Can data offer a solution to our public health issues? Is data, indeed, the right solution? Most experts in the field will tell you that the answers depend on what you measure and what measures you implement. But to get the answers right, it turns out, peer learning is essential.
What We Measure
“What we measure informs what we do. And if we measure the wrong thing, we will do the wrong thing.” The economist Joseph Stiglitz said these words famously in his critique of using the GDP metric as a proxy for social progress.
Public health in the United States faces a similar issue, given that historically we’ve defined community health through a narrow and often negative lens. Think of average life expectancy, rates of obesity, or morbidity. These indicators are useful for telling us whether something is amiss – but they don’t tell us why.
In other words, a narrow, medicalized view of human health suggests that health problems are mostly rooted in chemical and biological changes in the human body. This overlooks the social, political, and historical contexts that make those physiological changes more likely in the first place.
Highlighting this trend in the field of mental health, for example, the psychiatrist Bessel van der Kolk wrote that “[t]he brain-disease model overlooks” the fundamental truth that “we can change social conditions to create environments in which children and adults can feel safe and where they can thrive.”
In fact, a narrow focus on the physiological indicators of health often leads to interventions that improve people’s lives in an immediate way but, shortly after, return them into the social conditions that had contributed to their poor health. A narrow measure of health often leads to half-measure solutions. Or as the President and CEO of the Robert Wood Johnson Foundation put it in a recent report, “It’s impossible to fix what isn’t measured.”
It follows that our response to matters of public health needs to be rooted in broader social, political, and historical explanations as much as in physiological ones. Experts in the field know this well, and an increasing focus on social determinants of health (SDOH) – such as education, employment, and food security – has been a testament to our evolving understanding of the complexities of health.
“We can change social conditions to create environments in which children and adults can feel safe and where they can thrive.”
The Measures We Take
The task then, it would seem, is to collect data on all the various factors that contribute to the health and well-being of communities. Collect and exchange as much data as you possibly can from a broad array of sectors, the thinking goes, and you’ll arrive at a more nuanced picture of our communities. A more complete story will surely lead to more complete solutions.
But just as a medicalized view of health overlooks the broader political, social, and historical contexts, a technological view of data often omits questions about how data systems were created, how they operate, who owns them, and who has access to them.
As the recently published Rising Equitable Community Data Ecosystems (RECoDE) report explains, “more data doesn’t mean better outcomes.” This is especially true in social settings where “[d]ata systems built to track housing, health, education, and employment are largely rooted in racist systems and discriminatory assumptions.
What’s more, when we deploy data to better understand the health and well-being of our communities, we’re primed to imagine that the solutions must also be data-driven. This is akin to the view that in a technologically advanced society every social ill must have a technological solution; as if to say that technology determines social progress. However, ignoring the broader context can lead us to implement data-driven health measures that simply reinforce the social systems our data-sharing work was meant to address.¹
To borrow from the ecological economist Tim Jackson, “Language sometimes situates itself a little too close to the object of its scrutiny.” In our case, initiatives that espouse the language of data with a narrow-minded enthusiasm are bound to collect and exchange data for data’s sake. The unfortunate result is that in the scheme for improving human affairs, we might overlook human beings.
It follows that we must try to free ourselves from the limits of our assumptions and biases as we think about measuring health and implementing better health measures. As the above-mentioned RECoDE report reminds us, “Technology isn’t a solution. Nor is data. They are tools that can help drive solutions, but they are meaningless without first considering the human needs.”
Building New Paradigms
Ultimately, we need to integrate social, political, and historical considerations into the data-sharing work we do. We need to go beyond our conventional ways of thinking. So, how do we do that? How do we think outside the box?
One useful approach that has been gaining traction is to engage with communities, organizations, and initiatives who share our goals but who can also challenge our understanding of both problems and solutions. In a word: peer learning.
In a recent paper,² Ruchi Patel, a Northwestern University master’s student and Illinois Public Health Institute (IPHI) intern (2021-2022), examined the benefits of peer-learning within the All In network’s Affinity Groups.
For those who aren’t familiar, All In is a nation-wide network of communities that share their successes and challenges from their data-sharing efforts with each other. (You can sign up to their monthly newsletter here.) DASH helps manage the All In network with its partners.³
Affinity Groups within the All In network are designed to foster peer learning around specific topics. Patel examined Affinity Group sessions that took place in 2021 where topics ranged from law and data, behavioral health, technology solutions, and housing. The discussions were led by the participants, while subject matter experts facilitated activities for each group.
Reviewing metrics on session recordings, feedback from surveys, and having gathered quotes and video clips, Patel evaluated levels of engagement and the reported usefulness of these Affinity Group discussions.
When asked if they would join the Affinity Groups again, 89% of respondents said that they would be somewhat or very likely to participate in future sessions. 69% of responses also indicated that the Affinity Groups would influence the data-related practices of the participant to some extent or a great extent.
As the findings suggest, peer learning has the potential to change what kinds of data various collaborations choose to collect and what measures they might implement. It’s no small conclusion that peer learning can challenge participants in data sharing projects to think critically about their own assumptions and biases. Put simply, peer learning helps us think outside the box.
As one participant shared, “I was able to present a challenge from our work for Affinity Group members to weigh in on, providing insights, resources, and strategies. Easy access to a broad network of initiatives actively engaged in multi-sector collaborations of organizations, community members, local and state government agencies and other stakeholders working together to improve health and well-being through data sharing has truly been invaluable for our work.”
Patel points out that this type of collaborative learning has, of course, room for improvement. Among others, she recommends allocating more time to certain discussions, but, in turn, have those discussions take place less frequently. Equally, smaller group sizes often lend themselves to a more engaged exploration of a topic. Ultimately, what works and what doesn’t work should be evaluated on a case-by-case basis.
In the end, however, it is clear that “peer-to-peer learning is a valuable method in this type of setting, as stakeholders from organizations can come together to discuss similar issues across the nation,” says Patel.
To ensure the integrity of the work, Patel met frequently with the research team at DASH to review the qualitative data she had collected and analyzed. As part of her master’s degree coursework, Patel also presented her research to academics at The Feinberg School of Medicine at Northwestern University.
Patel remarked, “I gained a great deal of experience from my coursework” and “gained more information about what All In and IPHI was through my applied experience project I did from April to August of 2021, which was more focused on creating virtual methods to increase engagement with webinars, affinity groups, the national meeting etc.”
If you’d like to delve into more detail about Ruchi’s and the DASH research team’s findings, keep an eye out for Part 2 of our Affinity Group blog series.
¹ This sentiment is echoed in the publication A toolkit for centering racial equity throughout data integration published by the Actionable Intelligence for Social Policy initiative at the University of Pennsylvania.
² Patel, R., 2022. How Peer Learning Advances Multi-Sector Collaborations: An Evaluation of the All In: Data for Community Health Affinity Group Program. Northwestern University.
³ DASH, in turn, is co-led by the Illinois Public Health Institute (IPHI) and the Michigan Public Health Institute (MPHI).