AI-powered prior authorization: A catalyst for building a more equitable healthcare system

Updated on May 17, 2024

I look forward to a time when a person’s zip code and income level do not dictate their life expectancy. That’s the core principle of health equity: ensuring everyone has a fair and just opportunity to live a healthy life. Unfortunately, in many communities, some patients have significantly worse outcomes. The Centers for Medicare & Medicaid Services (CMS) has prioritized advancing health equity to eliminate disparities in healthcare access and outcomes.

I have focused on health equity throughout my career. As a physician at Federally Qualified Health Centers (FQHCs), I’ve seen both sides of the coin: communities in which health equity flourishes, and others in which it’s sorely lacking. This experience equipped me with a better understanding of the factors that contribute to, or hinder, achieving health equity. It has also strengthened my passion for ensuring all patients get the quality medical care they need and deserve. To understand why health equity is so vital, we should examine the challenges of pursuing equitable healthcare for all.

The power of social determinants of health data

A patient’s zip code is often an indicator of their overall well-being. Social determinants of health (SDOH), like access to transportation, safe housing, and healthy food, play a huge role in a person’s outcomes. Effectively capturing and utilizing this data is crucial for attaining health equity.

The first hurdle? SDOH data collection itself. Currently, a significant gap exists: less than 2% of Medicaid and commercially insured patients have documented zip codes, which are used to analyze SDOH data. This lack of information creates a blind spot, making it challenging to identify and address the specific needs of vulnerable populations.

Beyond a patient’s medical condition, SDOH significantly impacts their care journey. Inpatient prior authorization requests, for example, are often driven by a patient’s SDOH risk factors. Lack of caregivers, social support, or a safe home environment can influence whether a patient is admitted as an inpatient or observation-level care and determine their discharge plan. Studies have shown that SDOH can account for 70-80% of a patient’s overall care needs, highlighting the profound influence these social factors have on a person’s health.

But there’s hope. The dream of equitable healthcare starts by capturing SDOH data at a granular level. Once we have this information, we must ensure it is interoperable and standardized so patients, physicians, and health plans can easily access and share this critical data. In my prior practice, we screened patients for food insecurity and used their zip codes to connect them with local food banks; this was especially helpful for a patient with uncontrolled diabetes who had trouble accessing and affording nutritious food. The patient frequently relied on eating macaroni and cheese, which increased their a1c (a measure of diabetes control) to dangerously high levels. After we helped the patient access nutritious food resources, their diabetes significantly improved. These kinds of interventions can make a real difference in a patient’s health and recovery.

Of course, challenges remain. While capturing and sharing SDOH data is a vital first step, it is not enough. Many SDOH factors, like transportation assistance, are not reimbursable for physicians. Physicians often struggle to find enough time to care for their patients’ medical issues; adding transportation to a physician’s already full plate can take time away from physicians providing necessary care to their patients. This disconnect between data and resources must be addressed.

The good news: innovation and equity are now at the forefront of the healthcare industry. By harnessing the power of SDOH data and building a system that incentivizes addressing these social factors, we can

move closer to a future where healthcare is equitable. Health plans can play a critical role in reaching this goal by using AI technologies to analyze SDOH data like access to transportation and social services. By using AI, health plans can leverage this information to create personalized patient care pathways and reduce administrative burdens for physicians. 

Here are three key ways AI-powered prior authorization can promote health equity:

Personalized care pathways:

Traditional prior authorization processes rely on standardized criteria that do not necessarily consider a patient’s unique circumstances; this can disadvantage individuals who face more significant SDOH burdens. AI algorithms, however, can analyze vast amounts of data, including clinical history, socioeconomic factors, and claims data, to create personalized care paths for patients. These paths would outline evidence-based treatment options tailored to the patient’s needs and address potential SDOH barriers to care. For example, AI might recommend a diabetes management program with transportation assistance for a low-income patient struggling to access fresh and healthy food. This data-driven approach ensures that patients receive the most appropriate interventions regardless of their background.

Empowering physicians with clinical nudges

Physicians often lack real-time information on a patient’s eligibility for specific treatments or potentially more cost-effective options. This can lead to care disparities since physicians may hesitate to request authorization for treatments they perceive as unlikely to be approved. Here’s where AI-powered clinical “nudges” come in.

AI-powered nudges can be presented to physicians during the prior authorization process to suggest alternative treatment options–such as lower-cost services or in-network physicians–that are likely more accessible to the patient. Physicians also receive immediate feedback on the likelihood of approval based on the patient’s medical history and the plan’s guidelines. Additionally, AI technology can explain the rationale behind specific approval criteria, fostering greater communication and collaboration between physicians and health plans. These nudges empower physicians to navigate complex prior authorization requirements and advocate for the best treatment for their patients.

Streamlined care with episodic authorizations

Fragmented care delivery, wherein multiple authorizations are needed for different components of a single treatment plan, causes significant delays in approvals, leading to gaps in care and poorer health outcomes. AI can be used to group related services into a single authorization request, particularly benefiting patients in underserved areas with limited transportation options. AI can aggregate related services, like pre-surgical consultations, surgery, and post-operative care, into a single “episodic authorization” request. This eliminates the need for multiple submissions and streamlines the approval process. Episodic authorizations improve care coordination and reduce administrative hurdles, benefitting at-risk populations who may have difficulty navigating the healthcare system.

Responsible AI: A tool, not a solution

While AI-powered prior authorization holds great promise for promoting health equity, it’s crucial to acknowledge its limitations. For example, algorithmic bias in training data can worsen existing disparities, so health plans must use diverse datasets. AI should NEVER be used to deny care to patients. Finally, ensuring access to technology is crucial, as digital literacy and access vary. Health plans need alternative methods to reach patients who face technological barriers.

The path to achieving health equity is complex, but with innovative solutions like AI-powered prior authorization, we are starting to see a clearer way forward. Patients deserve a healthcare system where SDOH serve as opportunities to intervene and improve a patient’s overall well-being. By leveraging the insights gleaned from SDOH data and empowering physicians with intelligent tools, we can create personalized care plans that address a patient’s medical condition and the social factors that influence their health journey. This will ultimately improve outcomes for all patients.

Mary Krebs
Mary Krebs
Medical Director of Primary Care at Cohere Health

Dr. Mary Krebs serves as Medical Director of Primary Care at Cohere Health. In addition, she teaches residents and medical students at a family medicine residency program in Dayton, Ohio. She earned her medical degree from the Ohio State University College of Medicine in Columbus and completed a family medicine residency at Miami Valley Hospital in Dayton, Ohio. Previously, Dr. Krebs was in solo practice at a rural federally-qualified health center and co-ran Family Practice Associates, an independent rural practice.