Insights Series Issue 14: "Precision diagnosis in the COVID-19 era and beyond"
This thought leadership paper describes the current challenges of diagnostics and presents four matching approaches to overcoming them, both for COVID-19 and for the practice of medicine.
Insights Series Issue 14 siemens-healthineers.com/ expanding-precision-medicine Stephen Bruin Stephen Bruin The New Normal 2021 4 2020 --- -- < 2014 << 2008 Actionable < 2006 4 Comprehensive 3 Data 2 True-to-life < 1999 < 1998 Precision diagnosis in the COVID-19 era and beyond Timeline A thought leadership paper on how to improve diagnostic accuracy SIEMENS Healthineers Preface The Insights Series The Siemens Healthineers Insights Series is our preeminent thought leadership platform, drawing on the knowledge and experience of some of the world’s most respected healthcare leaders and innovators. The Series explores emerging issues and provides you with practical solutions to today’s most pressing healthcare challenges. We believe that increasing value in healthcare – delivering better outcomes at lower cost – rests on four strategies. These four principles serve as the cornerstones of the Insights Series. Expanding Transforming Improving Digitalizing precision care patient healthcare medicine delivery experience Our Insights portfolio is an integrated collection of events, speaking engagements, roundtable discussions, and an expanding array of print and digital platforms and products all carefully curated to share ideas, encourage discussion, disseminate original research and reinforce our position as a healthcare thought leader. Please visit siemens-healthineers.com/insights-series Executive summary The COVID-19 pandemic has brought an intense focus on diagnostics, highlighting the relevance of accurate and timely diagnosis for increasing the quality of care and reducing healthcare- and community-related costs. The pandemic has exposed four key factors that limit the ability to make precise, data-driven diagnoses: low-quality data, non-comprehensive data “snapshots”, non-actionable information, and a siloed approach. This thought leadership paper describes these challenges and presents four matching approaches to overcoming them, both for COVID-19 and for the practice of medicine. Practical recommendations include: ensuring hospitals generate timely and accurate data, obtaining a comprehensive and longitudinal view of the patient, transforming data into relevant diagnostic insights, and ensuring hospital leaders support an enterprise-wide approach to the relevance and the systemic execution of diagnoses. In the case of COVID-19, the learning and implementation of these four approaches have been leveraged by the overwhelming pressure on delivering a timely and accurate diagnosis. The solutions described here are generalizable beyond COVID-19 and can have a positive impact across all specialties. Siemens Healthineers Insights Series · Issue 14 3 Challenges The challenges 1. Low-quality, inaccurate, or incomplete data/test results COVID-19 presents numerous diagnostic challenges: the need for rapid turnaround time for results, the surge in demand for test kits, the diagnosis of asymptomatic patients, and the limitations of the tests themselves (i.e., false negatives during the “window” period just after infection). In addition to these known challenges, there are unknowns – does 2. Diagnostic data that misses exposure confer immunity, what is a patient’s long- the “big picture” term prognosis after COVID-19, and what is the “best” diagnostic approach? Diagnostic challenges are not unique to COVID-19. By some estimates, in the medical practice patients may be medically misdiagnosed 10%-15% of the time¹, making it likely that most people will experience at least one diagnostic gap in their lifetime. Equally important, ? the diagnosis might not be specific enough to identify diagnostic subcategories to guide a tailored treatment for the patient. 3. Diagnostic information that is too complex or specialized to be actionable The disruptions to healthcare in the wake of COVID-19 have only served to highlight the importance of accurate diagnosis for every medical condition and for every patient. 4. Siloed and departmentalized diagnoses 4 Issue 14 · Siemens Healthineers Insights Series “Half of the medicines we use don’t work. The reason is we are actually treating the wrong disease.”² Paul Rothman, MD, PhD, Dean of the Medical Faculty & CEO, Johns Hopkins Medicine, U.S. 1 Low-quality, inaccurate, or incomplete data/test results Test-related factors like low sensitivity or specificity or data subjectivity limit diagnostic accuracy. In the case of COVID-19, the window period with polymerase chain reaction (PCR) and low antibody specificity might lead to underestimates or overestimates of infection/exposure due to false positives or false negatives. And limited access to test collection supplies, coupled with limited availability of both reagents and appropriate analyzers, prevents some hospitals, clinics, or even regions from assessing the true extent of the COVID-19 pandemic. Reimbursement issues and lack of contact tracing infrastructure can keep public health officials from learning who needs to be tested for COVID-19 in order to limit further community transmission. The challenges of data quality and quantity are not unique to COVID-19; they are common across the practice of medicine. Siemens Healthineers Insights Series · Issue 14 5 “Acquired data often insufficiently represents the individual patient in front of the physician.”³ Prof. John D. Halamka, MD, MS, Professor of Medicine, Harvard Medical School 2 Diagnostic data that misses the “big picture” Lack of patient longitudinal data limits diagnostic and tracing efforts, making it more difficult to prevent outbreaks in a patient’s community or workplace. Too often, diagnostic data is only a snapshot, a picture of a specific factor or system at one point in time. When the COVID-19 pandemic erupted, a patient’s medical history and travel history were not always available when the patient arrived at the care facility, making it more challenging to accurately diagnose COVID-19. A specific sub challenge is the lack of access to comprehensive information at the point of decision. The data infrastructure in many healthcare organizations makes it difficult to bridge the information gap between departments and organizations. In many countries, the need for an integrated digital infrastructure that makes all relevant information available to the caregiver and the patient has become painfully visible during the COVID-19 pandemic. Limited information on patients’ behavioral history and family life can hide risks, potentially limiting patients’ compliance to specific treatments or to preventive measures like self-isolation. To deliver the best care to the patient and their community, a more thorough longitudinal understanding of that patient as a person and as a member of their community should be available to healthcare providers. 6 Issue 14 · Siemens Healthineers Insights Series “Data doesn’t do you any good until you can turn it into information, and that is really our challenge.”⁵ Stanford Medicine (2017): Health Trends Report: Harnessing the Power of Data in Health 3 Diagnostic information that is too complex or specialized to be actionable COVID-19 challenges include assessing a high volume of data from many patients. Complex tasks, like assessing chest imaging with suspicious readings, as well as support in interpreting conflicting results (on molecular and antibody tests, for example), must be completed in a timely manner in order to appropriately diagnose and care for the individual patient. ? In medicine in general, the total volume of data is growing at a rate of 48% per year, presenting a huge challenge to healthcare providers, who may be acting on outdated information. Many hospitals and health systems lack the scalability, performance, and analytic capability to support clinical decision-making and to make timely and targeted care interventions.⁴ Siemens Healthineers Insights Series · Issue 14 7 4 Siloed and departmentalized diagnoses Many tests for PCR and antibodies were developed, but not all were properly validated. Even now, there is no Diagnosis is too often seen as a siloed department-level test that is 100% accurate, so we need to use the most process, not a systemic hospital-wide one centered on accurate tests available to limit inaccurate PCR and the patient. The issues with organizational awareness antibody results. on the relevance of diagnosis, as well as limitations of digital infrastructure mentioned in our previous Insights There are still a lot of unknowns – for example we don’t publication on public health, play a part here, limiting know if antibodies confer immunity and if so, for how the speed and specificity of many diagnoses. long; and we don’t know how specific immunity may be to mutated virus strains. Continuously improving the Only 19% of the organizations surveyed – 1 in 5 – are quality of data that we generate on COVID-19 is vital to fully equipped to make enterprise-wide data-driven enhancing our understanding of the disease and how to decisions, based on a recent survey.⁶ prevent and treat it. The relevance of these factors on non-transmissible COVID-19 and the need for precision diagnosis conditions is built upon the same pillars for improving diagnostic accuracy in medicine. For example, in At the outset of the COVID-19 pandemic, many questions oncology: “The first step in identifying the right cancer concerning when, how, and who to test needed to be treatment and achieving the best possible outcome is addressed nimbly, across organizations, cities, and even making the most accurate and precise diagnosis,”⁷ said countries. Clinicians had no choice but to quickly learn Marshall Hicks, MD, professor and head of the Division of to improve overall accuracy and efficiency in the Diagnostic Imaging at MD Anderson. diagnostic process. 8 Issue 14 · Siemens Healthineers Insights Series Solutions to improve diagnostic accuracy 1. Improve data quality 3. Generate actionable insights from large and complex data sets 2. Provide comprehensive longitudinal 4. Build executive support and enterprise data at the point of decision culture for precision diagnosis Siemens Healthineers Insights Series · Issue 14 9 Solution 1: Improve data quality The two main types of testing for COVID-19 are PCR testing for active infection and antibody (or serologic) testing to determine recent or prior infection with SARS-CoV-2. Together, these two tests can help health agencies get a clear picture of the state of the pandemic. For accurate diagnosis, these tests must be validated and must offer a sensitivity and specificity close to 100%. Lower specificity in areas with low exposure to the virus might lead to over- estimates of community exposure due to false positive results. Accurate and widespread testing helps assess community status and identify infection “hot spots.” PCR testing Molecular testing detects How much COVID-19 testing is needed? The recommen- viral RNA, indicating an dation from health organizations and thought leaders is active disease state to conduct widespread testing in order to identify and contain outbreaks.⁸ In many countries, an initial gap or learning curve was identified between the ideal number of tests for miti- gation and suppression and the actual testing being done. Further strengthening of testing programs was Serologic (antibody) necessary to close these gaps. testing Detects patient΄prior Beyond improved tests, data should transition, when exposure to SARS-CoV-2, possible, from non-structured or qualitative into the virus that causes structured or quantifiable. Quantifiable data allows COVID-19 for more precise diagnosis and subclassification. For example, generation of quantitative or semiquanti- tative results for COVID-19 antibody titers will be important to characterize the immunity, if any, conferred by prior exposure to SARS-CoV-2. 10 Issue 14 · Siemens Healthineers Insights Series Solution 2: Provide comprehensive longitudinal data at the point of decision a) Getting the true picture of the individual patient COVID-19 taught us that the diagnostic process does not need to occur strictly in a practice or hospital. It can begin at home, for example, using a smartphone or a PC, bolstered by digital tools such as patient portals, digital tracers, screening algorithms, point of care technology, remote access, and telemedicine. This broader approach to diagnosis, and to understanding the “whole patient” incorporates lifestyle data, which can be crucial to mitigating the spread of infectious disease. For patients with suspected COVID-19, the picture of the “whole patient” includes previous tests, travel history, and tracing data. Smart watches can provide additional data, including heart rate and temperature changes, that can be important in early screening the potential presence and severity of infection. In other conditions, similar insights can be gained by incorporating sensors that track exercise, diet, sleep patterns, heart rhythms, and more. This data can be analyzed and visualized to present a more complete longitudinal picture of the patient to allow more tailored and proactive interventions. Comprehensive longitudinal data is especially important in a rapidly evolving situation like COVID-19, where our understanding of the virus and the disease it causes are developing quickly. Access to comprehensive data can help identify important trends that emerge as our understanding increases. Siemens Healthineers Insights Series · Issue 14 11 b) Making comprehensive data available at the Digital enablers and infrastructure are needed to point of decision guarantee secure access across settings (e.g., The relevant longitudinal patient data discussed above ambulatory, hospital admissions, home care, cell should be readily available when making urgent phones). Healthcare provider organizations must invest decisions in any healthcare setting, including telehealth in digital enablers for secure information transfer, and or home care settings. more importantly, must develop an enterprise-wide strategy for secure data access and sharing. Seeing the whole patient: Gathering data and insight across multiple traditional Transmit data Evaluate data Evaluate data and remote care settings Collects data to health care and notify and notify on patients provider in a health care health care remotely different providers providers Task (e.g., at home) location as needed as needed Radiologyist analyzing imaging (e.g., CT scans, x-rays) Urgent care physician conducting a virtual visitin with a patient Asthma patient using a movile app to track symptoms for her personal review Patient communicating with primary care physician via a secure patient portal Hypertensive patient using a remote blood pressure cuff that automatically transmits information to physician Diabetic patient using a mobile app to track blood surage readings that are transmitted to endocrinologist Parkinson‘s disease patient wearing a sensor that measures bradykinesia and transmits information to neurologist Modified based on Harvard Business Review, July 2020: How to Make Remote Monitoring Tech Part of Everyday Health Care, Sanders MD et al. 12 Issue 14 · Siemens Healthineers Insights Series “The more comprehensive and structured the available information is, the better the basis for decision-making and the better the decisions made on this basis are, as a rule.”⁹ Schneider, UK: Einrichtungsübergreifende elektronische Patientenakten: Zwischen Datenschutz und Gesundheitsschutz Solution 3: Generate actionable insights from large and complex data sets High-quality data is important, but what we do with it is treatment. Incorporating factors that may influence patient even more so. Physicians must be able to translate data into compliance or, in the case of COVID-19, disease-spreading actionable insights to deliver personalized medicine. behavior, is a vital part of precision medicine. Digital tools can help evaluate patients’ likelihood to adhere to For example, consider chest computed tomography (CT) treatment and can also be used to promote improved scans or X-rays. AI-powered decision support systems can adherence. help radiologists identify suspicious areas for evaluation or even suggest diagnoses that could be considered, based on analysis of many similar cases. This is especially important for patients with rare conditions, atypical presentations, or confounding physiological or pathological factors. COVID-19 transcends the definition of a “respiratory disease.” Its symptoms can mimic other conditions (i.e.: gastrointestinal disease, flu, conjunctivitis), making it exceedingly challenging to diagnose and treat. Clinical decision support algorithms can help healthcare providers arrive at the right diagnosis in the face of such complexity and may help avoid ruling out COVID-19 in atypical presentations. As mentioned earlier, it’s vitally important to incorporate other aspects of the patient’s life and health history. We’ve learned from COVID-19 how the economic and behavioral consequences of the pandemic triggered an unprecedented impact in mood problems and stress, affecting many activities of daily life. A proper evaluation of mental health is key to accurate and actionable diagnosis and effective Siemens Healthineers Insights Series · Issue 14 13 Did you know? AI-powered tools can automate quantitative measurements of imaging data. There are tools that can even perform risk stratification to guide treatment. Real-world ways to make complex data actionable Proactive identification of higher-risk subpopulations: Digital enablers: AI-based tools can support complex An infectious disease suppression strategy includes integration of multiple data sources and comparison of aggressive, consistent testing of high-risk persons so individual patient data with aggregated data sets to communities can contain emerging COVID-19 case streamline and improve clinical decision-making. Today, clusters faster. there are tools available that aid in a specific field, as in aiding in interpretation of imaging data, and other tools Integrating data sources from multiple regions and that incorporate multiple data sources (imaging, lab, from as many cases as possible allows national health EHR, etc.) to suggest a treatment pathway. organizations to form effective strategies for identifi- cation of and outreach to high-risk populations. This can apply to population health concerns as well. Streamlined access and integration of multiple data “I’m much more interested in figuring out who is likely sources, coupled with automation, can accelerate to be infected and bringing the tests to them rather than contact tracing in the wake of COVID-19 diagnoses, testing an entire state,”¹¹ explains Jennifer Nuzzo, an which is key to suppressing potential outbreaks. epidemiologist at Johns Hopkins. Robust testing of high- risk groups could help pinpoint diagnoses, as well as “Despite technical and regulatory challenges, the appa- facilitate prevention and early detection. rent clinical and economic utilities of CDSS (clinical decision support systems) must lead to greater engage- ment. These tools play the key role in realizing the vision of a more ‘personalized medicine’, one characterized by individualized precision diagnosis...”¹⁰, Arnaud Belard, Chief of Operations – USU at The Henry M. Jackson Foundation, MD, U.S. 14 Issue 14 · Siemens Healthineers Insights Series “The diagnostic process is a complex, patient- centered, collaborative activity that involves information gathering and clinical reasoning with the goal of determining a patient’s health problem.”¹² Solution 4: Build executive support and enterprise culture for precision diagnosis Precision diagnosis is a systemic process, requiring buy-in During the COVID-19 pandemic, engagement and at every level of the organization. Top management support at the executive and enterprise level accelera- involvement and endorsement are essential to garnering ted the realization of the relevance of the diagnosis. In support for and adoption of a precision diagnosis culture parallel, the CDC and other agencies provided templates and the necessary integration of care teams and infra- for addressing the overall diagnostic process, including structure to enable that culture to thrive. “This process subclassification of patients based on clinical severity occurs over time, within the context of a larger healthcare and guidance for appropriate management. These were work system that influences the diagnostic process.”¹² important first steps toward adopting an integrated, enterprise-wide approach to precision diagnosis. For COVID-19, the flow of real-time diagnostic Diagnosis: Its systemic components information across departments and healthcare professionals allowed the needed integration of care teams in the diagnostic process in different clinical Organization settings (home care, nursing care, ER, surgery, etc.). culture and leadership Physical and digital Patient and HC Enabling environment Professionals technologies Diagnostics guidelines Siemens Healthineers Insights Series · Issue 14 15 Conclusions COVID-19 has put a spotlight on specific challenges 3. Give clinicians help in assessing complex data. AI-based in diagnosis and the rewards of improving diagnostic clinical decision support systems can help physicians accuracy and speed. The ability to make a data-driven process the potentially overwhelming burden of diagnosis is one of the most fundamental skills in the complex data in order to understand underlying healthcare armory. Low data quality, non-compre- pathologic processes and treat patients accordingly. hensive data, overly complex or non-actionable data sets, and a lack of organizational infrastructure can 4. Obtain C-level buy-in on the enterprise-wide relevance hamper physicians’ ability to consistently make data- of precision in diagnosis. Leaders of healthcare driven diagnoses. A fragmented view of a patient enterprises should clearly demonstrate the value of must give way to a more comprehensive, longitudinal a systemic approach to diagnosis to the heads of perspective for precise characterizations of diseases. departments to foster an integrated approach. And a fragmented view of diagnostic processes must Encourage a patient-centric approach and cross- give way to an integrated, enterprise-wide approach departmental collaboration to deliver the best results. to precision diagnosis. The way forward is clear and rewarding, but it is by no We identified four action-oriented insights to be means easy. Some of these solutions are departures from considered by executive leadership at hospitals “business as usual” and require executive support, as we and others healthcare institutions: experienced with COVID-19. 1. Ensure that hospitals generate accurate, high- We are at a crossroads, where the digitalization of quality data in terms of specificity, sensitivity, medicine is making integrated data-driven approaches and quantification. to diagnosis and treatment a real possibility. Technology is also enabling decentralized care and precise remote 2. Consider processes and technology to capture monitoring, so that physicians will have greater insight comprehensive and longitudinal data. This includes than ever before into the mechanisms of disease and continuous data at non-hospital locations (home, how they affect patients’ lives. gym, at work, ambulatory, ER, ICU) and leveraging new wearables, like smart watches and other sensors, Now is the time for healthcare leadership to adopt a when appropriate. While complying with privacy and precision diagnostic mindset, to embrace the integrated security, a more comprehensive, longitudinal approach approach to diagnosis, and help their care teams develop to patient health data supports proactive, preventable, robust insights for effective and proactive diagnosis and and early diagnosis decisions. prevention. 16 Issue 14 · Siemens Healthineers Insights Series Suggested readings Lasalvia L., Merges R. (2020) Insights Series, issue 11: Moving toward precision in managing pandemics: 5 critical domains for success in public health. Available on siemens-healthineers.com/insights/news/ moving-toward-precision-in-managing-pandemics.html i Information The Siemens Healthineers Insights Series is our preeminent thought leadership platform, drawing on the knowledge and experience of some of the world’s most respected healthcare leaders and innovators. It explores emerging issues and provides practical solutions to today’s most pressing healthcare challenges. All issues of the Insights Series can be found here: siemens-healthineers.com/insights-series Contact If you have further questions or would like to reach out to us, please do not hesitate to contact our expert directly: Reto Merges Global Head of Expanding Precision Medicine [email protected] Siemens Healthineers Insights Series · Issue 14 17 About the authors Luis Lasalvia, MD Reto Merges Vice President, Global Medical Officer Global Head Expanding Precision Medicine Siemens Healthineer Siemens Healthineers Dr. Luis Lasalvia’s experience in clinical practice combined with his With more than 10 years’ leadership experience in healthcare versatile career in pharmaceutical and medical device technologies marketing, Reto Merges has a strong track record in building has been inspiring and driving the delivery of concrete high value. effective teams for clinical and innovation marketing. In addition, He’s led a variety of numerous scientific, business and partnership he has four years of work experience in China, ramping up efforts teams, served as an advisory board member, and developed large for research collaborations in China and South Korea. He holds a programs and new ventures, including more than 30 novel and large degree in electrical engineering and information technology from programs in the U.S. and multiple countries in Europe. the Karlsruhe Institute of Technology, Germany, and has studied at the Nanjing Normal University, China. His scientific background is in Dr. Lasalvia has been a keynote guest speaker, panelist, and the field of medical imaging, where he has authored many moderator at about 500 events and conferences around the world, publications and holds multiple patents. has authored more than 50 peer-reviewed papers and articles in prestigious publications, and has submitted multiple patents in Europe and the U.S. He lives in New York and is a Medical Doctor from University of the Republic in Montevideo. He holds a Master’s degree in International Business from Pompeu Fabra University in Barcelona; and Executive education at The Wharton School of Business in Philadelphia, New York University, New York City, and Harvard Business School in Boston. 18 Issue 14 · Siemens Healthineers Insights Series References 1. Frelick M, Vega CP (2020) How Big a Problem Is Misdiagnosis in Medicine? Medscape Education Clinical Briefs Medscape.org. 2. Rothman, P (2019): Siemens Healthineers Executive Summit, Expanding Precision Medicine. 3. Halamka, J (2019): Siemens Healthineers Executive Summit, Expanding Precision Medicine: nexus of measures, models and forecast. 4. Katz (2015) How Healthcare Big Data Lakes Aid Population Health, Patient Care. 5. Stanford Medicine (2017): Health Trends Report: Harnessing the Power of Data in Health. 6. Digital Diagnostics Transformation: What’s next? Harvard Business Review (2020). 7. MD Anderson News Release. mdanderson. org. (2014). MD Anderson Opens Diagnostic Imaging Center in West Houston. 8. NPR. As Coronavirus Surges, How Much Testing Does Your State Need To Subdue The Virus? (2020). 9. Schneider, UK: Einrichtungsübergreifende elektronische Patientenakten: Zwischen Datenschutz und Gesundheitsschutz (2016). 10. Belard A et al. Precision diagnosis: a view of the clinical decision support systems (CDSS) landscape through the lens of critical care. J Clin Monit Comput. (2017) Apr;31(2):261- 271. 11. Coronavirus by the numbers (2020) NPR. As Coronavirus Surges, How Much Testing Does Your State Need To Subdue The Virus? 12. Improving Diagnosis in Health Care, National Academies of Sciences Engineering, and Medicine. (2015). Improving Diagnosis in Health Care. Washington, DC: The National Academies Press. Siemens Healthineers Insights Series · Issue 14 19 At Siemens Healthineers, our purpose is to enable healthcare providers to increase value by empowering them on their journey towards expanding precision medicine, transforming care delivery, and improving patient experience, all enabled by digitalizing healthcare. An estimated five million patients worldwide benefit every day from our innovative technologies and services in the areas of diagnostic and therapeutic imaging, laboratory diagnostics and molecular medicine as well as digital health and enterprise services. We are a leading medical technology company with over 120 years of experience and 18,500 patents globally. With about 50,000 dedicated colleagues in over 70 countries, we will continue to innovate and shape the future of healthcare. Siemens Healthineers Headquarters Siemens Healthcare GmbH Henkestr. 127 91052 Erlangen, Germany Phone: +49 9131 84-0 siemens-healthineers.com Published by Siemens Healthcare GmbH · Printed in Germany · 0920 · ©Siemens Healthcare GmbH, 2020
- presicion medicine
- precision medicine
- patient experience
- value promises
- expanding precision medicine
- reduce unwarranted variations
- Tranforming Care Delivery
- digitalizing gealthcare. improving the patient experience
- increase workforce productivity