Cancer Network spoke with Dr. Scarlett Lin Gomez about the role gender, socioeconomic status, race/ethnicity, and neighborhood characteristics have on health outcomes in cancer.
Today we are discussing disparities in cancer with Scarlett Lin Gomez, PhD, a professor in the Department of Epidemiology and Biostatistics at the Helen Diller Family Comprehensive Cancer Center at the University of California, San Francisco. Dr. Gomez studies the role of various factors, such as gender, socioeconomic status, race/ethnicity, and neighborhood characteristics, on health outcomes in cancer and other diseases. She is also the director of the Greater Bay Area Cancer Registry, a part of the California Cancer Registry and the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) Program. At the recently held American Association for Cancer Research (AACR) meeting, held March 29 to April 3, 2019, in Atlanta, Georgia, Dr. Gomez discussed how to create infrastructures that address multiple factors that can contribute to cancer health disparities.
Cancer Network: First, can you talk about some of the factors, beyond healthcare itself, you believe are most important in influencing health and health equality in the context of cancer?
Dr. Gomez: I think it’s interesting to acknowledge that it was only fairly recently that we as health researchers, particularly those of us focused on studying health disparities, have started to appreciate the importance of our neighborhoods and how they might potentially impact health and health equality. George Kaplan, a notable social epidemiologist, was among the first to empirically demonstrate a link between neighborhood environments and health outcomes. He said, "Health is location, location, location." But what does this really mean for us? Is it just about access to healthcare, which is the most studied factor? Well, access is certainly a part of it, but research has started to show that, in the grand scheme of things, healthcare is actually only a small part of our health.
Several conceptual models have been developed over the past several years, including the "self-society" model. This model posits that factors ranging from individual biology, including everything from our genetics to our health behaviors, to social networks, neighborhood characteristics, and broader structural and institutional forces, together interact to impact health outcomes. Other conceptual frameworks illustrate this in a more nested conceptual way, in which individual factors are nested within layers of networks, which are nested within neighborhoods, and in turn nested within our societies. Regardless of which framework researchers use to guide their research, one commonality is that institutional factors are conceptualized as a fundamental cause of health disparities. Institutional factors, such as segregation, structural racism, and policies, define the social determinants of health, and that is why we often consider inequalities in health outcomes to be inequities in health outcomes.
Cancer Network: Have you found that there are either misconceptions about or lack of awareness of the importance of these factors in influencing the health of individuals?
Dr. Gomez: Yes, absolutely. We think that structural forces, such as segregation, have a biological impact on health through an embedding process. Over one’s lifetime, stress that is associated with chronic exposure to things like discrimination, racism, and other adverse experiences can "get under the skin" and impact biological processes, such as epigenetic changes, and cause shortening of telomere length, which is the sign of biological age. We think that these chronically stressful experiences, especially if they occur over one’s lifetime, can also impact health through coping behaviors. For example, adverse childhood experiences, such as violence and abuse, have been linked to unhealthy behaviors later in life. Social and psychobehavioral scientists have been studying these phenomena for some time, but only recently have we, as epidemiologists, come to the table to collaborate in a multidisciplinary manner to study how these factors and processes might be linked to disease.
Cancer Network: You study social and other environments and how these factors can influence health. Can you provide an example of how social and other types of environments can influence cancer diagnoses or outcomes?
Dr. Gomez: Usually when researchers think about place and health, they think about access to healthcare. But, as David Williams, another notable social epidemiologist, said in the documentary Unnatural Causes, the choices of individuals are often limited by the environments in which they live. This means that health behaviors play a major role in health and disease. However, while some interventions have been successful in helping individuals to change their health behaviors, others have not, particularly those conducted in vulnerable and underserved populations. This is because these earlier generations of interventional studies have not accounted for the context within which individuals are embedded.
As an example, it can be much harder for some communities to access food, to maintain a healthy diet, and to engage in physical activity if there is simply no availability of affordable fresh fruits and vegetables. In some cases, neighborhoods are unsafe for walking or for children to play outside, or residents may generally keep to themselves because it is unsafe outside or because the neighborhood structure is not conducive to interactions. Also, if residents have to work multiple jobs that require long commutes, it is much more convenient for them to pick up something quick and cheap rather than spend time cooking. We think that, given the importance of context of place and neighborhood, individually-directed interventions to improve health behaviors are unlikely to have a positive impact if these types of social environments and contextual influences are not also identified and addressed.
Cancer Network: I imagine that it’s difficult to study the influence of these social and infrastructure factors on cancer outcomes. In your talk at AACR 2019, you discussed a breast cancer study you and your colleagues conducted. Can you talk about the structure of this study and your findings?
Dr. Gomez: It is definitely a challenge, as our research group has come to appreciate as we have been studying neighborhoods for the past 10 to 15 years. Every time we embark on a study, we learn more about how we can improve our methods for the next one. There are many different ways to measure neighborhoods and neighborhood attributes. What we have come to appreciate is that it is really important to think critically about which measures and methods are most appropriate and meaningful for the research question at hand. A good part of the work in my lab is collaborating with investigators to build resources and infrastructures that integrate data across multiple levels, from self to society.
At the AACR talk, I highlighted 4 such infrastructures. The breast cancer one, the Pathways Study,[1] was conducted among breast cancer survivors. This is an ongoing study that is co-led by Lawrence H. Kushi, ScD, of Kaiser Permanente, and Dr. Christine Ambrosone, PhD, of Roswell Park Cancer Institute. This is a survivor cohort in which we recruited over 4,500 women who were members of Kaiser when they were diagnosed with breast cancer. We are assessing aspects of survivorship, such as quality of life, side effects of treatment, health behaviors, and others. My team has been assembling a dataset to characterize the social environments of the women in this study. We are focusing on using three different types of data collection. First, we are geo-coding and then linking the addresses and locations of where the women live to existing geo-spatial data that we then use to characterize social and built environment attributes. Second, we ask women on one of our recurring surveys to report on their perceptions of their neighborhoods. Third, we audited their neighborhoods virtually by using omni-directional images from Google Street View. If the participants moved, or stayed in the same location but the neighborhood changed in terms of its characteristics over time, we will have the ability to study that.
In terms of findings, our first analysis was based on the existing geo-spatial data. We found that breast cancer survivors who were overweight were more likely to live in neighborhoods of lower socioeconomic status, and these neighborhoods were also more likely to include more ethnic minorities, higher traffic density, more fast food vs non–fast food restaurants, and more unhealthy vs healthy food outlets. We also found that breast cancer survivors who were obese were more likely to live in neighborhoods of lower socioeconomic status. In fact, there was a striking two-fold difference when we compared the highest socioeconomic quintiles to the lowest socioeconomic quintiles. Obese breast cancer survivors were also more likely to live in neighborhoods with ethnic minority compositions, and in those with more traffic, fast food restaurants, and unhealthy food outlets and fewer recreational facilities. We were interested to see whether these neighborhood factors actually explained racial and ethnic differences in overweight and obesity among breast cancer survivors and found that some, but not all, of them did. This initial study certainly points to the importance of neighborhood characteristics in terms of maintaining a healthy weight after diagnosis for breast cancer survivors.
Cancer Network: Are you or other researchers currently working on additional studies focused on cancer disparities?
Dr. Gomez: One study we are launching and starting recruitment for very soon is the Research on Prostate Cancer in Men of African Ancestry: Defining the Roles of Genetics, Tumor Markers, and Social Stress (RESPOND) study. This multi-site study is being led by Christopher Haiman, ScD, of the University of Southern California. The goal is to study the drivers of aggressive disease among African-American men diagnosed with prostate cancer. This is probably one of the most long-standing and perplexing disparities in cancer. To date, despite studies that have looked at individual aspects of this problem, we still don't have a good sense of what is driving the higher rates of aggressive prostate cancer among African-American men.
With this study, we are really taking a multi-focal approach. We will be looking at the impact of multi-level stressors across men’s lives, including at their genetics, genomic aspects of their tumor, and the inflammatory environment around the tumor. We are looking to recruit 10,000 African-American men in 7 states. Men across the nation will have the opportunity to volunteer as well and, again, we are just starting this study. We are looking forward to working with our collaborators and our community organizations. My team’s role will be studying the impact of neighborhoods and multi-level stressors, things like structural racism, experience with discrimination, and adverse experiences over the lifetime, and how those factors might be impacting the genetics of aggressive prostate cancer.
Cancer Network: Thank you so much for joining us today, Dr. Gomez.
Dr. Gomez: Thank you.
2 Commerce Drive
Cranbury, NJ 08512