Erasing Bias in Clinical Decision Support to Achieve True Health Equity

David Beckam

Many have claimed that technology will be the great equalizer in healthcare. While technology is objective and purely data-driven and without prejudice, there can still be inherent biases that hamper the ultimate goal of health equity.

Many have claimed that technology – from virtual care solutions to clinical decision support (CDS) systems – will be the great equalizer in healthcare.

These tools are viewed as an ideal solution for bridging the chasm that has prevented medically underserved communities – racial and ethnic minorities, those who live in rural and socioeconomically disadvantaged areas – from receiving the same quality of care as others. Yet, while technology is purportedly objective and purely data-driven and without prejudice, there can still be inherent biases that hamper the ultimate goal of health equity.

With healthcare data accounting for 30% of the world’s data volume, it’s no surprise providers can feel overwhelmed at the point of decision making, especially given time constraints. When challenged with assessing too much data in a limited period of time, our brains look for patterns and fall back on gut instinct, which can bring with it unintended biases.

CDS systems have been designed to help providers assess the vast amounts of data that can play a role in care decisions. But CDS should not be all about guideline-driven care; it should also facilitate health equity. As new technology is developed and current capabilities are advanced, it’s critical we seek to combat bias through the guidelines themselves and the underlying clinical algorithms that drive CDS.

Overcoming Bias in Healthcare

Health equity implies, “a system designed and built for inclusion and equal access to the best health outcomes for all,” including socioeconomic status, such as education, income and occupation, according to the Agency for Healthcare Research and Quality (AHRQ).

However, the agency believes, in reality, the definition is more expansive, referring to “inclusion and best health outcomes for all no matter their physical environment, personal behaviors and abilities, or social circumstances (gender identity, military services, sexual orientation, citizenship status, social status, history of incarceration, culture and tradition, social connectedness, work conditions, early childhood education and development).”

Overcoming all of these factors can be difficult for the average provider. One study utilizing the Implicit Association Test (IAT) revealed that 80% of sample clinician groups exhibited evidence of implicit bias. These biases are most pronounced among racial and ethnic groups.

For example, a groundbreaking Institute of Medicine report found that Black Americans, who are only 3% to 6% more likely to deny treatment than other populations, received significantly less needed care than white Americans, even when controlling for factors such as insurance coverage and patient income.

The result: Black Americans are 40% more likely to die from breast cancer, 20% less likely to receive treatment for depression, and two times more likely to receive a less desirable treatment for diabetes, such as limb amputation.

While much of this problem can stem from providers’ attitudes, implicit biases and discrimination, it’s likely that clinical algorithms, tools and guidelines are factoring into the biases as well. The Kaiser Family Foundation noted that algorithms for CDS could introduce bias if they’re built on data that is not reflective of a diverse population.

Refining Clinical Decision Support for Greater Health Equity

CDS has been defined as a “process for enhancing health-related decisions” that provides “clinicians, staff, patients or other individuals with knowledge and person-specific information, intelligently filtered or presented at appropriate times, to enhance health and healthcare.”

However, to truly support both guideline-driven and equity-driven care, the developers of the algorithms and the CDS systems themselves must strive to eliminate bias.

A study completed in 2019 explained how a widely used commercial healthcare algorithm reflected serious racial bias. Specifically, the risk score used healthcare expenditures as a proxy for health status. Researchers found that Black patients were consistently and significantly sicker than white patients with the same score. Since the algorithm was designed to predict healthcare costs, rather than illness, it was flawed because unequal access to care means that less money is typically spent caring for Black patients than white.

Additionally, the National Kidney Foundation (NKF) recently announced changes to the eGFR calculator, the most widely used kidney test, after many years of pushback from physicians and medical students about the flaws of its race data, which removes the race variable from the equation that estimates kidney function.

In spring of 2021, the AHRQ, at the request of Congress, sought public comments on clinical algorithms used in medical practice that may introduce bias into decision making or negatively affect access, quality or health outcomes among racial and ethnic minority groups. From the responses, AHRQ plans to commission an evidence review to critically appraise whether race/ethnicity and prior utilization are included as explicit variables in commonly used algorithms, and how algorithms have been developed and validated. They also plan to identify and review approaches to clinical algorithm development that avoid the introduction of racial and ethnic bias into clinical decision making and resulting outcomes.

Done right – through careful design, greater awareness and a thoughtful approach to eliminating potential biases – algorithms and CDS systems can mitigate such problems. And in the process, they can decrease disparities in care to achieve greater health equity.

Additionally, when algorithms are developed based on data that accurately represents the diversity of the patient population to which it is being applied, they can also achieve greater patient outcomes. Therefore, demographic, clinical, socioeconomic and even patient preference data should be included when leveraging CDS systems. This data may not be available in the EHR directly, and can require retrieving SDOH and PRO data from other sources.

Looking ahead, it is vital that designers of CDS solutions think about and plan for health equity from the start of development. By taking this approach, CDS can truly deliver on the promise of fostering both guideline-driven and equity-driven care that improves outcomes and the overall health of everyone.

Lucienne Marie Ide, M.D., PH.D., is the Founder and Chief Executive Officer of Rimidi, a leading clinical management platform designed to optimize clinical workflows, enhance patient experiences and achieve quality objectives.

Next Post

Opinion | Tough Decisions About End-of-Life Care

To the Editor: Re “A Better Way to Face Death,” by Dr. Daniela J. Lamas (Opinion guest essay, Jan. 6): Too often, advance directives are construed as immutable guides to how one dies. We have found advance directives to be more helpful if we focus on how one wants to […]
Opinion | Tough Decisions About End-of-Life Care

You May Like

Subscribe US Now