With the rapid development of artificial intelligence (AI), state legislatures have been attempting to keep up with the technological advancing surge. Most recently, Texas enacted Senate Bill (SB) 815, a law imposing greater restrictions on the use of AI in utilization review (UR) and coverage determinations. This is not the first state to address the use of AI in UR and the insurance landscape, as California Gov. Gavin Newsom approved a similar bill on Sept. 28, 2024. In fact, it had been expected that the California law would set national precedence on AI-based regulations on insurers for the expedition and efficacy of coverage determinations, with additional informationpreviously reportedby Holland & Knight. The impending limitations on AI in the Texas insurance landscape will impact how healthcare providers and insurers approach the UR process statewide as the law takes effect Jan. 1, 2026.
AI Software Implications and Use in the UR Process
Insurers have begun to look to AI-based algorithms to expedite the medical necessity determination, which is the part of the UR process through which health plans consider and approve or deny coverage for particular patient services. However, Texas, like California and other states, has recognized the significance of continued human review.
The newly amended Texas law provides that a UR agent may not use an automated decision system to make an adverse determination either in whole or in part. An adverse determination includes a determination that healthcare services provided or proposed to be provided to a patient are not medically necessary or appropriate or are experimental or investigational.
Nonetheless, the law does not prohibit the use of an algorithm, AI system or automated decision system for administrative support or fraud detection functions. Instead, the amendment defines an AI system as any machine learning-based system that infers from the input received how to generate content, decisions, predictions and recommendations that can influence physical or virtual environments. An automated decision system means an algorithm including one using AI that uses data-based analytics to make, suggest or recommend certain determinations, decisions, judgments or conclusions.
Importantly, the notice provided to the insured regarding an adverse determination must include a description of the source of the screening criteria and review procedures used to make the adverse determination. In addition, the use of AI in UR may be audited or inspected at any time to ensure compliance with the stringent regulatory standards set forth under SB 815.
Looking Ahead
Similar to the predicted ramifications of the California law, only time will tell if integrating AI into the UR process will prove beneficial. Though AI is typically sought to streamline and expedite decisions, considering the limitations imposed by the regulations, it is possible that the impact of AI will be lessened as adverse determinations will require review procedures beyond AI-based algorithms and technologies.
It is important that AI be used in UR decisions only for administrative support or fraud and abuse detection with human oversight and review and that its use be disclosed in the notice to the insured. For insurers planning to utilize AI for UR in Texas, there will likely be strong oversight and potential review of adverse determinations made with AI to ensure that the final determination is not the flawed result of a technological error. It is likely that other states will follow Texas and California in enacting AI-focused regulations in the near future.
Stay tuned for updates as the potential capabilities and flaws of generative AI for insurance coverage review continue to develop upon this evolving regulatory landscape. For more information or questions, please contact the authors.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
