AI Prior Authorization for Payers Market Map
This is part of Elion’s weekly market map series where we break down critical vendor categories and the key players in them. For more, become a member and sign up for our email here.
Last week, we focused on how providers are using AI to improve prior authorization (PA). On the payer side, PA is hardly any more beloved: important for utilization management (UM) and financial planning, but a nightmare to manage.
After PA requests are sent in, payers have to process the request from various channels into a case, match it with clinical criteria, validate required information, review and make the PA determination, then respond and iterate through any appeals.
As payers update their procedures with new clinical evidence guidelines and new diagnostics and treatments, managing policies and creating decision trees for PA decisions requires enormous effort.
This week, we’re diving into how AI can enhance the payer side of the PA equation with AI Prior Authorization for Payers.
AI-Enabled Workflows
Payer-facing AI prior authorization platforms have a huge opportunity to streamline these workflows by:
Improving rules-engine generation and updates from unstructured policy documents and PDFs
Automatically processing PA requests across all channels
Matching the case to the right clinical criteria, validating data completeness, and determining if policies have been met through machine learning models and generative AI
Generating decision documents and automating appeals correspondence with providers
Many Strategies with One Goal
Several vendors are already making strides in AI-enabled PA for payers, but they all seem to be carving out niches and areas of expertise as opposed to handling the entire process from start to finish.
Banjo Health offers tooling both for the PA request and clinical decision policy creation workflows for health plans, TPAs, and pharmacy benefit managers.
Basys.ai is a newer player, but uses LLMs to offer rapid ingestion of policy documents for fast integration and up-to-date policies.
Cohere Health is one of the more mature vendors in the space, and offers additional products for UM.
Co:Helm is a generative AI platform for payers, with initial use cases around enabling UM nurses to make complex PA decisions more quickly.
Finally, GenHealth.ai has built their own large medical foundation model and is using it to enable both rapid policy ingestion and PA review.
Beyond these niches, we see PA as the first place where payers will incorporate advanced clinical AI. Although PA should never be fully automated for payers, vendors that win here stand to become a part of payers’ source of truth in clinical decision-making. And with that, the race is on.