Often times, processes created with the objective of benefitting the system, end up creating barriers and administrative burden for multiple healthcare stakeholders. Prior authorization is one such process.
Prior authorization (PA) plays a vital role in obtaining timely approvals for the provision of specific healthcare services to a patient covered by a health plan. The process is intended to manage the utilization of healthcare resources, reduce overuse or misuse of services, improve the quality of care, and control the overall spending on healthcare.
However, according to a 2020 physician survey, from the American Medical Association (AMA), PA issues are associated with 94% of care delays and contribute to a negative impact on patient clinical outcomes and create administrative inefficiencies. The same survey also demonstrates that an overwhelming 79% of prescribers believed that PA requirements may lead to abandonment of treatment.
While ePAs (electronic Prior Authorisation platforms) have certainly improved the speed and efficiency of the process, they do not necessarily impact approval rates. That requires analyzing and modeling, both, payer and provider behaviour as well as understanding the details of every submission. Clinically expert PA professionals have to read through anywhere between 50-200 pages of PA case forms consisting of patient records, charts and doctor notes to analyze the case.
That’s where AI/ML can be put to good use to ease the burden of clinicians and their staff. By training algorithms on thousands of records, NLP can intelligently extract and classify data from unstructured medical records and PA case forms into easily analyzable data. Inbuilt domain-specific variable classifiers enable the analysis and mapping of cause and effect for instances of prior rejection. Combine it with medical experts – and you can take the data and ensure effective interpretation and actionable insights with recommendations to ensure higher PA approvals and faster time to market.
Of course, we tried it ourselves at Indegene in the real world on real PA case forms from a client... it didn't let us down. We were able to drive the following tangible outcomes for our client:
32% improvement in the number of PAs approveds
10% reduction in incorrect PA rejections
9% reduction in PA rejections due to discrepancies or insufficient documentation
Many #FutureReadyHealthcare organisations are considering this approach and we're excited with the outcomes we are able to deliver to them. Are you experiencing similar PA challenges? How are you solving them?