The Future of Revenue Cycle Management (RCM): AI and Data Analytics

By Candice Hoshi, VP of Revenue Cycle, UCHealth In today’s rapidly evolving health care landscape, sustainability depends on quickly addressing key areas within revenue cycle management. The pressure to deliver value and reduce costs is constant. At UCHealth, we’re leaning on data analytics and artificial intelligence (AI) to help uncover problems, streamline operations and improve patient experience and financial performance. At UCHealth, our mission is to improve lives. Staying congruent with our mission was key to adopting the tools rapidly and proficiently. Leading with data to improve processes and performance The revenue cycle often starts with patient registration and ends with patient billing, but there are many steps along the way. Each step presents potential for delays or errors, which can have a variety of negative impacts. In 2023, we developed an analytics tool to investigate the mismatch between the authorized and billed CPT codes. The tool revealed that CPT mismatches were the cause of many prior authorization denials – in other words, what we told the insurers we were performing was not what was billed. On the surface, it may appear to be an administrative issue; however, in the example of colonoscopies, up to 50% of screening exams result in some level of diagnostic intervention like a polyp removal. We implemented a retro authorization process to notify insurers of CPT changes within 48 hours of service, and the result is quicker payment and less rework in appealing for services based on medical necessity.  We also launched “DPCs” – denial prevention committees – led by our regional operators. We have enhanced the CPT mismatch analytics tool by adding denial data, and our operators are engaged in helping resolve concerns related to prior authorization and registration-related denials. Collectively, they have improved staff/provider education, corrected IT setup issues, and made other valuable improvements. Most importantly, our operators feel connected and empowered to address issues. The results? Over a two-year period, we’ve reduced prior authorization denials by 47% and registration denials by 41%. Getting paid on first pass is a double win – for patients whose bills quickly adjudicate for billing transparency, and for providers who receive timely compensation for the care they provide. AI use cases in document scanning, CDI and process intelligence A few years ago, our scanning specialists were working around the clock to ensure documents were available in the electronic chart within 24 hours. This manual process was not sustainable, so we turned to an AI partner. The tool “reads” images using optical character recognition (OCR) technology, and our HIM teams trained it to index the document in the patient’s chart with a high degree of accuracy. Approximately six months after going live, the tool improved scanning productivity by 1100% and accuracy over 95%.  Not only did we improve our operational metrics, but we also improved clinician and patient satisfaction. Clinical documentation integrity (CDI) is a hot spot in RCM right now. Hospitals are discovering that not only are they potentially leaving money on the table, but their quality metrics also may be understated because of incomplete or inadequate clinical documentation. CDI work is resource-intensive and relies on expert clinical acumen. We knew there were gaps with manual processes and short stays where patients aren’t under our care long enough for queries to finalize. We chose an AI company that looks for both coding and CDI querying opportunities, and flags for a final review prior to billing. In the 8-month pilot across three hospitals, the tool identified almost 10,000 more query opportunities than our baseline. Approximately 20% of total queries were attributed to the clinical validation of sepsis, acute blood loss anemia and respiratory failure. One hospital improved its sepsis mortality index by 43% and all pilot sites similarly experienced a rise in quality and financial impacts. We will soon expand these tools to additional hospitals. The revenue cycle is riddled with many steps, and some occur between the “ping-ponging” of info exchanged between providers and payers. Operational Intelligence is a newer offering from an AI partner that manages daily automations such as prior authorization and enriched claims status. Their Operational Intelligence solution finds the “diamonds in the rough” – those one-off workflow inefficiencies that are difficult to find. For example, it found unnecessary account transfers between billers and coders on hundreds of accounts, which routinely delayed $1M-$2M in monthly cash flow. Another hidden insight resulted in a 96% reduction in coding review transfers, saving valuable time. These insights have led to new automations, staff training and process improvements. Conquering fear and finding the “why” Embracing new and creative technologies can be difficult and scary. Data analytics are powerful, but they require humans to validate information. Fears of job cuts or erosions to current state performance are why some technological adoptions fail. Using data to objectively evaluate outcomes allowed our teams to adopt and champion the benefits. In the case of CPT mismatch, we were solving prior authorization denial root causes that led to significant rework, patient complaints and A/R write-offs. For HIM scanning, we wanted to deliver a faster, quality service that also improved customer satisfaction and resource costs. We began with the end in mind, and in virtually every case, the patient was at the center of the “why”. At UCHealth, our mission is to improve lives. Staying congruent with our mission was key to adopting the tools rapidly and proficiently. Our people are more empowered than ever with the tools and insights they need to succeed in a challenging environment. Looking Ahead As Generative AI technologies mature, the potential for further transformation is huge. Clinical documentation automation and agentic AI are examples of trending technologies. As our industry’s adoption grows, RCM will change with it. Realizing this potential requires thoughtful investment in infrastructure, talent and change management. Success depends on leadership commitment, cross-functional collaboration and a willingness to embrace continuous improvement. The post The Future of Revenue Cycle Management (RCM): AI and Data Analytics appeared first on HealthTech Magazines.

Source: https://www.healthtechmagazines.com/the-future-of-revenue-cycle-management-rcm-ai-and-data-analytics/