Aim 2. Multiplatform Social Media Analysis for COVID-19 Clinical Trial Experiences
01
Study Design
Aim 1 of this project systematically mined U.S.-based COVID-19 clinical trial data from ClinicalTrials.gov. The process began by filtering for relevant keywords, conditions, and MeSH terms related to COVID-19. Key attributes such as study type, phase, design, conditions, and interventions were extracted. Natural Language Processing (NLP) techniques were used to identify additional information on minority populations and social disparities.
​
The data was organized into a COVID-19-specific dataset, linked to geospatial information, and managed using Python scripts and a data lake in S-3. Secondary data from sources like the American Community Survey and CDC WONDER were integrated for multivariate statistical testing on health equity and community-level variables, focusing on underrepresented communities in COVID-19 research.
​
Geospatial analysis examined the distance from clinical trial sites to identify clusters of health equity interest. The project resulted in a continuously updated nationwide database of COVID-19 clinical trials, with all software made available as open-source on GitHub.
02
Key Findings
-
37,043 US sites for 1,654 COVID-19 clinical trial studies listed on CT.gov, with 70% being interve4ntonal and over 30% of these being Phase 2 trials.
-
The Los Angeles and New York City areas were statistically significant hot spots across all study types.
-
Clinical study sites were more likely to be found in counties with higher proportions of Asian American (p<0.001) and Native American residents (p<0.001).
-
Areas with greater concentrations of African American residents had significantly lower concentrations of observational (p<0.001) and government-sponsored COVID-19 studies (p=0.003) in national analysis and significantly fewer concentrations of study sites in both Los Angeles (p<0.001) and New York (p=0.007).
​
03
Project Deliverables
-
Publication (Paper) - Demographic Disparities of COVID-19 Clinical Study Site Proximity in the United States: A Geospatial Analysis (Under Review)
-
Conference Poster - Identifying Spatial Disparities in Online Discussions About Clinical Trials for COVID-19 Vaccines. APHA Annual Meeting & Expo 2022, Boston Nov. 6-9. Abstract Poster