual pollutant and allergen exposures that are work-, school-, or home-dependent. Indoor pollution exposure to particulate matter and increased nitrogen dioxide concentrations also increase exacerbations.^20,21 A number of indoor air quality sensors, both stationary and wearable, are available commercially and could provide more precise exposure histories and real-time information.²² Early studies have found that HEPA filtration can reduce personal exposure to fine particulate matter (PM < 2.5 μm in diameter) by ^~50%, supporting the potential value of monitoring indoor air quality given that there exists a practical solution to address elevated particulate levels.²³

A combination of two or more of the technologies mentioned above, as was done with Lynnae, can enable a closed-loop system that fosters personal empowerment of the individual with respiratory disease. For example, in the U.S. AIR Louisville program, the combination of smart inhaler use tracking and monitoring of local outdoor air pollution levels led to a 78% reduction in rescue inhaler use and a 48% improvement in symptom-free days.²⁴ The development of new systems of care, built around a menu of technologies to create personalized, closed-loop solutions for each individual, and supported by 24/7 tiered connectivity and on-demand educational content, has the potential to improve outcomes and lessen the burden of care activities while minimizing exacerbations and improving quality of life.

During a recent uptick in COVID-19 cases in her community, Lynnae’s husband, Leonard, through his participation in a clinical study, was alerted of early signs of a potential acute respiratory infection based on changes from his normal baseline resting heart rate, movement, and sleep. He was prompted to get a PCR test, which came back positive, leading him to quarantine to avoid exposing Lynnae to COVID-19 and exacerbating her asthma.

Early Identification of Acute Respiratory Illness

As was the case with his wife, Leonard’s use of DHTs was central to early detection and corrective action. His viral illness diagnosis was possible through knowledge of his usual physiologic response during routine daily activities rather than waiting for his vital signs to fall outside of the range of the population norm (e.g., respiratory rate >20 BPM or temperature > 38°C) (Figure 3). Early data from the U.S. and other countries, most recently gained through COVID-19–related research programs such as the one Leonard was participating in, support the potential value of identifying subtle deviations in a person’s physiology and behaviors as an earlier indicator of an acute viral illness, in some cases even before any noticeable symptoms.^25–28

Figure 3 -.

A number of studies, such as CovIdentify (https://covidentify.covid19.duke.edu) and DETECT (https://detect.scripps.edu) have demonstrated that subtle but individually detectable changes in multiple physiologic parameters and behaviors via wearable sensors are associated with acute COVID-19 infection, occasionally preceding the presence of any symptoms.

While the COVID-19 pandemic rapidly accelerated the implementation of DHTs in research programs, it also surfaced important challenges. For example, coincident with the COVID-19 pandemic, several popular consumer smart watch companies released new functionality to monitor blood oxygen saturation. While timely, these functionalities were released as “wellness” rather than “healthcare” tools, with no evidence of their accuracy reported to, or examined by, regulatory agencies such as the U.S. Food and Drug Administration (FDA).²⁹ Particularly confusing, several of these devices have other functionalities that are FDA-regulated (e.g., irregular heart rhythm detection), making it difficult for patients and practitioners to understand how to interpret the blood oxygen saturation or other biometric information being reported.

The Need to Better Address Existing Health Disparities

Another challenge highlighted by a large number of wearables studies in COVID-19 was the reliance on a bring-your-own-device approach in which only those who already owned a wearable device could participate. The major benefit of this design was that it rapidly accelerated the collection of data from a large number of participants without typical geographic barriers. However, because owners of wearable devices are not representative of the general population, results of these studies would be more likely to contribute to, rather than reduce, existing health disparities.³⁰ As a specific example, optical sensors to measure blood oxygen saturation may fail in people with more melanin and for people with the sickle cell trait,^31,32 both characteristics common in Black populations. Therefore, there is a clear need for digital health studies that sample from populations that are representative of the target population where the sensor is intended to be used. For example, a recent, small study focused on employing a connected smart inhaler intervention in Medicaid-enrolled children with asthma in southwest Detroit found a significant reduction in rescue inhaler use.³³

The Need for Pragmatic Clinical Effectiveness Trials

FDA clearance of DHTs in the U.S. reflects that the technology in question measures the variables as described in their labeling. Their relation to clinical outcomes and value in patient management, however, require additional study. Although gradually increasing, the current number of clinical validation studies of DHTs in respiratory conditions is quite small, with few having evaluated clinical effectiveness in improving health quality in real-world settings (Table 2). While effectiveness trials are necessary and require broad support going forward, the current scarcity of such studies should not diminish enthusiasm for their potential, especially considering that the certainty of evidence supporting the majority of guideline-recommended therapies in the 2020 Asthma Guideline update were low to moderate.³⁴

Table 2.

Studies showing the number and type of validation studies of each type of digital health technology-based measure.

	Registered on clinicaltrials.gov			Indexed in PubMed
Physiologic Measure or Device	Number of studies	[Digital Health Technologies} AND (“digital health” OR “mobile health” OR “home”)	[Digital Health Technologies} AND (“pragmatic” OR “real-world” OR “real world”)	Overall Analytic Validation studies	Overall Clinical Validation studies	Clinical Validation Studies in Respiratory Conditions
Auditory Diagnostics	535	58	14	39	9	8
Breath	7.829	889	82	7	2	2
Cough	4,920	542	78	27	7	5
Spirometry	3,276	430	46	45	11	9
Pulse oximetry	1,658	141	10	77	16	2

This table is not the result of a systematic review. Rather, our search was conducted as follows:

Clinical Trials: clinicaltrials.gov was searched in November 2023 using the search terms listed in the column headers, where DHT in columns 3 and 4 corresponds to the row name in column 1. For example, the data entry in column 3 row 1 uses the search terms Auditory Diagnostics AND (“digital health” OR “mobile health” OR “home”).

PubMed publications: These numbers come from a November 2023 search of HumanFirst’s Atlas database, a comprehensive library of studies using digital measures and technologies. Publications are flagged for review based on curated search queries aiming to surface DHT studies. Each study is then manually reviewed for the clinical population(s) evaluated, the digital measures captured, and the technologies used to capture them. If all can be identified, including the specific make and model of technology, it is included in the Atlas library.

Changing Roles for Clinicians and Patients

Digital health data are unique in their continuous and longitudinal format, which is unfamiliar to most patients and clinicians and, as noted in Table 1, will require both patients and clinicians to take on new roles. Digital health implementation will require increased patient agency and empowerment in their own care; thus the patient will need to recognize ‘alarms’ from devices, and clinicians will need to respond to increasing patient concerns about these deviations from a person’s baseline. However, with primary care providers already overburdened, requiring an estimated ^~27 hours a day to just provide guideline-directed primary care, it is unrealistic to expect that digital health data can be broadly incorporated into current systems of care in a meaningful way.³⁵ To address this obstacle to innovation, several health systems have developed dedicated digital health programs,³⁶ while a growing number of virtual care organizations have been developed, with at least one specifically focused on people with respiratory conditions (https://www.wellinks.com), built around remote monitoring and virtual care.³⁷

Data may come in the form of a single measurement per day or per week (i.e., daily or weekly summaries) or may be higher-frequency, up to the resolution of one or several measurements per minute or per second. Given the newness of this data type in clinical practice, analysis and interpretation challenges remain. One solution has been the development of “digital biomarkers” that serve to transform high-volume digital health technology data (e.g., minute-by-minute measurements of blood oxygen saturation, or SpO₂, from a pulse oximeter) into indicators of health outcomes (e.g., COPD risk). The utility of the digital biomarker is to convert large streams of multivariate data into a single indicator that is useful as a threshold for clinical decision-making and to lend context (for example, if a patient is borderline versus an extreme case). One example of this was the development of a COVID-19 Decompensation Index using wearable sensor data to alert the care team if an individual was at risk for decompensation requiring hospitalization following an acute COVID-19 diagnosis³⁸ (Figure 4).

Figure 4 -.

Detection of early indicators of decompensation optimized through machine learning with continuous data streams from a multiparametric sensor patch in individuals recently diagnosed with acute COVID-19. The COVID-19 Decompensation Index (CDI) model, based on the unique individual interactions between vital signs and behaviors, has the potential to detect early signs of decompensation relative to any single vital sign. (Developed using from Ref. 38)

Beyond the practical challenges of implementation are the emotional ones. Digitally-enabled systems of care can disrupt the traditional power balance between patients and physicians.³⁹ This participatory medical model can return agency to patients, allowing them to manage their condition in partnership with their provider, and even on their own in a more informed manner supported by individualized objective data. Given this disruption to a power dynamic established over centuries, this transition will neither be immediate or easy for patients or providers.

This is still very much a nascent field that is rapidly evolving to meet the practical needs of clinicians and patients. Cross-disciplinary collaborations bridging across the different areas required for digital biomarker development (i.e., statistical, mathematical, computational, and respiratory health expertise) are needed. Further, tools that allow for collaboration in improving algorithms, validating known digital biomarkers, and discovering new digital biomarkers will enable much-needed standardization and interoperability in this burgeoning field.⁴⁰ Finally, implementation scientists and human factors engineers will need to assess the user experiences and areas where human error may obfuscate the utility of the digital biomarker in practice. Other aspects to be elucidated include the legal consequences of the medical applications of DHTs, and the inclusion and evolution of artificial intelligence in such applications.

Summary & Future Expectations

As a leading cause of morbidity and mortality globally, respiratory disease represents a significant management challenge and highlights the vulnerability of all people to environmental and pathogen triggers of decompensation. To reduce the global burden of respiratory illness will require taking advantage of a range of transformational DHTs. At the individual level, continuous measures allow for analytics comparing the present moment to well-established historical data and alerting the patient and their healthcare providers that they are moving out of their normal operating zone and allowing for changes in management. The prospects for avoiding emergency room visits and intensive care unit stays are good but there is evidence that underserved populations do not have the access to these technologies to benefit similarly to the well-served by healthcare. Further, such DHTs and physiological monitoring stand to impact not only respiratory healthcare but also a broad range of primary care and specialties areas, as deviations from personal baselines may be harbingers of a health event in all aspects of personal health. It is likely that summing digital data across communities (see https://covid19pvi.niehs.nih.gov/) will provide public health officials with the means to monitor emerging outbreaks and pandemics and may reduce morbidity and mortality from emerging environmental exposures due to climate and industry as well as from emerging infections and pandemics. Lynnae was lucky that she, her family, and her providers were early adopters of DHTs—now for the rest of the world to follow.