Toward equal footing

A conversation with health equity expert Alyce Adams

Featured Media for Toward equal footing

The passion Alyce Adams, PhD, has for improving health outcomes for people in marginalized communities was inspired by the suffering she witnessed of chronically ill elderly relatives in California, Oklahoma and Arkansas.

Adams’ desire to understand how adults who had health insurance could have such catastrophic health outcomes eventually led her to Harvard University, where she earned a doctorate in health policy.

She has since spent two decades at academic and research institutions conducting research to inform policies that address and prevent suboptimal treatment for people with multiple chronic conditions, particularly in minority and low-income communities.

Now, Adams, who holds the inaugural Stanford Medicine Innovation Professorship, is a leader in the Stanford School of Medicine’s efforts to improve the quality of health care and outcomes for minoritized individuals and those from underserved communities.

In addition to teaching and mentoring, Adams co-directed the Stanford Medicine Commission on Justice and Equity’s working group on health equity excellence. She is also the Stanford Cancer Institute’s associate director for cancer health equity and the Department of Health Policy’s associate chair for health equity and community engagement.

Priya Singh, Stanford Medicine’s chief strategy officer and senior associate dean and a member of the Commission on Justice and Equity, recently talked with Adams about her health disparity discoveries and her vision for forging meaningful health equity advances at Stanford Medicine and beyond.

Singh: What motivated you to do the work you do today?

Adams: Growing up, I saw the devastating impact that chronic conditions like diabetes had on my own family members. Worse still, I was troubled by the acceptance that these outcomes, such as repeat amputations and blindness, were just the consequences of being a Black person with chronic illness. It sparked my interest in health policy and its potential to improve health care and address these disparities.

Singh: Why does chronic disease disproportionately affect marginalized groups? What can be done to help?

Adams: The drivers of disparities include factors at the patient, interpersonal, health system, community, societal and policy levels. For example, let’s take medication adherence: Historical racial injustices toward Black patients have led many to distrust doctors and the treatments they prescribe. Similarly, physicians’ personal beliefs about whether a certain patient can pay for newer medications may influence what is prescribed.

“Including people from communities affected by adverse health conditions enhances the quality of the work and its potential to improve health.”

Alyce Adams, PhD, a Stanford Medicine leader and expert in health equity policy

Health systems also might not have enough interpreters to help people who aren’t fluent in English navigate their care options. Further, in some communities, the rising cost of housing and food can prevent a patient from being able to buy medications or follow through on their treatments.

Focusing on a single driver, such as access to high quality health care, is not enough to eliminate these disparities. We have to look at broader health and social policies as well — something we are doing at the new Department of Health Policy, where we take a multidisciplinary approach to developing policy solutions to address health inequities.

Singh: Can you talk about how we can include more medically underserved people in medical research and why that’s important?

Adams: At its core, impactful research involves asking the right questions and finding answers. To have the most impact, engaging with patients, communities and practitioners is critical.

As scholars, we are experts at posing questions based on our theoretical understanding of the world and available evidence to support or refute those theories.

However, people with lived experience and practitioners working to address their needs are better able to help us prioritize research questions with the greatest potential for impact. For example, community partners can identify environmental factors that may contribute to disparities in cancer and other conditions.

Too often, the people who are selected as research participants or who volunteer to be part of the research team are not those who would most benefit from the research.

Including people from communities affected by adverse health conditions enhances the quality of the work and its potential to improve health. We are partnering with community care providers and advocacy groups to educate communities about clinical trials and make patient participation easier.

In another collaboration, the Stanford Cancer Institute’s clinical trials office and Stanford Health Care are evaluating clinical trial proposals with an eye toward diversity, equity and inclusion. For example, we are encouraging scientists to consider opening trials to people who have more than one illness — who are typically excluded — to increase participation from patients who are more ethnically and geographically diverse.

Singh: You have spoken about the need to have “equal opportunity for positive health outcomes.” Can you elaborate on this concept and explain how we can achieve it? 

Adams: People with the greatest need are often the last to benefit from health care innovations. Therefore, in addition to improving access to high quality care, we need to identify strategies within and outside of health systems that can accelerate getting the benefits of treatment innovations to the most vulnerable patients.

The key to making this happen is building partnerships between researchers, patients, communities, practitioners and policymakers to create interventions that have equity at the forefront of innovation.

For example, our research lab is working with cancer patients, survivors and clinicians to develop and evaluate an algorithm to identify which patients are at risk for chemotherapy-induced neuropathy. We are also working with these groups to understand patient perspectives on balancing the risks and benefits of chemotherapy.

“In the same way that we have to be open to the possibility that our hypotheses are wrong, we have to be open to accepting where our methods may be flawed.”

Alyce Adams

Singh: You mentioned using algorithms in cancer care. How else can new technologies such as artificial intelligence help address health disparities?

Adams: Exceptional scholars here in the Department of Health Policy and elsewhere have rightly drawn attention to the potential of algorithms to create or exacerbate bias in clinical care and outcomes.

At a minimum, biomedical ethics should be applied to the development, deployment and adaptation of these algorithms to reduce potential harms and maximize potential benefits.

Toward that end, we can ensure that the data used to train algorithms represents those affected by the algorithm, engage community and clinician partners in the development of workflows for implementation and monitoring of algorithms, and use artificial intelligence to augment rather than replace autonomous decision-making by patients and providers.

We are also paying attention to how these algorithms might inform the development of more equitable health care practices. Some of our work with colleagues at Kaiser Permanente, Harvard and Northeastern, for example, uses causal inference and machine learning to emulate randomized controlled trials to identify intervention combinations that might address underuse of clinically effective medications due to cost.

Singh: Where are you seeing the most progress and what gives you hope? 

Adams: I am heartened by the current focus on structural and systemic drivers of health inequities and efforts to improve research practices. As scientists, we have to be willing to cast a critical eye on our own work and change our approaches to acknowledge and address previously unacknowledged biases. For example, some journals are requiring publications of machine learning algorithms to include a discussion of potential biases toward minoritized subgroups.

In the same way that we have to be open to the possibility that our hypotheses are wrong, we have to be open to accepting where our methods may be flawed. That willingness to be introspective and curious about the future is at the heart of what we do.