School of Pharmacy’s Data to Action Research Team (DART) provides student pharmacists with hands-on experience in data science, conducting epidemiological research to support evidence-based practice.
How are the conditions of daily life and health disparities playing out in the COVID epidemic? What are some data sources available that can help us understand what makes an individual more vulnerable to COVID? Can the data also tell us how to deploy a vaccine?
These are the questions that the Data to Action Research Team (DART) housed within the DatAction! program in the Department of Pharmaceutical Health Sciences Research (PHSR) at the University of Maryland School of Pharmacy seeks to tackle.
Soon after the COVID-19 pandemic began, DART started to engage pharmacy students at the School in evidence-based research focused on the relationship between the social determinants of health (SDOH), health disparities, and COVID-19 trends.
To conduct its research, DART leverages a cloud-based data and visualization platform, which better facilitates real-time COVID-19 epidemiological research. In the spirit of “learning by doing,” student pharmacists have an opportunity to tackle this real-world issue, putting their learning into action, gaining much needed data science skills, and producing research with actionable results.
To-date, the team, which is led by Fadia T. Shaya, PhD, MPH, professor and director of the Center on Drugs and Public Policy at the School of Pharmacy and director of informatics and implementation research for the University of Maryland, Baltimore (UMB) Institute for Clinical and Translational Research (ICTR), and includes Apoorva Pradhan, MD, MPH, postdoctoral fellow with DART and the Behavioral Health Resources and Technical Assistance (BHRT) program at the School of Pharmacy, and nine Doctor of Pharmacy (PharmD) students, has started to examine the following domains of SDOH in Maryland.
Health and Health Cares: Assisted by student pharmacists Anthony Iacovelli and Alice Lu, DART is studying the impact of Medicare fee-for service benefits on the prevalence and mortality related to COVID-19. Preliminary results have shown a positive correlation, with counties that contain more Medicare recipients having higher rates of both COVID-19 cases and related deaths. Similar trends were also uncovered when the team examined the number of Medicare recipients with preexisting health conditions, such as chronic obstructive pulmonary disease (COPD), diabetes, or coronary artery disease. Other variables currently under study include poor mental and physical health. The association of poor mental and physical health with COVID-19 cases and deaths is still unclear, possibly because they are confounded by a number of related variables, such as poverty, health care access, education, broadband internet access, population density, and other comorbidities. The team’s adjusted models will help address this question.
Social and Community Context and Disparities: Race and ethnicity are important factors in both the social and community context of the SDOH. Results from preliminary analyses conducted by student pharmacists Christina Ha, Amy Bao, and Olamide Olujohungbe show that a higher prevalence of COVID-19 was observed in Maryland counties with a higher percentage of African American residents. Similar trends were observed in the area of COVID-19 related mortality. More than half of counties with low percentages of African Americans residents (0-20%) experienced a lower COVID-19 related death rate. Another factor that the team is examining is violent crime. Violent crime rates are affected by numerous factors that differ between locales, including, but not limited to: poverty level, education level, degree of urbanization, family cohesiveness, and attitudes towards crime. Some evidence suggests that crime rates have decreased during the pandemic, especially during those times when COVID-19 lockdowns were in place. However, regarding violent crimes — especially domestic violence or child abuse — the evidence is unclear. Results from data collected in the state of Maryland show some correlation between low violent crime rates and low COVID-19 infections and deaths, but no evidence of any strong correlation between high violent crime rates and high COVID-19 infections and deaths was observed. Further analysis is ongoing to help the team ascertain the exact relationship between COVID-19 and violent crimes. Olujohungbe is also examining disparities in infection and mortality rates. As disparities in enrollment in clinical trials are also noted, this topic remains relevant for ongoing COVID-19 vaccine trials.
Neighborhood and Built Environment: Student pharmacists Meghna Bhatt and Allison Thompson are exploring the relationship between this SDOH and COVID-19 based on food insecurity and internet access. Access and consumption of healthy food is associated with lower risk of chronic health conditions. However, lockdowns due to COVID-19 have led to additional barriers that prevent many people from accessing healthy food, including access to transportation, which put further strain on food deserts due to hoarding of food and supplies, and personal economic strain. Preliminary analysis of the data indicates that there may be a positive association between food insecurity and prevalence of COVID-19, underscoring how food insecure populations may be at an increased risk for various negative health outcomes and health disparities. Internet access could have a significant impact on employment in the current work-from-home environment. The team was discovered that reliable access to the internet varied across different counties in the state of Maryland, ranging from 74-95 percent. Preliminary analysis did not find any correlation between internet access and COVID-19 related mortality or prevalence. Further investigation into different types of internet access – for example, broadband versus dial-up – health literacy, and the impact of digital health equality is ongoing.
Education and Health Literacy: When this research was initially conducted, the team found that in the two Maryland counties with the lowest infection rates — Garrett and Allegheny counties – more than 40 percent of the population has a high school diploma. In contrast, the two areas with the highest infection rates — Baltimore City and Prince George’s counties — only 29 and 25 percent of residents have a diploma, respectively. The U.S. Bureau of Labor Statistics reports that only 10 percent of people without a high school diploma are able to work from home, compared to 37 percent of people with a bachelor’s degree. The typical work setting of individuals without a high school diploma, as well as associated rates of health literacy and attitudes towards the virus, seem to be possible reasons for the link, although many socioeconomic factors are also likely at play. The team – which is led by student pharmacist Griffin Sauvageau — cannot yet point to any one population characteristic, but team members continue to look at a number of variables
Economic Stability and Financial Literacy: U.S. economic activity has slowed as a result of lockdowns, delayed re-openings, stay-at-home orders, and quarantine mandates, leading to a high number of individuals losing their jobs and filing for unemployment. In the state of Maryland, the unemployment rate was highest in Worcester County (7.4 percent), second highest in Somerset County (6.1 percent), and the third highest in Baltimore County (5.1 percent). However, preliminary results from this team led by student pharmacist Christina Ha show that higher unemployment rates did not necessarily correlate with a higher number of COVID-19 cases or related mortality. Furthermore, the preliminary analysis of the relationship between the rate of cases and mortality related to COVID-19 in Maryland and the proportion of people living in poverty also shows no significant correlation. However, it is important for researchers to examine longer term trends.
Individual associations between each SDOH variable and COVID-19 prevalence and mortality have been assessed; however, in order to evaluate the impact of SDOH on COVID-19, the team plans to move forward with assessing the relationship between prevalence and mortality associated with COVID-19, while simultaneously adjusting for all of the variables identified above.
DART invites individuals from the School of Pharmacy and UMB’s six other professional schools, to join its work. Team members have found student pharmacists are particularly interested in learning more about informatics and data science. With its approach of “learning by doing,” DART members meet weekly to discuss relevant clinical questions, including how those questions relate to prevention, treatment, public health, and SDOH; translate these questions into research questions; select and apply methods to help answer the questions; and derive actionable results. DART members learn and practice how to implement the results uncovered into practice to help inform pharmaceutical and health policy and action. Thus, the name: Data to Action Research Team — DART!
For more information, or to join DART, please contact fshaya@rx.umaryland.edu.