Despite the Centers for Medicare and Medicaid Services’ call for an end to faxing in healthcare by 2020, the practice remains prolific across care settings. Healthcare organizations exchange more than 9 billion fax pages each year at an additional cost to the healthcare system of $125 billion.
Faxing is problematic, as 30 percent of tests must be re-ordered due to lost faxes and 25 percent of faxes fail to arrive prior to a patient’s first visit. Manual indexing is often required to integrate faxed documents into electronic health records (EHRs) and other health information systems—a time-consuming process requiring human and financial resources that few healthcare organizations can spare.
The good news is that fax indexing is well-suited for robotic process automation (RPA) — which excels at automating mundane and repetitive tasks—and generative AI. While it is unrealistic to expect digital workers to accurately index 100 percent of faxes, use of RPA can significantly decrease the burden on manual indexers. In fact, the best outcomes are realized when digital workers are paired with human counterparts who manage exceptions and handle specialized information and requests until digital workers have undergone sufficient “on-the-job training” to handle these higher-level tasks.
When properly designed and implemented, a hybrid automated fax indexing model accelerates processing and eliminates the care delays caused by late and lost faxes. It relieves the strain manual indexing places on increasingly scarce resources, reducing costs and freeing internal staff to focus on higher-value tasks.
A Well-Equipped Digital Workforce
A well-designed RPA model is the foundation of a successful automated fax indexing program. The first step is to map the necessary technologies, a process that is ideally informed by shadowing human indexers for a comprehensive understanding of the process and any unique needs, followed by implementation planning and feasibility testing.
Automated indexing involves equipping digital workers with an array of advanced automation tools, such as optical character recognition (OCR) to download and digitize faxed documents, GenAI and natural language processing (NLP) to filter and classify faxes, and pre-defined rules specific to the sender’s documentation format to extract data, determine its confidence threshold, and either index it into a documentation management system or EHR or divert it to the validation workflow for manual processing.
Machine learning allows digital workers to adapt to new document formats and categorize data according to providers' templates and styles. Further, each processed fax enhances accuracy, efficiency, and capabilities while reducing exceptions.
Feasibility testing and monitoring effectiveness are crucial. For the latter, establish clear KPIs, for example the volume of faxes indexed per day, indexing accuracy, turnaround times, and productivity levels.
Making its Mark
Automated fax indexing may be a relatively new entry into the burgeoning field of healthcare AI, but it has quickly made its mark. For one health system, automating their fax indexing has reduced the number of manual indexers required to process its fax volume, enabling key team members to instead focus on higher-value tasks.
The system currently indexes approximately 33 percent of all inbound faxes, encompassing 14 unique document types from 90 high-volume senders. It has achieved a 98% accuracy rate and 24-hour turnaround time, reducing annual expenses by roughly $2 million and increasing daily fax volume processed fourfold.
As the digital workers “learn” over time, the automation rate will increase. By the end of Q1 2025, the organization expects to have 80 percent of its physician practices’ fax indexing workflow to be fully automated, reducing the number of manual indexers required by more than 50 percent.
A fax-free healthcare system isn’t in the cards, at least not for the foreseeable future. However, by leveraging readily available automation and AI tools it is possible to digitize the process and alleviate its associated cost, productivity, and patient safety issues.
Contact us to learn more about how to transform your fax indexing process with automation.
Author
