Happy anniversary! (Not!)
As we head into the COVID-19 lockdown anniversary — a non-celebration marking one year separated from family, friends and workplace community — one thing is clear. A full year into a pandemic that’s killed more than 500,000 Americans and put a pause on the world economy, we are still struggling with shortages of accurate, high-quality, cost-effective test options. Yet this is perhaps the single biggest key to recovery.
Most of the easy ways we do spot testing — temperature checks or symptom questionnaires — offer a 50/50 chance of identifying someone as COVID-19 positive. Those are terrible odds. And from what we know, COVID-19 is just the first pandemic we can expect to encounter in the years ahead. The time to solve the issue of readily available (and affordable), rapidly returned, and highly accurate screening options is now. We must add to our arsenal of pandemic management tools — or we are doomed to a repeat performance.
Stay Calm and Scream On
COVID-19 is testing the resolve and courage of governments. What should have been a common-sense response to the pandemic has turned out to be shockingly chaotic. Now, with Texas, Mississippi, Alabama, Montana and Iowa lifting restrictions and casting a blind eye toward the reality of sickness and risk, all of us need to remain on guard. While the US Centers for Disease Control and Prevention (CDC) have issued new guidance that people who are fully vaccinated can enjoy small indoor social exchanges unmasked, we all must remain vigilant. With the voice of science quashed and digital and racial divides starkly apparent, apolitical health groups such as the American Heart Association have rushed forward to warn citizens to take heed as COVID-19 will continue to kill.
Countries that tested early and often fared better in saving lives. The Wall Street Journal ranked and reported countries that responded well to the pandemic. A key insight in this analysis points to the continued importance of testing:
When it became clear that people without symptoms could spread COVID-19, South Korea tested early and aggressively, conducting more than twice as many tests per capita as other countries in the pandemic’s first weeks. Along with other measures, including extensive and highly effective contact tracing and quarantine, this kept cases from increasing rapidly.
Perhaps sensing the need for a creative solution to individual frustration and collective chaos, Dutch inventor Peter Van Wees created a COVID-19 “shouting booth,” which offers an alternative screening approach. On the surface, the idea seems…odd: people enter and scream inside an airlocked cabin, and the presence of the virus is confirmed using a nanometer-scale device. While screaming is emotionally therapeutic, the imaginative “COVID-19 scream test” takes three minutes, which is a lot of screaming, even for the most angst-ridden among us. And how would you like to be the second person to use the booth — let alone the 75th? Jokes aside, beyond grabbing headlines, is this approach clinically realistic or scalable?
While the scientific jury is still out on the Van Wees scream test and its accuracy, the Netherlands National Institute for Health acknowledges that “a faster, easier and accurate method is welcome.” It’s hard to imagine that airports and malls will be installing scream booths anytime soon, but Van Wees is on to something: our voices — vocal biomarkers — may be a smarter, sensible frontline COVID-19 identifier to enhance what’s in place now.
Voice as an Emerging Digital Health Science
Have a voice? Have a smart phone? That’s all that may be needed to test vocal biomarkers, a non-invasive, democratic way to track disease symptoms.
Dr. Bertalan Meskó, aka The Medical Futurist, wrote four years ago in Vocal Biomarkers: New Opportunities in Prevention:
Vocal biomarkers have an amazing potential in reforming diagnostics through their accuracy, speed and cost-effectiveness. They are able to detect some diseases earlier than an average check-up process; and an earlier diagnosis could essentially be the difference between life and death in relation to certain illnesses.
Long studied for mental health, neurological and respiratory conditions, vocal biomarkers are under consideration for COVID-19 screening. Already, an app-based smartphone interface can analyze how individual vocal biomarkers align with collective artificial intelligence (AI) data, arriving at a statistically significant negative or positive COVID-19 result. This is done by recording users counting up to 50 or 70, and while not as headline-grabbing as the scream booth, the data are more encouraging and worthy of far greater attention.
Follow Vocal Biomarker Science
More and more, researchers look to vocal biomarkers to develop accurate, affordable, accessible and consumer-friendly screens, which can be done anytime and give live-time analysis. This type of modality could address the public health challenge that asymptomatic people may unknowingly spread infection, even if vaccinated. MIT researchers reported in July 2020 that asymptomatic people differ from healthy individuals in how they cough. While not apparent to the human ear, these differences can be picked up by AI when compared to the vocal biomarkers of healthy individuals.
In their paper published in the IEEE Journal of Engineering in Medicine and Biology, the MIT team used the AI model to distinguish asymptomatic COVID-19 patients from healthy people through forced-cough recordings submitted voluntarily through smartphones:
…the model achieves COVID-19 sensitivity of 98.5% with a specificity of 94.2% (AUC: 0.97). For asymptomatic subjects it achieves sensitivity of 100% with a specificity of 83.2%. Conclusions: AI techniques can produce a free, non-invasive, real-time, any-time, instantly distributable, large-scale COVID-19 asymptomatic screening tool to augment current approaches in containing the spread of COVID-19. Practical use cases could be for daily screening of students and workers.
Vocalis Health, in collaboration with the Municipal Corporation of Greater Mumbai, announced results of another study in February 2021 that included more than 2,000 people speaking a variety of languages including English, Hindi, Marathi and Gujarati. The unblinded validation set of 288 participants demonstrated accuracy of 81.2%, sensitivity of 80.3% and specificity of 81.4%.
While the MIT and Vocalis Health studies are not comparable, their encouraging data suggest vocal biomarkers are an easy, inexpensive and accessible way to call upon digital health technologies to provide highly accurate “quick-check” tools to clear people for confident return to the workplace and public spaces, alongside the expected prevention methods.
Europe Takes the Digital Health Lead — Again
Today, there are no digital vocal biomarker technologies FDA-approved for broad use. However, in Europe, which once again outpaces the US in use of and access to digital health, these technologies are already making a difference. Countries like Germany are in the vanguard, reimbursing physicians for prescribing digital health therapies and technologies. Vocal biomarker digital devices have secured the prized Conformitè Europëenne (CE) mark for use within the European Economic Area (EEA).
Meanwhile, in the US, it’s research only. Perhaps that could change. FDA Emergency Use Authorization (EUA) can allow unapproved medical products that may diagnose, treat or prevent serious or life-threatening diseases such as COVID-19. The potential of digital vocal biomarker technology to improve screening certainly makes it a good candidate for EUA.
While we haven’t yet been able to access this useful technology, there is no reason to go into a closet and scream in frustration. To realize the potential of vocal biomarker technology, allies are needed.
Right now, billions of dollars are being invested in new medications and health ideas that have less of a chance of success than digital vocal biomarker technology. In a situation driven both by immediate urgency and long-term need, savvy investors have an opportunity to collaborate closely with digital heath pioneers, as well as people willing to quite literally raise their voices to participate in vocal biomarker clinical trials. With Apple, Google and Microsoft welcoming these apps to their platforms and regulatory officials outlining next steps, this is an area where much can be accomplished for all stakeholders, especially for the people — three billion plus globally and quickly growing — who depend on smartphones.
It is a given that testing and self-monitoring are critical in tracking viral spread and ensuring when it is safe for people to travel or to enter schools, stores or social settings. At the current rate of vaccination, even if people mask up, we may not reach herd immunity until early 2022. Even afterward, we will need to continue to use sensible, scientific approaches in determining who might be contagious and need to isolate. Consumer-friendly diagnostics are simply a must, and will be so for years to come.
The potential of smartphone-based digital vocal biomarker technology to provide the basis of expanded, accurate testing should be pressed forward. It has the capability to be useful in the case of COVID-19 in the long term, and the adaptability to be repurposed for detecting the presence of additional diseases that could result in future pandemics. In that sense, it provides the basis of a growing, global, personal-testing network. With public health authorities, investors, innovators and patients working together, we can make this happen, ideally sooner than later, because in raising our voices for access to digital health options to diagnose COVID-19, we shouldn’t have to scream ’til we’re blue in the face.
[Appreciation to Arielle Bernstein Pinsof and John Bianchi for their counsel in preparing this Being Well Medika Life article, to Shira Friedman for her thorough review, and to Dr. Robert Turner for his inspirational piece on vaccine passports as creating a global divide among nations.]
Disclosure: Vocalis Health is a client of my employer, Finn Partners. All content expressed here is my own.