In an emergency, every minute counts. Coordinated, real-time electronic reporting from hospitals and other health care facilities to state and federal agencies enables better operational decision-making and ultimately saves lives. Such technology exists today and the time to implement it is now.
For the past two months, based on a request by Vice President Pence, hospitals and states have begun reporting capacity, utilization, and “in house” laboratory testing data to the Federal Government. The reporting process is burdensome for hospitals and states, who are requesting better processes and a reprieve from the requirement. The Situational Awareness for Novel Epidemic Response (SANER) Project is an industry-wide collaboration that aims to revolutionize outdated and unreliable data-sharing processes, reduce the burden on hospitals and states, and improve real-time situational awareness during the COVID-19 pandemic and all future public health emergencies.
The Initial Request
On March 29th, 2020, Vice President Pence sent a request to hospitals on behalf of the President and the White House Coronavirus Task Force. Recognizing the critical role of hospitals on the frontlines of defending against the COVID-19 pandemic, the letter asked all hospitals to report on (1) COVID-19 testing performed in “in-house” laboratories, and (2) data about Patient Impacts and Hospital Capacity. The testing data is reported daily by 5pm, by spreadsheet, to the Federal Emergency Management Agency (FEMA), while the impact and capacity data is requested through the National Healthcare Safety Network (NHSN) at the Centers for Disease Control and Prevention (CDC) at the same time each day.
The NHSN collection is a series of three forms, or “pathways” as they are referred to by CDC. These pathways cover data elements related to patient impacts and hospital capacity, healthcare worker staffing, and healthcare supplies. Each pathway is estimated to take about 25 minutes to complete and can be delivered either by manual data entry, a CSV file by an individual facility, or a bulk CSV file upload for multiple facilities.
On May 8th, 2020, the Centers for Medicare and Medicaid Services (CMS) published their interim final rule requiring CMS-certified long term care facilities (LTCFs) to report a COVID-19 module at least once every seven days. Facilities were required to submit the first set of data to CDC no later than May 17th, 2020, and were subsequently granted a two-week grace period.
Today, just a short period after Vice President Pence’s request, figures from the CDC COVID Data Tracker do not match figures reported by individual states.
As someone who has participated in several federal-level disaster response efforts, here are a few likely reasons for this discrepancy:
- Manual data entry leads to errors. From experience, I can tell you that manual reporting (particularly via spreadsheet) is rife with errors and omitted information. If you don’t want to take my word for it (understandably) then check out this review or this review – turns out this is a well-studied phenomenon.
- Manual data entry takes time. Each of the reporting pathways is expected to take about 25 minutes for a cumulative total of 75 minutes – or more than an hour. The testing report and the module for long term care facilities do not include a time estimate, but it is safe to say the reporting time is probably not negligible. For hospitals, LTCFs, and other healthcare facilities that are concerned with staffing shortages and patient surge, every minute and every hour is valuable. This is likely one several reasons that more than half of the states have a facility reporting rate of less than 60 percent.
- Reporting is not mandatory. Apart from the new LTCF reporting requirement, data reporting is voluntary. The consequence is missing data. Missing data can lead to bias and an unrepresentative sample, leading practitioners to draw inaccurate inferences about what the data actually means. While CDC notes that statistical methods including “weighting (to account for non-response), multiple imputation (to account for missing data), and a running 7-day smoothing technique (to account for daily fluctuations and updated responses in reporting to NHSN)” have been applied, they do not provide additional information regarding the effect of these limitations on operational decision-making.
- Reporting requirements evolve. In disasters and public health emergencies it is very common for knowledge to evolve over time. When reporting processes do not keep up, data quality can be compromised. For example, the COVID-19 testing spreadsheet asks for “Cumulative Positive COVID-19 Tests” but does not distinguish whether that figure should include diagnostic tests, serology tests, or both. Presumably, this is because diagnostic testing was much more prevalent than serology testing at the time the request was made.
So why are facilities being asked to report this way? Well, because tough decisions must be made and we haven’t invested in a better system. The reality is that, local, tribal, state, and federal governments currently use a patchwork of clinical and public health reporting systems and few are interoperable with one another.
On the clinical side, there has been over a decade of investment in the advancement of EHRs. Today, modern health IT systems routinely share patient-level data seamlessly within networks. However, rarely do hospitals and healthcare systems share patient-level data with one another, and rarely must they share the data with state and federal partners. In fact, the need for better information sharing among competitors has been cited by health executives as a barrier to healthcare system preparedness.
LTCFs, who were not part of early meaningful use incentives, are not as advanced in their health IT infrastructure. While almost every single hospital in the country has demonstrated meaningful use of EHR, just 64 percent of skilled nursing facilities used an EHR in 2016; of those, only 30 percent sent or received clinical health information. Experts have recently called for the expansion of regulation and incentives to apply to nursing homes, labs, and state health departments.
Speaking of state health departments, while reporting components such as lab results, immunizations, and syndromic surveillance have been supported by promoting Interoperability/Meaningful Use overall public health budgets have seen continuous cuts as has funding specifically for preparedness and response activities. As we are now seeing, the decimation of public health infrastructure has had major ramifications on their ability to respond to this global pandemic.
At the crux of the issue is the lack of a national approach to an electronic public health situational awareness network. You may be surprised to learn that Congress has supported such a network in legislation since 2006. The Pandemic All Hazards Preparedness Act (PAHPA) calls for HHS to “establish a near real-time electronic nationwide public health situational awareness capability through an interoperable network of systems to share data and information to enhance early detection of rapid response to, and management of, potentially catastrophic infectious disease outbreaks and other public health emergencies that originate domestically or abroad.” This call was re-upped in the law’s 2013 reauthorization as well as in its 2019 reauthorization.
Unfortunately, such a system has yet to be realized. In 2010, the Government Accountability Office (GAO) published a report citing HHS’ lack of progress, and in 2017, they published a follow-up report with a similar conclusion.
State and federal authorities need a higher level of confidence in the data they are using to make critical operational decisions about the allocation of healthcare. Now, in the middle of a global pandemic, the need for a more resilient healthcare system and more robust data sharing systems has become obvious. However, the lack of interoperability standards and policy related to sharing critical healthcare information in disasters has long hampered the disaster response community. Whether it is COVID-19 or a hurricane, a mass shooting or an earthquake, healthcare capacity reporting is essential for state and federal governments to make decisions about the allocation of scarce resources, facility evacuations and closures, and implementation of policy waivers.
The good news is that modern technology is quite capable of more reliably moving data, securely and in real-time, from point A to point B. And, the global response to COVID-19 has mobilized the public and the private sectors in a unified direction. The time to make progress is now.
While not a panacea, efforts like the The SANER Project are a huge step in the right direction. The SANER Project, led by Audacious Inquiry, focuses specifically on improving state and federal response operations through real-time situational awareness of the healthcare delivery system. The project leverages shared interoperability standards based on HL7 FHIR, through a critical collaborative to develop an HL7 Fast Healthcare Interoperability Resources (FHIR®) Implementation Guide. Healthcare facilities, critical infrastructure and government response authorities can leverage these standards to expose data that is critical for controlling the spread of the disease and managing limited healthcare resources across the country. These data include number and status of specific hospital bed types, availability of staffing, the number of ventilators, and other critical supplies.
For those like me who do not have a tech background, what this means is NO MORE SPREADSHEETS (plus faster and more reliable access to data, a dramatic increase in data quality, a near-real-time ability to update desired data elements, and an overall greater likelihood of making the right decisions at critical moments). In other words, we can save time and, more importantly, WE CAN SAVE LIVES. With widespread participation, The SANER Project can provide the foundation for an automated data sharing system that enables seamless transition of critical capacity data from healthcare facilities to local and state health departments, as well as federal response agencies. While these data are but a subset of the data required for the “near real-time electronic nationwide public health situational awareness capability” envisioned in PAHPA (and PAHPRA and PAHPAIA) they are a very critical subset.
To truly prepare our country for all-hazards, including pandemics like the one we now face, we need a standardized, real-time, data sharing system that leverages modern technology. Until federal and state governments invest in and develop interoperability standards for the modernization of IT infrastructure in healthcare, public health, and emergency management, our national health security will be impaired. It is time to do things smarter.
If you would like to learn more about The SANER Project, check out some of our recent news and press:
- Hospital Vital Signs: The EHR Doesn’t Know Everything
- Local First, A SANER Approach
- HL7 International and Audacious Inquiry Collaborate to Support COVID-19 Response
- Audacious Inquiry Launches the SANER Project to Modernize Data-Sharing for COVID-19 Response through HL7® FHIR®
Director, Emergency Preparedness and Response
Lauren Knieser recently joined Ai after eight years in federal service supporting disaster preparedness, response, and recovery work, and she leads Ai’s Patient Unified Lookup System for Emergencies (PULSE) portfolio. In her previous role at the Office of the Assistant Secretary for Preparedness and Response, Lauren led the development of the Regional Disaster Health Response System and contributed to policy, research, training, and information management initiatives.