Clinical Pathways at the Crossroads of AI and Value-Based Care

J Clin Pathways. 2026;12(3):e004.

Employer plans, whether self-funded under the Employee Retirement Income Security Act of 1974 (ERISA) or fully funded through a commercial plan, will likely have a proactive mindset that will lead to greater coverage modeling scenarios, plan accountability, and a focus on outcomes to successfully support their workforce while controlling costs beyond 2027. There are numerous efficiency tactics to address this, but more than ever the spotlight has been focused on artificial intelligence (AI) as a tool to control costs. Today, the health care industry has become a major target for AI integration and development as the need for operational efficiency has grown in importance due to pricing and cost pressures. As I mentioned in the previous entry of this column, employers are facing ever increasing premium costs for their health plans, which are negatively impacting their balance sheets.¹ Many business leaders see today’s market challenges as structural and not temporary, which has led them to prioritize optimizing operations and improving worker productivity. Overall, market stakeholders are reviewing their care expenditures and seeking more granular insights into the care delivered.

The promise of AI to deliver unique and timely insights for a variety of applications beneficial to businesses of all types could not only accelerate health care change but do it in ways not foreseen by traditional thinking. This column will explore the traditional role of clinical pathways in health care and how changes in the industry are forcing leaders to look at the potential of AI and how its many applications and iterations—particularly integration in clinical pathways—have swept over the marketplace.

Clinical pathways are evidence-based clinical care road maps that assist institutions in reducing variation in care, thereby promoting best practice and providing better outcomes for patients. From a financial risk perspective, a pathway strives for standardization to improve patient outcomes and create institutional or plan coverage efficiency. However, clinical opponents fear that this standardization is against the goal of personalized or precision medicine and creates “cookbook” clinical medicine, threatening the autonomy of physicians.² 

Much of clinical pathway strategy has its roots in clinical quality improvement (QI), which can improve both clinical and financial outcomes. This includes allowing for the abandonment or discontinuation of a pathway if patients no longer meet clinical requirements. Institutions (hospitals or health plans) can build flexibility into their clinical pathways to ensure that enough patients within a specific population will fit within the pathway framework.

The variety of uses for pathways, and subsequent users with differing purposes, has created frequent tensions among clinical care and plan administration users. For instance, clinical pathways can aid in documenting key outcomes by provider entities for plan sponsors or their third-party managers. However, there is resistance to third-party clinical pathways by clinical practitioners while they are embraced by financial managers.

A previous survey published in the Journal of Clinical Pathways identified various ways in which pathways are utilized and for what purposes,³ and prior authorization (PA) was found to be a common use. PA requirements are on everyone’s radar screen to approve medical or pharmacy claim approvals, which plan administrators have sought to limit inappropriate or more costly products or services. As new products or services have continued to emerge for PA through clinical care with claim costs at higher costs to a plan, this tension has caused backlash from patients who pay increasingly higher out-of-pocket costs.

Understanding the potential of a clinical pathway also requires recognizing the limitations, resistance, and operational challenges associated with their development and use. As the health care industry increasingly embraces AI, emerging technologies may help transform clinical pathways from static frameworks into more dynamic tools that support both personalized care and efficient resource utilization.

Year-over-year plan cost trends have caused premium and cost-sharing increases and limitations on care networks and/or drug formularies, putting more pressure on care entities and plan administrators. As a result, business leaders have used different tools to help manage their QI goals. One example of these tools is the Pareto Principle or 80/20 rule. Taught routinely in business and health profession schools, a Pareto diagram helps a QI or financial team concentrate its efforts on the factors that have the greatest impact. It also helps a team communicate the rationale for focusing on certain areas along with other QI tools.⁴ 

The use of AI is another option for business leaders to keep up with industry trends and changes while maintaining quality standards. For instance, with the use of AI, it is possible to analyze individual patient data with diagnostic and clinical data against aggregated experience in a cloud environment to predict appropriate use. There are also AI solutions that allow for routine, real-world studies with n-of-1 data analysis that demonstrate the benefit of personalized medicine on an individual patient. A similar determination of inappropriate use can find nonresponsive or at-risk patients that can avoid adverse effects or financial waste.⁵ By design, these AI analyses resemble or iterate as a pathway of care that some might call a patient care journey. While there is more than one application tying AI solutions to clinical pathways as a general term, the purpose may be different from traditional clinical practice applications.

Such solutions being developed push aside traditional research norms, including large clinical studies used for marketing approvals that have now been explored by the US Food and Drug Administration (FDA) for rare conditions.^6,7 For example, in small groups or populations that are not amenable to larger patient counts, novel analytic models allow for elegant extrapolations from multiple sources of data than can be applied down to a population of one. Safety determinations by the FDA can be accomplished along with the expected effectiveness in delivering an optimal clinical outcome. In addition, academic centers and other health systems are already exploring the same AI-driven analytics in their own patient populations to optimize clinical care resources, individual care outcomes, and financial efficiencies.

As AI-driven analytics become more sophisticated, clinical pathways may evolve from static population-based tools into dynamic, patient-specific frameworks that improve both clinical outcomes and financial stewardship.

Empowering patients has become more common given the current comparative state and business importance of US health care for other industries. As such, health insurance trends for both medical and pharmacy coverage may see movement from traditional one-size-fits-all coverage decisions toward less restrictive and more tailored, with higher shared cost (eg, deductibles, copayment, or coinsurance) in 2027-2030. It seems likely that employers will seek more uses for clinical pathways, including real-time applications incorporating clinical pathways into strategic plan coverage options such as improving access at the point of service while documenting care outcomes.

Government agencies and commercial entities are actively pursuing options for coupling AI research with new research models and leveraging predictive analytics with real-world evidence studies. To date, those application efforts include both clinical and financial domains. Looking ahead, clinical pathway uses will continue straddling both domains. What will change is the speed of analytics, design and development for more specific applications, and the facilitation of appropriate patient access closer to real-time.