One of the most daunting quandaries in healthcare is managing the abundance of data that are generated every day. The challenge isn’t just the vast volume of information, but the fact that most data are unstructured. Unstructured data require a considerable amount of manual intervention to read the contents and determine where the information needs to be integrated for the most efficient action.
These data challenges multiply as organizations struggle to compile information across various platforms and sources, from interoperable electronic health records (EHRs) that share structured data to faxes, scanned documents, and handwritten forms that are not structured. Managing structured data interoperability is challenging for small organizations, such as skilled nursing facilities, behavioral health clinics, birthing centers, and substance use disorder clinics that don’t have certified EHRs. These facilities weren’t incentivized to purchase an EHR under meaningful use, and they treat some of our most vulnerable patient populations. In order to have health data equity, we also need to consider tech equity. It’s an area where advanced interoperability and artificial intelligence (AI) solutions could make a substantial and pragmatic difference.
Why Tech Equity Matters in Healthcare
Despite recent advances in digital healthcare tools, the technological capabilities for processing unstructured data differ wildly between healthcare settings—leaving under-resourced organizations without timely access to longitudinal patient data.
These tech disparities contribute to health inequities, such as when a patient being discharged from a hospital to a post-acute care facility is unable to receive medication in a timely fashion because the records didn’t arrive with the patient, or are on page 4 of a fax document. They also underscore the increasing importance of adopting a multi-channel approach to interoperability, with the goal of seamlessly exchanging health information regardless of an organization’s digital maturity.
Interoperability is not a sending data problem, it’s a receiving data problem. The sender needs to use a multi-channel approach that takes into account how the receiving end can accept data in the most efficient and actionable manner. Any proposed data-sharing options emphasize the use of a standard like Fast Healthcare Interoperability Resources (FHIR) using the Trusted Exchange Framework and Common Agreement (TEFCA) to streamline how data is structured and shared. Unfortunately, these standards could exacerbate tech disparities and health inequities by pigeonholing smaller facilities that lack the resources needed to adopt sophisticated EHRs and national standards into a specific path for information exchange.
Organizations that can’t afford to implement an EHR can still exchange data across complex systems in a simple send and receive platform, leveraging technology they likely already own: digital fax. While fax has been referred to as a legacy technology by some, such a solution gives “digital have-nots” a pathway to share important content from one care setting to the other. This enables them to provide more coordinated care even if the data is unstructured. The impact: better outcomes at reduced cost.
The most effective platforms enable providers to automate and streamline information exchange to track the patient care journey as close to real time as possible. The right platform also will empower providers to manage messages from multiple sources and file types through a workflow they already use, such as a centralized intake portal or dashboard. When unstructured data – from faxes, scanned documents and PDFs – can be extracted to integrate into tools the providers are using today, the data are received quickly and become more actionable. This includes Direct Secure Messages to records requests and scanned paper documents, with the ability to retrieve documents on demand with the most critical data extracted.
This is where AI could come in to close gaps in interoperability. For example, most small and under-resourced facilities rely on pragmatic technologies like digital or even paper fax to exchange information. When AI in the form of machine learning and natural language processing (NLP) is applied to a digital fax, it can transform unstructured data from open notes, scanned images, and even handwritten text into structured data. That data can then be sent as a Direct Secure Message, FHIR, X12 or HL7 message. For those systems that can’t send native FHIR, this solution means the receiving end can consume data that can automatically be mapped into a workflow for immediate action without the need for manual intervention. It can level the playing field for information sharing in a pragmatic way, empowering real-time data exchange to support better care.
Addressing Inefficiencies and Inequities
For skilled nursing facilities, critical access hospitals, behavioral health clinics, substance use disorder clinics, and birthing centers, advanced technologies are financially out of reach. There is a need to depend heavily on pragmatic solutions like digital cloud faxing to send and receive information because it offers the privacy and security to meet strict HIPAA regulations and is an easy send-and-receive data workflow.
Traditionally, fax technologies still require labor-intensive workflows, requiring administrative staff to tediously re-enter unstructured information into structured systems that make health data more accessible. The time it takes to manually enter all this data can impede critical decisions by obstructing the flow of information and care while adding room for human error and other inefficiencies. For facilities in underserved markets treating the most vulnerable populations, these hurdles not only hamper data exchange but also undermine health equity.
While there’s value in developing national data frameworks to overcome these hurdles, not all care settings have the technology to support these standards. What we’ve heard continually from our conversations with healthcare leaders in these settings is the underlying need for tech equity as a precursor for interoperability. Without access to technologies that can send and receive FHIR-based messages, these settings simply can’t participate in modern data exchange, such as TEFCA.
Applying AI to Enable Health Equity
Converting unstructured information into structured data doesn’t necessitate an extensive or expensive digital transformation; it just requires the right extraction and translation technologies.
When digital cloud fax is combined with intelligent data extraction technologies that leverage AI to convert unstructured information into structured data fields, this data can then be configured to the most efficient consumption on the receiving side. That example does not require hefty investments, helping to level the tech equity playing field.
About Bevey Miner
Bevey Miner serves as Executive Vice President, Healthcare Strategy & Policy for Consensus Cloud Solutions. With over 20 years’ experience in healthcare technology and digital health, she has been instrumental in leading strategy, product management, business development, marketing and commercialization. Bevey has been influential, leading innovation in care coordination, patient engagement, population health and interoperability as well as advocating for policy change with the federal and state governments.
