Open Source Solutions For Public Health Case Reporting and COVID-19

Noam H. Arzt, Ph.DThe United States is continuing its slow emergence from a nation-wide shut down imposed to slow down the spread of COVID-19. Most states have started to reopen, with bars, restaurants, and many workplaces starting to fill. As people begin to spend more time together again, it is critically important that public health agencies do everything they can to help prevent further spread of the infection and continue to monitor the level of infection within the population.

Data is an important tool that public health has to understand what is going on in the country. Years of limited government investment and neglect of current systems has limited public health's ability to meet the challenges of managing both localized outbreaks and pandemics. A recent article in the New York Times about the Centers for Disease Control and Prevention (CDC) response to the pandemic points to antiquated data systems as one underlying problem. This is exacerbated by the distributed nature of public health in the US and the ability of each state, territory, and sometimes even county to establish its own processes and deploy its own systems to support its activities.

Data collection for COVID-19-related infections and testing has been confusing at best. Currently, CDC makes available an Adobe Acrobat-based reporting form (Figure 1), but clearly instructs clinical sites to contact State and local public health authorities to determine how to report. In reality, reporting varies widely from state to state, with many jurisdictions relying on paper forms, spreadsheets with patient data (or summarized data), and comma-separated values (CSV) files from testing laboratories and clinical care sites. CDC released new guidelines for laboratory reporting in early June.

Public health surveillance

A key element of public health surveillance is the reporting of infectious and certain non-infectious conditions to state, local, territorial, and tribal public health agencies (PHA) around the United States. The purpose of this reporting is to allow these public health agencies to monitor, control, and prevent the occurrence and spread of diseases and conditions within the population.

Today, most case reporting is done manually. That is, when a clinician suspects that a reportable condition exists or may exist, a paper form provided by the local public health agency is completed and typically faxed to the public health agency for review. Often, forms are incomplete (or even illegible) and may require follow-up by the public health agency to complete or understand. Data from these forms may be manually entered into one or more information systems at the public health agency for case determination and follow-up or statistical tracking and analysis. In addition, reporting requirements vary from jurisdiction to jurisdiction with no proscribed national reporting directly to CDC.

A national project has been underway to implement electronic case reporting (eCR), which is defined by CDC as "the automated identification and transmission of reportable health events from the electronic health record (EHR) to state and local public health departments." The CDC, through the Council of State and Territorial Epidemiologists (CSTE), has funded the Reportable Conditions Knowledge Management System (RCKMS) which deployed an Authoring Tool tailored to eCR that allows jurisdictions to author rules that determine whether a patient's encounter is reportable for certain conditions.

In addition, the RCKMS software includes a Decision Support Service (DSS) to which the authored rules are deployed. The DSS runs the deployed rules against data sent from a patient record by a clinical care site and determines if the patient's encounter is reportable to public health, for which condition(s), and to which jurisdiction(s). The RCKMS software was developed by HLN Consulting under the LGPL v3 open source license. You can read more about this project in a previously-published article in Open Health News.

Why open source?

The open source approach has been a great model for collaboration and product development for HLN Consulting and its customers. Much of our work originates with public health agencies, so open source development allows us to take that government-funded work and make it available to a wider audience. Many of these agencies have similar core software requirements with some specific local differences. Our open source model means that all agencies can share the core code, and then customize it as needed. The open source model also reduces the users' dependence on a single vendor in a turbulent world through access to source code if necessary. Open source can also promote more modularity in complex systems.

Our open source projects are conducted in a collaborative and transparent way. For RCKMS, HLN Consulting is working with CSTE, a national membership organization of public health epidemiologists, to develop the requirements and test the system before deployment. While currently RCKMS is only deployed centrally, it is likely that over time local health information exchange (HIE) organizations or health systems will deploy local copies of the decision support service while relying on the centrally-developed rules distributed by CSTE.

Impact of COVID-19 on case reporting

Deployment of eCR nationally began very slowly in the fall of 2019. As COVID-19 accelerated, there were not very many clinical sites that were ready to implement eCR. Though the necessary lab and other clinical data is often available in electronic health records (EHR), most EHRs are not prepared to implement the Health Level Seven (HL7) standards required to make eCR work. Still, as of the end of May 2020, CDC reported that over 400,000 electronic case reports submitted from over 2,000 facilities in 17 healthcare organizations.

RCKMS rules authoring screenTo accelerate eCR use specifically for COVID-19 reporting, the Association of Public Health Laboratories (APHL) launched "eCR Now" to further enable EHRs to participate in eCR. The project is developing an open source back-end SMART app to be deployed in EHRs that are capable of supporting a newer HL7 standard, Fast Healthcare Interoperability Resources (FHIR), even though they are unable to produce the current electronic exchange document for eCR (the Electronic Initial Case Report, or eICR). The app converts the FHIR-based query response from the EHR to a Clinical Document Architecture (CDA) eICR for submission to the RCKMS decision support service. EHRs should be able to begin using this new app soon.

We still have a long way to go. A June 2020 Department of Health and Human Services (HHS) Report to Congress on the COVID-19 Strategic Testing Plan indicates that, "...it is critical that collection of complete data and automated electronic exchange of those data be included in the planning. This includes defining the critical data elements that must be collected during patient registration and ordering at every testing site, how those data will be collected, and how they will be electronically transmitted to the testing lab; defining required data elements, file formats, and automated transmission methods for having test results reported to the health department; and how all de-identified line level testing data will be reported from the health department to HHS" (p. 41).

The need for sustainted investment in public health software

As the use of eCR and eICR become more prevalent we should be able to have more timely and more accurate disease reporting. Sustained investment in public health software is crucial to enabling modern information technology to help fight the pandemic and position us to fight the next. For this funding to be successful it must be ongoing. Typically, public health funding is episodic - it rises when a disease outbreak takes place, and then it withers as the public forgets about the disease outbreak(s).

The other fundamental issue is that Congress should consider funding open source, collaborative projects when such projects exist. First of all, if taxpayer money is to be invested in pandemic prevention and response software, it should benefit all Americans. This was the crucial point made by Congressman Pete Stark in his famous H.R. 6898, the health IT bill that should have been passed in 2008. This point is well made by Dr. Bruce Wilder in this article.

Second, funding for open source software would mean that states and localities can collaborate on the development of all the common core elements of the software and platforms, yet at the same time have the ability to customize the software to fit their existing state and local public health requirements.