Digital Defenders: COVID-19 Creates a Spotlight on Digital Health Tools
Digital health tools like telemedicine, AI algorithms, and connectivity are playing an increasingly important role in helping prevent and manage disease.
By Michael Barbella, Managing Editor - Read original article here
Inder Singh’s tipping point arrived fully loaded on Day Eight.
Sidelined with a mysterious illness for the previous week, Singh decided that day to take matters into his own hands, vowing to deliver the world from the “dark ages” of infectious disease control and prevention.
“We have no idea where illness is starting or spreading,” Singh told Forbes last spring. “I realized this at a system level...and then on a personal level when I came down with a mysterious illness. It’s shocking when you realize how much real-time, crowdsourced information we have about so many aspects of our lives—from how much our neighbor’s house costs to which side street will get us home fastest. And yet we have no idea what’s circulating around us—something that would have helped when doctors were trying to treat me after eight days of 104 [degree] fever. Infectious illness is the leading cause of death in young children around the world...it’s the biggest problem in healthcare globally. The only way to stop illness from spreading is to know where and when it’s starting.”
To find those epicenters, Singh founded a company that invented an artificial intelligence-powered “smart” thermometer capable of real-time illness tracking. The thermometer is equipped with an app that allows users to self-report symptoms and can estimate flu-like maladies throughout America at any given time. The instrument also can predict disease patterns weeks before the U.S. Centers for Disease Control and Prevention (CDC), with equally accurate results.
Developed by San Francisco-based Kinsa Inc., the smart thermometer aggregates tens of millions of anonymized data inputs from nearly 1 million households and uses AI to triage illnesses based on symptom severity, duration, and communicability. Kinsa then uses that collected data to create a real-time health map of disease outbreaks according to zip code.
Singh’s company has distributed its thermometers to thousands of elementary school student families to help them better track the spread of illnesses in their communities. Pharmacies have used Kinsa’s data to determine the kinds of drugs to stock, and healthcare providers have found the technology helpful in keeping non-critically ill patients from seeking emergency treatment.
Ultimately, Singh hopes the data will be used by public health agencies to better track disease. “The sad fact is, we have almost zero accurate information about where and when disease is starting or spreading. It’s all based on models that are pretty inaccurate,” Kinsa’s CEO said in a February 2019 Fast Company interview. “How do you curb the next epidemic if you don’t know where and when it’s starting? You can’t.”
It’s virtually impossible, actually, as the world is just now discovering with the COVID-19 pandemic. The most effective mitigation efforts thus far are those rooted in comprehensive testing: South Korea, home to the second-highest number of Asian cases (behind China), quickly ramped up its testing efforts upon confirmation of its first case. Within two months of its initial infection, the country had conducted roughly 300,000 novel coronavirus assays and was performing 15,000 nasal swabs daily.
Compare those numbers with U.S. testing output: 60,000 total and 7,000/day between Jan. 20 (first reported case) and early March. Not surprisingly, South Korea has flattened its infection curve and avoided the draconian lockdowns currently in place throughout America.
“You cannot fight a fire blindfolded,” World Health Organization Director-General Tedros Adhanom Ghebreyesus said at a March 16 news briefing in Geneva, Switzerland. “We have a simple message for all countries—test, test, test.”
Certainly, more testing will be necessary to flatten the curve, but digital health technologies will be equally as important in slowing transmission rates. Case in point: A smartphone app enables South Koreans to self-report symptoms and government officials to track quarantined citizens, while text message is the dominant delivery tool for newly-infected patients.
Similarly, China took full advantage of the latest digital technology to bring COVID-19 under control, using web-based tools like Weibo, Tencent, and WeChat to disseminate accurate information about the virus, and moving half of all medical care online to help reduce in-person clinic visits. Moreover, Chinese hospitals have begun using AI software to detect visual signs of COVID-19-related pneumonia on lung CT scan images.
Other countries are following suit: Israel, the United Kingdom, and the United States are increasingly digitizing patient visits, and a newly-formed task force at Stanford University has pledged to work with digital health firms on COVID-19 prevention efforts and treatments.
“The coronavirus, it’s such a negative aspect of our lives [in terms of the effect] on other people,” virtual reality surgery pioneer Dr. Shafi Ahmed said in March, “...actually, I think there is an opportunity here to rethink the whole model of the way we see our patients, certainly around technology.”
Maybe so, but the virus is just as likely to reshape the future role of digital health. Since the turn of the millennium, advancements in data collection and analysis, medical connectivity, and wireless technology have been used to improve patient outcomes. Healthcare digitization also has proven effective in assessing disease burden, tracking outbreaks, and supporting infection control/prevention measures, though it has never truly been tested on a massive scale.
Digital health has become a powerful weapon in the fight against COVID-19, as it has been harnessed for diagnostic test design, genome sequencing, and vaccination development. The Vere-CoV detection kit from Singapore-based Veredus Laboratories, for example, is based on portable lab-on-chip technology capable of detecting MERS (Middle East Respiratory Syndrome), SARS (Severe Acute Respiratory Syndrome), and COVID-19 in a single (blood) test.
Aidar Health’s assay employs cloud-based AI and machine learning to spot and measure early COVID-19 symptoms such as fever, cough, and shortness of breath. The company’s MouthLab device features complex firmware, a digital thermometer, and breath sensors that gauge breathing rate, coughing patterns, and respiratory flow cycle morphology to assess shortness of breath.
AI, however, could hold the key to defeating the virus.
Within the first month of its birth, scientists had sequenced COVID-19’s genome (SARS took several months), and by early March, Google’s DeepMind lab had applied its deep (machine) learning algorithms to predict the structure of proteins associated with SARS-CoV-2, the virus that causes COVID-19 (such knowledge is crucial in vaccination development).
AI also is tackling the mountain of literature that has been generated about the virus to help communicable disease specialists and epidemiologists answer “high-priority” scientific questions related to COVID-19. More than 29,000 articles have been written about the disease, more than 13,000 of which have full text.
“This valuable new resource is the fruit of unselfish collaboration and now offers the opportunity to find answers to important questions about COVID-19,” Dr. Dewey Murdick, director of Data Science at Georgetown University’s Center for Security and Emerging Technology, said in announcing the cross-team data mining effort on March 16. “Once the crisis has passed, we hope this project will inspire new ways to use machine learning to advance scientific research.”
‘Just a Matter of When’
In many respects, the worsening COVID-19 pandemic is validating the value and efficacy of digital health tools. Both the CDC and WHO, for example, are urging medical facilities to expand their use of telehealth services—otherwise known as remote or virtual care—to help triage sick patients and minimize the influx of non-critical ailments. WHO also has set up its first Digital Health Department (its inaugural meeting was held last fall), and the agency has partnered with WhatsApp and Facebook on a public messaging service that potentially could reach a quarter of the world’s population (two billion people). The service provides the latest news/data on COVID-19 including details on symptoms and protection measures, as well as the latest situation reports and real-time numbers. It was developed in collaboration with Praekelt.Org using Turn machine learning technology.
Telemedicine and public messaging is just the tip of the digital health iceberg. though. The rise of digital technologies such as AI, robotics, virtual reality,/augmented reality, portable sensors, wearables, smartphone apps, gamification, and remote monitoring has led to the development of innovations like cancer-detecting smart probes and connected inhalers to atrial fibrillation-identifying algorithms and automated population health trackers.
“Digital health is equivalent to the term mobile health in that it utilizes web, mobile, mathematical, and statistical technology as well as general computer technology to listen to data and to use various algorithms that can be simple or complex to drive care processes or disease management,” explained Dr. Mark Clements, chief medical officer at Mountain View, Calif.-based Glooko, creator of a diabetes management solution that provides an app for the person with diabetes, remote patient monitoring, and healthcare provider tools for decision support and population health management. “It’s also a tool for marrying the person with chronic disease with interventions that themselves may not be digital. I view it as putting more of the power of technologies in the hands of the person with diabetes to support their self-care without having to come for a clinic visit every time.”
Glooko’s diabetes app syncs with 95 percent of all diabetes devices. Designed for easy data tracking, the app helps improve clinical workflow by aggregating data from nearly any device type—blood sugar monitors, insulin pumps, continuous glucose monitoring systems, fitness devices, and smart insulin pens.
Downloaded to a smartphone or tablet, the app shows patterns and trends in blood sugar readings and allows persons with diabetes to create food/meal logs and set reminders for glucose checks or medication dosing. Important information like blood glucose levels, meals, physical activity, and medication is displayed in a logbook screen that shows a near real-time history. Importantly, the app allows syncing readings from multiple devices.
Clinicians can access the data in person (office visits) or remotely. They can also create notification alerts for specific events like fluctuating blood sugar levels. “We’re solving a cognitive burden for the person with diabetes by helping them make sense of the voluminous data that is generated by their own body when monitoring glucose [levels], recording their insulin doses, and recording their foods, etc.,” Clements noted. “We have a real opportunity here to distill that down into a narrative and a message that is easily digestible. Otherwise when you’re looking at hundreds of data points a day, unless you are a data scientist it becomes impossible to digest. I think one of the other problems we’re solving is the issue of getting care more proximal to the person with a chronic disease so they are empowered to do more to improve their health and to engage more and rely less on the healthcare system and clinician.”
Self-empowerment and greater patient engagement are just two of digital health’s numerous by-products. Others include new disease identification, more efficient disease management, lower healthcare costs, and improved quality of life. One of the more obscure benefits is the potential insights that would be gained through the creation of vast data repositories. Through AI and data analytics, researchers could one day definitively connect the dots between lifestyle and chronic health conditions—associate, for instance, factors like diet, profession, residency, and medications with specific medical conditions or illnesses.
The possibilities are truly mind-boggling: Last year, Stanford University Medicine enrolled more than 400,000 Apple Watch users in the United States to study the device’s irregular heart rhythm notification (the tool occasionally checks cardiac rhythm and notifies users of irregular patterns indicative of atrial fibrillation). As part of the study, participants with irregular heart rhythms received a notification on their Apple watch and iPhone, a telehealth consultation with a doctor, and an electrocardiogram (ECG) patch for additional monitoring.
Study results showed 0.5 percent of study participants received an irregular heart rhythm notification; the pulse detection algorithm (indicating a positive tachogram reading) had a 71 percent positive predictive value. Eighty-four percent of those receiving the notifications were experiencing atrial fibrillation at the time, and 57 percent sought medical attention. One third (34 percent) of participants who followed up with the ECG patch later were found to have Afib.
“The results of the Apple Heart Study highlight the potential role that innovative digital technology can play in creating more predictive and preventive health care,” Lloyd Minor, M.D., dean of the Stanford School of Medicine, said in announcing the study results. “Atrial fibrillation is just the beginning, as this study opens the door to further research into wearable technologies and how they might be used to prevent disease before it strikes—a key goal.”
Surely, disease prevention is perhaps the holy grail of digital health, but better disease management an important objective too. Medtronic plc and Fitbit developed an app that enables type 2 diabetics to obtain their blood glucose levels and physical activity data in one streamlined application. Philadelphia medtech firm RTM Vital Signs LLC has developed a miniature implantable long-term cardiovascular monitoring system that produces data to support medication dosing, diagnostics, and even predictive analytics. The device was one of a half-dozen finalists that competed for top billing in the 2019 Boston Scientific Connected Patient Challenge; its cardiac monitoring device features a real-time implantable blood pressure sensor equipped with a long-term monitoring system that measures the arterial blood pressure waveform, the electrocardiogram, respiratory rate, tidal volume, hemoglobin oxygen saturation, and sounds of the heart, lungs and upper airway. Sensor data feeds predictive algorithms to improve diagnostics and medication dosing related to hypertension and other cardiac issues. Relevant data is sent to a smart device and a central monitoring station for advanced analytics by machine learning algorithms and a clinician.
Cardiac monitoring also is the intent of software developed by Voorheesville, N.Y.-based Lille Group. The company’s Cardiac Signals software is designed to help cardiology offices better manage patients and their implantable cardiac devices, improve outcomes, increase (office) productivity, and boost efficiencies. “Cardiac Signals is a specific medical software program that helps cardiology groups in hospitals and private practices manage cardiac patients,” said Dennis Vetrano, Lille’s sales and business development director. “Cardiac patients with a pacemaker or an ICD have to be monitored...the devices have to be monitored. There has been a push to get away from a server-based system. Our software helps nurses in the cardiology office and cardiologists themselves monitor these patients, manage any events, and bill out for these patients. There’s been a big push in the industry to get patients out of the hospital quicker and keep them at home, especially with COVID-19 pandemic. These are patients that are at higher risk to have serious complications if the virus was acquired, so keeping them home and monitoring their cardiac device and health remotely reduces their risk significantly by not entering a hospital or physician office.”
And that, in turn, frees up precious resources now needed to help control the pandemic. Using telehealth services to monitor high-risk patients also could save America’s bloated healthcare system money—as much as $46 billion by some estimates.
Some of the ways digital technologies can save healthcare systems money is by improving patients’ adherence to medications and reducing acute care episodes. Medication non-compliance costs the U.S. market more than $100 billion annually, and the global healthcare system over $270 billion each year, according to market analyses. Higher levels of adherence are associated with lower overall healthcare costs for numerous chronic diseases including diabetes, hypercholesterolemia, and hypertension, as well as chronic obstructive pulmonary disease and severe asthma.
“Digital health is integrating across the life sciences industry supporting today’s major health issues. We see digital health integrating into the pillar modalities of the Diagnostic, Therapeutic, and Connected Health continuum,” said Anthony Kalaijakis, strategic industry marketing manager for the Molex Phillips-Medisize MedTech business, an electronic components and solutions provider. “The top morbidity factors in medtech according to the World Health Organizations remain: Cardiovascular—heart disease; Diabetes mellitus—high blood sugar; Hypertension—high blood pressure; and COPD. These ‘lifestyle’-influenced morbidities can be significantly improved with digital health tools. Payers and providers are united in maximizing efficiency by utilizing the connected health continuum. Arguably, the major disruptors are the user interface (mobile device), the evolution of the cloud (data centers), and population health management (software/user interface). The ability to modulate from mass population data to individual personal needs brings an entirely new paradigm.”
Population health management is likely to dominate innovation efforts now with the COVID-19 pandemic in full swing. New telehealth services seem to be available daily, and a steady stream of new tools is flooding the market to help monitor the virus’ spread and facilitate better treatment. Digital solutions also are on the way from tech giants Amazon, Facebook, and Google, which recently met with WHO officials to define their roles in combating SARS-CoV-2.
“The current crisis certainly makes the need for digital healthcare solutions more apparent. Digital products will be a part of the healthcare ecosystem and we believe they will improve care. It’s just a matter of when,” said Brad Womble, chief of staff, senior director of Strategy and M&A for Jabil Healthcare, a global manufacturing solutions provider that delivers comprehensive design, manufacturing, supply chain, and product management services. “The current pandemic brings the need for more remote solutions to the forefront. Prior to March 2020, providers were opening up in small ways to telehealth, but quarantines are making it vital to care delivery.
Although COVID-19 is the overwhelming topic in health today, there’s still a population with other diseases and conditions who need to consult with their care teams.
The social distancing mandate has made remote care a reality today for many doctors and their patients. This crisis may prove to be a tipping point in the acceptance of remote care and digital health for those who have resisted change.”