Traditionally, the healthcare industry has moved slowly when embracing new technology innovations.
However, over the past few months, it has been forced to rapidly advance to confront the challenges posed by COVID-19. One area that has received increased attention is real-world data (RWD), specifically, the ways in which RWD can be used in clinical research to quickly reveal insights about treatments, outcomes, and risk factors for COVID-19.
As much of these data are already available in the form of electronic health records (EHRs), it enables us to illuminate clinically relevant trends and learnings that can be applied in real-time.
If you’ve been following the news over the last few months, one thing is abundantly clear: Things are changing daily as COVID-19 spreads around the world. As a result, frontline care providers are learning how to treat patients in real time and they need new information and additional guidance now.
Unfortunately, the speed at which we need to move is a direct juxtaposition to traditional research methods, which can take months or even years to complete. While randomized clinical trials remain the gold standard for ensuring drug safety and efficacy, they are not designed to provide real-time information to inform treatment for patients with COVID-19.
This virus is a constantly moving target, with a growing data set and a myriad of variables to consider. If we only rely on traditional research methods to improve our understanding of COVID-19, it’s essentially like entering a boxing match blindfolded for those on the front lines. By analyzing RWD from patients who have tested positive, including their treatment patterns and associated outcomes, we can begin to piece together some early learnings.
By analyzing trends in COVID-19 data, we’ve been able to glean information on what treatments might be making an impact—and which are not. For instance, one analysis showed a lack of efficacy of hydroxychloroquine among hospitalized COVID-19 patients, despite early statements supporting this treatment.
This desire for increased insights isn’t just on the frontlines—the FDA has been vocal about its desire to understand treatment outcomes in COVID-19 patients.
They’re actively working with organizations across the industry to gather actionable information. Another industry initiative that is working to put patient data to use is the COVID-19 Evidence Accelerator, launched by Friends of Cancer Research and the Reagan-Udall Foundation for the FDA. Collectively, these programs are working to fill the gaps in our understanding of COVID-19 so we can better triage at-risk patients and treat them more quickly.
Patient data can also be instrumental in ensuring that clinical trials continue on as planned. Due to the pandemic, a number of trials have been put on hold. However, RWD can be used to fill these gaps down the line so that ongoing clinical research isn’t slowed.
We are now seeing the acceptance of RWD increasing across various verticals in healthcare. Using EHR data of real patients, we’ve been able to identify risk factors and treatment outcomes trends. Having this information readily available to everyone—from frontline workers to the FDA—will be instrumental in treating COVID-19.
EHR enhancements, such as EHR templates and EHR triggers, are encouraging tools for promoting an understanding of clinical pathways to improve cancer care in the eastern region of North Carolina, according to a study published in Journal of Clinical Pathways.
The eastern region of North Carolina is a rural region of the state, where individuals are at an increased risk for more health issues, including cancer. According to the study, those individuals have lower survival rates than those who live in urban areas, and breast cancer is no exception.
“Patients with breast cancer residing in ENC [eastern North Carolina] are more likely to have delays in delivery of adjuvant chemotherapy and suffer from higher rates of breast cancer-specific mortality when compared with the rest of the state,” the authors explained. “A regional hospital system that serves this rural constituency is uniquely positioned to address these disparate outcomes.”
Researchers at Brody School of Medicine at East Carolina University wanted to ensure that newly diagnosed ENC patients with breast cancer received timely, evidence-based, and efficient care from their respective health systems by EHR optimization.
The researchers aimed to use the EHR as a tool guiding providers to follow certain clinical pathways.
Researchers integrated evidence-based clinical pathways for breast cancer care into the EHR at Vidant Cancer Care in Greenville, NC. The team also provided a means for rapid analysis of quality metrics.
Researchers collaborated with an EHR Advisory Group at numerous meetings to develop critical pathways. The stakeholders established 75 corresponding elements to individual treatment decisions and these critical pathways were then implemented into the EHR through separate approaches.
“First, standardized VBCC [Virtual Breast Cancer Conference] templates comprised of clinical pathway elements were developed and made live within the EHR,” the writers explained. “These templates were purposefully designed to enable rapid data collection and real-time analysis of variance from minimum quality standards.”
“Each element included in the templates was designed to be analyzable as a discrete variable via use of checkboxes and drop-down menus or as a qualitative variable via use of free-response text boxes,” they continued. “This design allows for rapid extraction of data elements via EHR queries. In addition, specific EHR triggers were developed to encourage compliance with clinical pathways and VBCC recommendations.”
Researchers analyzed 435 new breast cancer cases for two years prior to the implementation to gain a baseline. After three months, researchers will run an EHR query to analyze the results. This three-month re-evaluation will continue routinely at stakeholder meetings.
Early data hasn’t been particularly revealing, but researchers said there is upside to these EHR enhancements.
“Baseline data analysis revealed subpar compliance with multiple elements,” the authors wrote. “Gemba walks demonstrated that the standardized VBCC template is user-friendly. EHR triggers are currently in the beta testing phase. While VBCC participation by physicians at nonflagship hospitals has not been robust, EHR enhancements are a promising tool for promoting adherence to evidence-based clinical pathways, thereby improving the quality of breast cancer care in ENC.”
In particular, researchers said the EHR triggers will be an asset to this study once the triggers depart from beta testing phase.
“We anticipate that our EHR triggers will successfully improve compliance because similar triggers have been demonstrated to be effective mechanisms for identifying deviations from quality standards in cancer care,” the writers explained. “We intentionally limited the number of triggers to two in an effort to prevent “click fatigue,” which is associated with increased patient safety hazards and physician burnout.”
“Once beta testing is complete and the triggers go live, we will solicit feedback from providers about the perceived benefits of the EHR triggers relative to the additional burden on their workflow.”
Christopher A. Brown
A global health crisis can lead to global cooperation, if we work together to create sustainable change
The COVID-19 Pandemic has exposed a chink in the armor of the US healthcare system. It has become clear that our healthcare network is woefully fragmented, with the lack of cooperation between organizations underscored by the absence of data interoperability putting both the general public and care practitioners at a disadvantage. While the free sharing of data between patients, caregivers, and providers is not a panacea for all current and future health challenges, interoperability can certainly help us to flatten the curve.
COVID-19 is the very definition of a population health crisis. To truly understand this disease, we need to analyze it and see its effects from every angle: responses to drugs, mortality rates, comorbidities, patient responses to intubation and other treatments, and what happens after they come off ventilators and return to the general population.
Models are being developed that can leverage this information to produce actionable insights and stratify patients for risk, but we need data to use these tools effectively. Although COVID-19 is like nothing we have seen before, we can seek to refine information that has already been collected on the spread of infectious disease within and across populations. Some technology leaders, along with innovative healthcare partners are working to optimize and normalize this data for validation with the goal of revealing signals and trends which can result in better outcomes for patients.
The current options for data collection have one thing in common: they require a patient to show up. We ask them to log on and fill out a questionnaire, or to physically present themselves before a doctor. While such practices serve well enough for triaging of patients, we must look for ways to leverage the data that already exists, without relying on continually adding new data to the equation. This pandemic has uncovered the importance of true preemptive population health, rather than the reactionary “calling in sick” health.
Locking up the data
An individual’s healthcare records are commonly delivered across multiple settings. Throughout a patient’s journey, their medical records may lead from a private physician’s office to an imaging center, hospital, outpatient center, and back to the original physician, each step producing its own records. For quality care, accessible information must accompany the patient every step of the way. If all these twists and turns exist within just one patient’s medical journey, how much more so of a whole population? And how can we combat the spread of diseases among the population if this data is inaccessible and disorganized?
The infrastructure of our care system is fragmented. Imagine city planners designing New York City’s transit system without communicating – the roads, the subway, the train all built independently of one another would lead to chaotic intersections and more than a few deaths. So why do we provide healthcare to 328 million Americans based on data gathered in such an irrational way?
Other industries have long shown the way in creating a common and mostly secure platform for data access and sharing. Our advancing technology in healthcare now has the potential to transform our healthcare system into one that continuously learns and improves, using predictive analytics and decision support tools. However, without access to the necessary data, these tools will be useless.
Free sharing of data promotes interoperability
The barriers that exist when it comes to sharing data are not trivial. Issues relating to privacy, legality, ethics, economics, and politics must all be considered. While we do not have all the solutions to these obstacles, priorities must be determined, and the public health considered alongside these challenges.
The key to improving patient outcomes and managing population health is interoperability, the transfer of information seamlessly between multiple sources. Leveraging healthcare’s full potential requires the free (and secure) sharing of data with the technologies able to analyze it effectively.
For example, if a patient regularly sees a cardiologist, but between appointments is admitted to the ER, the cardiologist should automatically have this information available. Interoperability of data will allow us to avoid unnecessary duplication of services, reduce costs, and provide safer and more compassionate care. Interoperability will not only help individual patients but the collating and sharing of robust health records will support more accurate tracking and prevention efforts for a range of public health threats.
The free exchange of data affords providers lifesaving and cost-effective insights, enabling them to make beneficial care decisions. Improving patient outcomes and reducing cost is not just about having the data but putting that data in the right hands to provide actionable insights.
Looking to the future
The World Health Organization’s eHealth Resolution includes the aim “to foster exchange of data and information for promotion of health, health systems, and training of health-care workers.” Effective exchange of health information is vital to improving the capabilities of hospitals and the health of populations.
When the dust settles, health systems will be reeling from the physical, emotional and financial burden caused by COVID-19. Instead of a well-deserved break, staff will be inundated with a second wave of patients, this time from a backlog of innumerable medical procedures that were put on hold throughout this pandemic. This demand will hit providers hard, but it will also provide an opportunity to utilize data to stratify patients at-risk for specific diseases or downstream complications. We need to prioritize these patients, so hospitals can do the right work on the right people without wasting precious resources.
Failure to learn from this pandemic will mean we miss an opportunity for real advancement in medical practice. The tools which will be most important in the future are those that can analyze existing population data and provide answers without relying on a constant stream of new data. The future is in sight, but without better access to the health records currently held in silos, we cannot optimize these algorithms effectively. Sharing data freely is the key to curbing this pandemic and mitigating the effects of any future outbreaks.
Hospitals are having a hard time reporting data to public health agencies, according to a new JAMIA study, which finds patchwork data sharing, "often occurring via fax or phone."
Hospitals may encounter administrative or logistical hurdles when reporting data to public health agencies – which, in turn, can hinder essential information tracking in times of infectious disease outbreaks.
In a study published in the Journal of the American Medical Informatics Association, Harvard Business School researchers used data from the 2018 American Hospital Association Annual Survey and IT supplement to examine the barriers hospitals faced when trying to meet meaningful use requirements.
One significant challenge, researchers found, was the ability of agencies to receive the data hospitals were mandated to send.
"More than four in 10 U.S. hospitals report that public health agencies are unable to receive electronic data," researchers wrote.
"This finding may reflect the fact that substantial federal funding has been devoted to hospital information technology adoption, including the ability to send data electronically, without a concomitant investment in the ability of public health agencies to receive and act upon this data."
WHY IT MATTERS
The Centers for Medicare and Medicaid Services requires hospitals to send data to city or state health departments. These requirements, researchers noted, were implemented in part to allow agencies to respond efficiently to epidemics and pandemics.
Still, they write, reporting gaps exist, and the novel coronavirus outbreak has made those inefficiencies clear.
"Despite billions of dollars in federal investment in digitizing the U.S. healthcare system, aggregating information such as test results and potential cases was done in a patchwork way, with data sharing often occurring via fax or phone," researchers wrote.
"Had electronic data sharing been in place, hospitals could have quickly transmitted COVD-19 testing results and syndromic surveillance data to public health agencies to supplement their testing and provide greater clarity on disease prevalence and incidence," they continued.
Most of the 3,512 hospitals surveyed, the research team found, reported at least one barrier to sharing electronic data with health agencies, with public health agencies' capability being the most common hurdle. Interface-related issues, such as costs or complexity, were the next most common issues.
Other problems cited included challenges extracting the data from electronic health records, different vocabulary standards, hospital capacity and lack of information about which public health agency should receive the information.
"One state had no hospitals reporting public health agency inability to receive electronic data while several states had the majority of hospitals reporting that barrier," researchers explained.
"Differential funding levels priorities for public health agencies at the state and local level may explain some of this geographic variation," they continued.
Researchers also noted that some public health agencies may have only been able to receive data from certain EHR systems, which could explain the variation in answers even within the same state.
THE LARGER TREND
Researchers have noted the importance of data interoperability in containing the effects of COVID-19 around the country.
Earlier this month, a Duke-Margolis Center for Health Policy research team recommended better commercial lab reporting, improved access to clinical data and more reliance on the National Syndromic Surveillance program to manage the continued spread of COVID-19.
And although CMS has offered some flexibility for implementing interoperability requirements, experts say the crisis has only highlighted the need for efficient information sharing – including with patients.
"The COVID-19 pandemic gripping the nation underscores the importance of these regulations in enabling greater data exchange and providing patients with their information," said Ben Moscovitch, Pew's project director for health information technology, in April.
ON THE RECORD
"Many areas reporting barriers to public health receipt of electronic data are also projected to be overwhelmed by COVID-19 patients, indicating that it is not just low density or rural areas who lack critical IT infrastructure for electronic disease surveillance," said researchers in the JAMIA study.
"Policymakers should prioritize investment in public health IT infrastructure along with broader health system information technology for both long-term COVID-19 monitoring as well as future pandemic preparedness," they continued.