All of Us Research Program Adds Social Determinants Survey

The NIH All of Us Research Program has launched a survey tool to collect social determinant data. Cheryl Clark, M.D., co-chair of the research effort’s social determinants of health (SDOH) task force, recently spoke with Healthcare Innovation about how this data could enhance precision medicine research by complementing the genomic and clinical data already being gathered.

Currently, 428,000 participants have signed up and completed the basic consent process and 305,000 have completed the initial steps, including consent, EHR connection, DNA sample, first few surveys, and physical height and weight measurements. Eighty percent of participants self-identify into a group that has been historically underrepresented in biomedical research, including more than 50 percent from a racial or ethnic minority.

The survey, available in English and Spanish, asks participants for information about social factors, including neighborhood safety, access to food, experiences with healthcare, discrimination, and daily work and living environments. Responses will provide researchers with data to better understand the connection between these social environmental factors and overall health. Participant responses to the new SDOH survey are expected to be available to researchers in the program’s platform for data analysis, called the Researcher Workbench, by mid-2022.

Clark, an assistant professor of medicine at Harvard Medical School, has conducted research focuses on social determinants of health as explanatory factors for racial and ethnic disparities in cancer risk.

“What excited me about the All of Us program is its emphasis on equity and diversity,” Clark said. “Since the beginning of the program, the idea has been to make sure that the participants are as diverse as they can possibly be. Across the U.S., we wanted to make sure that we have multiple types of data — for example, electronic health records and biological samples, and that we have multiple kinds of people from all walks of life — people who have disabilities, people from different racial and ethnic backgrounds, etc.”

Part of understanding how all that comes together is to ask people questions about their experiences, she said. Beyond questions about health status and conditions or income levels, “the important and next logical step is to ask people to think more broadly about their life and health,” Clark said. “I think All of Us has been very forward-thinking in this idea of precision medicine — that it's not just the genome. It's not just in the biology. That is very important. But we also need to really understand the environment and the person. So that's been baked into the program. And it was a very natural extension to develop this module and to prioritize it early during the course of the study.”

The All of Us program has already seen the value of collecting this type of data, stemming from the results of its COVID-19 Participant Experience (COPE) survey, which was completed by nearly 100,000 participants from May 2020 through February 2021, and included information on COVID-19 symptoms, stress levels, social distancing, discrimination, and economic impacts.

The COPE survey found that 20 percent of respondents indicated that they worked more hours during the pandemic and 20 percent responded that they were not paid at all during the pandemic. Another 10 percent reported losing their jobs. These responses, coupled with the insights from the data set’s electronic health records, Fitbit, and other data, present a more complete picture of participants’ experiences during the pandemic, according to the All of Us researchers.

“When the pandemic hit, there was a really rapid mobilization to try to make sure that we captured as much information as we could about the pandemic,” Clark said. “Other researchers will talk to you about the contributions of the biological studies and how the All of Us program was able to understand the spread of COVID by looking at data from participants. But the COPE survey, in particular, was able to ask some questions related to social determinants of health to get a sense of how the pandemic was impacting people socially, for instance in terms of loneliness and isolation.”

I asked Clark if it was challenging for the task force to narrow down the number and type of questions in the survey. “We really tried to center equity in the way that we thought about this,” she replied. “If you are going to ask a million people questions, you want to make sure that you get more signal than noise. We have to be really rigorous, really specific about the things that have a long history or trajectory in terms of the resource and the use cases in the literature. We tried to be selective in terms of what we asked. The questions that wound up making it onto the surveys met that Venn diagram of things that were meaningful for equity, that were pulled from scientific validated surveys, and that were important to our participants. And we did have participant partners, folks who are members of the All of Us research program, not scientists, but lay folks looking at it, making sure it was relevant.”

Clark added that this type of data collection feeds into her own work as a clinician.

“I see people who come in during really vulnerable parts of their lives when they're hospitalized and feeling pretty ill, including patients who I took care of during the COVID pandemic,” she said. “Remembering to have that holistic view of people — taking care of their clinical needs, as well as thinking about their social and the other parts of who they are really helps to humanize people. And I think it makes care more effective. So this module really represents all of that. And as a researcher, it is really exciting, because not only can I think about this clinically, but I can start to contribute to the evidence base that helps us to think about the interface of all these issues. So it's really meaningful for me personally.”