In the spirit of Valentine’s Day, the Office of Research Integrity and the University Library have organized a series of events featuring data use in research across disciplines as part of worldwide Love Data Week, which runs from Feb. 10 through 14.
While the inception of international Love Data Week dates back to 2016, this is the first year that the University is participating.
The week is meant for both students and faculty who are involved with research, said Senior Director at the Office of Research Integrity Keri Godin, one of the event organizers.
Many of the week’s events encapsulate this year’s theme of “get to know the data specialists at your institution, the kinds of work they do and the data and associated issues with which these data specialists engage,” according to the University’s Love Data Week website.
Tuesday’s Love Data Week event at The Policy Lab, “Using Administrative Data to Investigate the Opioid Crisis in Rhode Island,” highlighted how data can be translated into policies, said Andrew Creamer, scientific data management specialist and a University librarian for the computer science and the cognitive, linguistic and psychological sciences departments. Creamer was also involved in planning Love Data Week.
The event featured Director of Data and Analytics at the Executive Office of Health and Human Services Kim Paull, who discussed how and why data can be used to get a fuller picture of the opioid crisis and channel energy and resources into effective solutions.
Paull got involved in state efforts to combat the opioid crisis when Governor Gina Raimondo’s Overdose Prevention and Intervention Task Force asked her to help coordinate the metrics for their strategic plan, she said at the event. As she learned more about the communities involved in this struggle, she realized that there are many metrics for success, and that linking data from diverse agencies could yield information that matches the complexity of the issue at hand.
“With integrated data, we tell a more complete story,” Paull said.
After the event, Paull told The Herald that this data is most useful in situations in which many different agencies want to tackle a problem, but don’t know how to help. “Linked administrative data helps us say, ‘Here’s the complete picture, and here’s how you all relate to each other … here’s how you can be most helpful,’” she added.
Integrated data analysis is particularly well-suited to tackling the opioid crisis because the cycle of addiction and recovery is “a very human process,” Paull said. “What’s happening is a confluence of external and internal factors, and … those things don’t live with one agency.” She added that “we have to bring as much data as we can to bear if we’re going to even attempt to tell the story and to help people in a way that meets them where they are.”
In an effort to tell this human story, Paull’s team linked Medicaid data for those with a diagnosed opioid use disorder (OUD) or overdose (OD) with other key data sets, including earned wages, state benefits and interactions with Corrections, to understand which factors correlate with enrollment in medication-assisted treatment (MAT) within six months of that signal event. While MAT is considered “the gold standard path to recovery,” Paull added, this is not the only path.
Paull and her team found that people often did not initiate or continue treatment due to bias, fear and discrimination, stemming from “a pernicious stigma” against OUD. They also found that factors such as anxiety, back pain and alcohol use disorder played a role in keeping people out of treatment.
“The biggest thing we can do … is to make sure that people who are responding at the point of overdose … support medication-assisted treatment,” Paull said, adding that the state has established “a number of programs supporting MAT-focused peers and empowering provider champions to say, ‘In my organization, this is important to me. We should be doing this.’”
Paull added that the Governor’s task force’s bold new initiatives required bold new metrics, which necessitated linked data that didn’t previously exist.
“No one dataset is going to tell the full story, and we have to respect the fact that humans are complex organisms, and if we’re not trying to connect the data, we’re not trying hard enough,” Paull told The Herald. “If we attempt to tell stories based on one dataset, we’re missing something pretty significant.”
Matthew Santacroce, chief of staff at The Policy Lab, emphasized the importance of Paull’s work to both governmental and non-governmental audiences. He called her team’s linked data analysis “an enormously valuable asset in the public policy conversation here in this state, both to folks in government — administering programs and making funding decisions — (and to) folks in academia.”
Santacroce added that this work will continue to shape the conversations around “pressing issues that are facing the state of Rhode Island, from the opioid overdose crisis to how we’re doing with workforce training and job development, to any number of things that we haven’t really even contemplated yet.”
Paull hopes event attendees came away from her talk thinking about how they could help in the fight against the opioid crisis in Rhode Island. “If you want to help, there’s a way to help, and it’s asking the right questions, and it’s caring enough to know that the folks that are suffering are complex individuals who are a hair’s breath away from where I stand right now,” Paull said.
Tom MacMullen ’20 attended Paull’s talk after taking a Wintersession class on the opioid crisis, which made him curious about potential solutions. “I want to know about how data can be applied to this issue,” he said.
Expanding on Paull’s answer to this question, Wednesday’s panel “Leveraging Big Data to Improve the Public’s Health,” featuring Associate Professor of Epidemiology Brandon Marshall’s research group, will further discuss the applications of big data to public health and the efficacy of resulting policies and interventions, including those relating to the opioid epidemic, Creamer said.
Through Love Data Week’s other events, the community can learn how to properly record, share and present data, Godin said.
But a lack of knowledge about how people should use their data and potential unwillingness to disclose all data could present a barrier, Creamer added.
Love Data Week therefore aims to introduce the University community to the numerous available resources that can help researchers follow data use guidelines. “Given our focus in the coming years on really starting to be more proactive about use of data and sharing of data, (Love Data Week) was a natural extension of that vision,” Godin said.
Creamer also hopes that having experts speak about their experiences using data could motivate others to follow in their footsteps.
Love Data Week will conclude Friday, but its organizers hope to make it an annual celebration at the University and have groups volunteer to participate and promote knowledge on data in research, Godin said.
“What we’re trying to do is demonstrate a true institutional commitment to research integrity,” Godin added. “We’re not just conducting research at Brown to get the most publications, to get the most funding. … We’re doing it because we are really committed to the conduct of research at Brown with the highest degree of integrity … because we want to have the most global impact.”