Digital Trials to Forefront, EHRs Uniquely Bury US Docs, Telehealth Helps APMs-The HSB Blog 12/22/20
isDigital Clinical Trial Technology Saves Money, Improves Speed to Market and Increases Trial Diversity
“Our Take”: The COVID pandemic has successfully made an impact in the adoption of digital clinical trials/decentralized clinical trials (DCTs) especially as people/participants had to socially distance themselves amid the quarantine. While DCTs have been around for a number of years, it is during this pandemic the importance and relevance of DCTs have really come to the forefront. With more than 905 trials in the U.S. postponed as a direct result iof COVID lockdowns researchers stressed the need for new strategies for conducting clinical trials as the then-current method wasn’t sustainable. This led pharma companies and clinical research organizations to look into a modernized clinical trial. The fast pace technology in digital trials has made drug discovery quicker, cheaper, more inclusive, and is expected to be maintained in the long term.
Description: Clinical trials have historically been done by a very labor intensive, manual process. Just to participate in trials, patients often have to travel long distances to centralized trial sites in order to undergo evaluations to determine if they meet appropriate eligibility criteria. Even when they meet clinical eligibility criteria, results from Health Union’s clinical trial programs indicate that 40%-60% of potentially qualified patients are excluded due to location restrictions. Once chosen, trial participants must then endure a long process of follow-up visits, and meticulously keep track of medications, symptoms and adverse drug reactions in a personal diary in order to report data to investigators. By contrast DCTs inherently overcome many of the challenges of traditional trials (in person follow-up visits, time and travel burden on participants, participant recruitment and retention limitations, and inflated costs). According to over three quarters of respondents from a Pharma Intelligence survey the pandemic has increased the use of remote trials and more than 90% said they expect the increase to be maintained in the long term. Moreover, digital trials have given rise to a number of interesting partnerships, For example Novartis and Microsoft have formed an alliance to work to develop drugs using AI, while Stanford, Apple and telehealth company Amwell have joined forces to study atrial fibrillation by collecting data on irregular heart rhythms through the Apple watch’s heart rate sensor. DCTs have the potential to make clinical trials more impactful by making research more patient-centric, creating more meaningful data, and increasing participation by underrepresented communities. The three most common “virtual components” used in DCTs are mobile technology (92%), web-based patient diaries (84%), and wearable technologies (82%).
Implications: DCTs can improve the trial process in a number of ways including: 1) significantly reducing trial costs (ex: site staff and operation, site-directed data collection and management); 2) shortening participant recruitment time and expense; 3) increasing participant retention; and, 4) decreasing time and travel burden on participants themselves. For example, given the average distance for patients to travel to trial sites is 50 miles and a typical requirement for them to visit these sites is approximately 15-20 times over a 6 month period, therefore patients can end up devoting 2-3 hours per week to travel alone. With the ability for patients and study coordinators to connect virtually to sites, DCTs have made the process much easier by eliminating travel time as well as reducing required time-off from work. In addition, technologies such as sensors and apps embedded in mobile devices make the tracking process easier and faster, eliminating the hassle of handwritten journals. Digital tracking and monitoring tools also improve compliance and side effect monitoring, allowing more rapid changes to trial protocols should they become necessary. Moreover, the application of digital tools such as applying machine learning to drug discovery in clinical trials, can speed up the process of drug discovery resulting in dramatic savings. For example, according to the U.S. General Accounting Office, the use of machine learning in drug discovery can yield cost savings of approximately $300-$400 million per trial (or approximately 20%-30% per successful drug). Lastly, with approximately 48% of trial missing enrollment targets and 49% of patients dropping out of trials before completion, the reduced burden on trial participants from the use of DCTs should not only allow sponsors to improve these numbers but also allow them to recruit from a more diverse patient pool thereby increasing the mix of what has generally been underrepresented groups in clinical trials.
The Impact of COIVD-19 on Virtual Trials; The Future of Decentralized Clinical Trials;Artificial Intelligence in Health Care: Benefits and Challenges of Machine Learning in Drug Development
U.S. Clinicians Spend 50% More Time in EHR than those in other Countries
Event: On December 17th, Healthcare IT News reported on two recent studies citing EHRs and frequent appointment overrun as main contributors to clinician burnout. The first found American clinicians spend an average of 90 minutes a day actively using the Electronic Health Record (EHR), while their non-U.S. peers spend about an hour a day in their EHRs. The second found the average primary care exam was and that the average primary care exam was 18 minutes, approximately 1.2 minutes longer than their non-U.S. peers. The authors of the separate studies in the Journal of the American Medical Association (JAMA) and Medical Care respectively, conclude that time spent inputting information into EHRs and the propensity of clinicians to run late due to appointment overrun are more severe for U.S. clinicians than for their peers.
Description: The JAMA article showed that Americans spend 50% more time on electronic health records than other countries. For this study, researchers from Harvard and Stanford University analyzed the data of 371 ambulatory care health systems around the world. The sample included all clinicians with scheduled patient appointments, including physicians and advanced practice practitioners. This study found that U.S. clinicians spend significantly more time than those not in the United States performing four clinical activities: notes, orders, in-basket messages and clinical review. U.S. clinicians composed more automated note text and received an average of 34 messages a day, compared with the non-U.S. clinicians' 13 messages a day. The Medical Care study shed some light on how the minutes using EHR translate to time spent with patients. It showed that primary care Wasappointments set by US clinicians typically overrun the allotted appointment slot. Visits scheduled for 10 or 15 minutes were more likely to exceed their allotted time than those scheduled for 20 or 30 minutes. These overrun appointments could be due to US clinicians spending more time on EHR use. As a result, researchers are investigating time spent on EHRs as a cause for overrun appointments and burnt out clinicians.
Implications: These studies suggest that US EHRs can be inefficient and can be causing delays in appointments. Whether these inefficiencies are technical or policy-related, they should be investigated and updated to allow clinicians to complete tasks in a timely manner. The US may lag behind in EHR efficiency because it does not have a single payer healthcare system like the other countries in the study do. As a result, U.S. clinicians' EHR time may be compounded due to America's complex and multi-faceted healthcare system. However, the studies also note that additional EHR time could be because US clinicians are under-trained. In situations where providers suspect this to be the case, hospitals should ensure their clinicians are regularly trained and aware of EHR usage and intractability.
Using Telehealth to Enhance Current Strategies in Alternative Payment Models
Event: The COVID pandemic dramatically increased the usage of telehealth in medical care as a means to keep patients safe and systems from becoming overwhelmed. In the spring of 2020, in-person visits declined due to quarantine-imposed precautions on visits to physical facilities dealing with the spread of the virus. Telehealth visits increased in virtually all fields to fill the gaps and were adopted across the country by provider organizations of all sizes. While originally deployed as a means of last resort in many cases, evidence supports the possibility that increased telehealth usage had driven lower emergency department visits and could be associated with expanded access. Moreover, researchers have theorized that while previous efforts to expand access via APMs have been resource attentive, policy makers can use telehealth to explore the promise of reducing acute care utilization via expanded primary care access. Strategies, such as on-demand visits, text-based symptom surveillance, and remote monitoring, could expand access more cost efficiently. Telehealth has proven to be useful for enabling access for hard-to-reach patients. Telehealth further advances the feasibility of alternative payment models, which are designed to improve quality and decrease costs. Telehealth allows providers to better coordinate care for patients and manage costs as a result.
Description: Telehealth and alternative payment models both focus on improving access to primary care. Existing research such as the Comprehensive Primary Care (CPC) initiative indicates that adopting certain standards for primary care can reduce emergency department visits, as patients can undergo preventive measures before the need for costly hospitalizations. Telehealth advances both the access and continuity principles of CPC, by allowing physicians to see patients where resources or distance may be a limiting factor. In addition, this could also apply where patient’s physical conditions limit their ability to see providers such as in the case of those with disabilities.Telehealth can overcome these limitations without the need for costly interventions like the need to provide transportation. Telehealth services can also support the population health management efforts that are part of accountable care organizations (ACOs). A form of APM, ACOs encourage providers to take responsibility for controlling costs in treatment they implement. Managing chronic illness in a patient population through telehealth strategies like remote patient monitoring can help reduce cost and improve quality of care by reducing or preventing complications through more engaged care. ACOs focused on cost containment have also adopted different strategies for discharge decisions. Post-surgical care can shift from inpatient to at-home with remote physical therapy or telehealth monitoring thereby allowing patients easier access to their providers at lower costs.
Implications: The rise of telehealth is promising for the future of alternative payment models. Telehealth services can increase access to care for vulnerable patients and enable better management practices for providers. The potential to reduce cost while improving quality is encouraging. Yet, in order for telehealth to become fully established as a means to increase access to care, the infrastructure on both the provider and patient ends must exist to support it. Providers need to ensure they adequately support technology necessary to maintain telehealth operations, and help bridge the gap between underresourced patients who may not have the same digital resources as other, more advantaged populations. In addition, providers also need to ensure that necessary accommodations are made for those with disabilities or seniors who may need additional help in using or implementing the necessary technology.
Amazon Wants to Offer Primary Care to Other Employers
Event: On December 16th, Business Insider reported Amazon plans to provide online and in person primary care to other large employers. According to the article, this is an extension of Amazon Care, a program originally offered exclusively to Amazon employees in Seattle beginning in September 2019. While representatives from Amazon have not yet commented on the extension, if true, this could be a major move both for Amazon and larger companies looking to provide greater access to care for employees at an affordable price.
Description: Currently, Amazon Care has only been introduced to employees in the Seattle, Washington area. Care is provided through an app that connects users with a clinician via video calls or through text messaging for mild health issues including cough and fever. Employees in the greater Seattle area have in person care access in addition to the virtual options as well as the option to have prescriptions delivered to their homes. With the expansion of this offering to other larger employers, the Amazon Care in-person services would become available to workers living within a certain proximity to the headquarters and virtual components similar to what is offered to those living outside of Seattle will be made available to those living in other cities or states. Currently, Amazon funds video care, care chats, and mobile care services for Amazon employees, but employees must be enrolled in Amazon healthcare insurance to receive services. Under this expansion of services, Amazon would charge employers fees based on the number of employees being covered and using the services on a monthly basis.
Implications: With the introduction of Amazon Care to large companies outside of Amazon, Amazon Care can offer convenient health care services competitive to those offered by traditional players. Additionally, this expansion is another vehicle Amazon can use to infiltrate the healthcare delivery system and use its services to determine the best way to take advantage of its artificial intelligence and data management skills and apply them to healthcare. Along these lines Amazon has a long history of undertaking pilot projects to learn industries and data flows and then determine the most advantageous ways to apply its skills to the industry. As noted in our piece on the Amazon HealthLake, Amazon views healthcare as a meaningful revenue contributor and growth business in the future and we view the provision of employer benefits as one piece of that equation.
Amazon Wants To Offer Primary Care to Other Employers & Amazon Plans to Provide Health Services to Workers at Other Companies
Study Reveals Disparities in Coverage and Accuracy Among Symptoms Apps
Event: On December 16th, MobiHealthNews reported on a study published in BMJ Open, that tested the coverage, accuracy, and safety of eight symptom assessment apps: Ada, Babylon, Buoy, K Health, Mediktor, Symptomate, WebMD, and Your.MD. The study noted that most apps did not correctly identify the possible conditions, about 38% of the time. It should be noted that while the study was peer reviewed, “all of the authors, with the exception of [one], are or were employees of, contractors for, or hold equity in the manufacturer of one of the tested apps (Ada Health GmbH).”
Description: Members of the public are increasingly using symptom checker apps to self-diagnose their ailments for both non-urgent and urgent symptoms. A peer-reviewed study conducted by a team of doctors and scientists led by the global digital health company Ada Health compared condition coverage, accuracy and safety of suggested conditions, and appropriateness of urgency advice to the level of human general practitioners (GPs). The study looked at how comprehensively the apps covered possible conditions and user types, finding that Ada was the most comprehensive app, providing a condition suggestion 99% of the time (compared to GPs providing 100% coverage). In comparison, other apps tested provided a suggestion 69.5% of the time, with the lowest scoring 51.5%. The study also found that the apps’ clinical accuracy was highly variable, finding that Ada was most accurate, suggesting the right condition in its top three suggestions 71% of the time, while the average across the other apps was 38%. With the exception of Ada, most apps did not correctly identify possible conditions, while human GPs are accurate 82% of the time. While most apps typically gave safe advice, only three apps, Ada, Babylon, and Symptomate (80%), performed close to the level of human GPs (97%).
Implications: According to a survey from 2010, a total of 12 countries have reported that 75% of people search for health information online, with two-thirds of patients noting that they “Google” their symptoms before going to the doctor's office. Symptom assessment apps support the clinical setting by reducing the burden on our healthcare system and improving outcomes. The utility of digital symptom assessment apps has seen rapid uptake by users in recent years due to their ease of use, convenience and availability. Symptom checkers can help relieve patients' concerns promptly and narrow their focus on conditions by explaining their symptoms but must return accurate results. Coverage is an important measure for digital health tools that might be deployed at scale because many tools exclude users who are young, too old, pregnant, or have pre-existing mental health conditions. The findings presented from the study above show that no digital tool can outperform human GPs, but potential enhancements can be made in AI-based symptom assessment technology. In addition, while the above research rank orders the apps based on accuracy, given the authors potential conflicts of interests (even though disclosed), further research by investigators unaffiliated with any of the companies under study would be recommended to validate the study’s findings and rankings.
Comments