How to Address the Relationship Between Health Equity and Patient Safety

By Christopher Cheney

Yale New Haven Health is striving to address the health equity elements of patient safety.

Health equity has become a top priority for healthcare providers nationwide. Last year, health equity was added as the Quintuple Aim for healthcare providers. In 2008, the Triple Aim for healthcare improvement was introduced, featuring improvement of population health, enhancement of the care experience, and reduction of costs. In 2014, the Quadruple Aim for healthcare improvement was created with the addition of workforce well-being as a fourth element to address healthcare worker burnout.

Linda Fan, MD, assistant professor of Obstetrics, Gynecology and Reproductive Sciences at Yale School of Medicine, and GME director of quality and safety at Yale School of Medicine and Yale New Haven Health, says health equity plays a significant role in patient safety.

“You basically cannot have patient safety without health equity. It is a glaring hole in how we have not been able to adequately address patient safety. In the past, we have believed that patient safety is an issue because the patient is not compliant—they did not listen to the directions or there is something intrinsically flawed with them. When you get to the root cause of a patient safety issue, it is possible that the patient did not understand directions, there could be language issues, or there could be issues with access,” she says.

Pairing health equity with patient safety is a healthcare trend, she says. “We are not used to looking at patient safety through the equity lens. Health equity is a different way of looking at patient safety. Once you start looking at patient safety through an equity lens, you cannot unsee it. We are at the beginning of incorporating health equity in patient safety.”

There are two primary approaches to addressing health equity issues in patient safety, Fan says. “When you are looking at quality and safety and health equity, one approach is to understand disparity indexes, where you might take something such as mortality related to myocardial infarction or how patients are treated for chest pain, and you compare how men and women are treated or how Blacks and Whites are treated. If there is a difference, then we should be trying to hone in on it and improve it. You need to be looking at information transparently and making sure that the data reflects disparities. Another approach is when you identify a safety event such as a medical error leading to a sentinel event, then you ask all the reasons why the error occurred, including health equity elements as part of the framework of looking at safety events.”

Health systems, hospitals, and physician practices should follow four steps to develop a systemwide process for equity data visualization of quality and safety measures, says Lou Hart, MD, medical director of health equity at Yale New Haven Health.

  1. With a multi-stakeholder, transdisciplinary team, launch a We Ask Because We Care education and training campaign directed to both patients and staff around the importance of sharing and collecting high fidelity patient self-reported demographic information to eliminate unwanted variation in care and patient outcomes. Start with REALD SOGI and social determinants of health data such as race, ethnicity, language, disability, sexual orientation, and gender identity. Co-create and update field values and selections after conducting focus groups with staff, patients, community advisors, neighboring health systems, state government agencies, and other stakeholders.
  2. Agree upon and ratify standardized mapping and reporting guidelines to ensure the health information technology system uses one single source of truth for assigning demographic categorization to patients that persist across encounters in their legal electronic medical record.
  3. Leverage existing or create formal quality and safety reports and dashboards, then calculate demographic subgroup-specific characteristics such as income, Zip code, language, disability, and race as process and outcome measures for the quality and safety metrics in question. For example, examine 30-day readmissions with subgroup readmissions and subgroup index admissions. Examine the Medicaid readmission rate by looking at patients with Medicaid experiencing 30-day readmissions and patients with Medicaid who had an index admission. Compare the Medicaid readmission rate to the commercial insurance rate by looking at patients with commercial insurance who experienced 30-day readmission and patients with commercial insurance who had an index admission. Start with metrics that are institutional priorities such as corporate objectives, risk-based contracted measures, and those that impact the most patients.
  4. With the group-specific outcome rates stratified, create an equity index showing relative risk or risk ratios across the subgroups. This should be viewed in a bar chart for ease of understanding. The control group should be dynamic and the group with the best outcome rate.

A recent example highlights collaborative change management led by Yale New Haven Health Pediatrics in partnership with Women’s Health, Hart says.

During the investigation, it was found that Black and Hispanic parents were disproportionately tested for substance use during pregnancy compared to White patients. Black and Hispanic parents were also less likely to be positive when tested than were White parents. Given this clinical disparity and the potential for bias-driven healthcare inequity leading to disproportionate child welfare system involvement with racially minoritized families, a partnership was formed with the regional child welfare system—the Connecticut Department of Children and Families (DCF)—to address the concern locally and structurally.

Yale New Haven Health partnered with patient advocates and multiple clinical departments and quality improvement experts. The health system also created a partnership with DCF to educate on the lack of medical necessity for routine newborn toxicology testing because it was redundant to parental testing, often did not change clinical management, and could lead to bias in future care.

These efforts led to two key changes, Hart says, noting the changes not only decreased the racial disparity in newborn toxicology testing but also decreased unnecessary testing in all racial groups for newborn toxicology testing.

  • There was policy clarification and procedural change to limit use of newborn toxicology to cases where it was purely clinically necessary and not used for social risk stratification or confirmation of parental testing. The health system created an electronic medical record clinical design support tool to guide clinicians in real time to highlight the need for informed parental consent being required to perform newborn toxicology.
  • There was broad internal messaging at the health system regarding the clinical disparity and to raise collective awareness among key stakeholders. There was internal and external training with hospital staff and DCF staff as well as re-evaluation of the statewide drug exposure reporting system to distinguish between mandated anonymous fetal substance exposure versus formal child welfare referrals for in utero substance exposure.

Christopher Cheney is the senior clinical care​ editor at HealthLeaders.