The Global Health Innovators Seminar — Prof. Elaine Nsoesie — A quick recap of ep #14!
Data, Health, and Racial Equity
This blog is the high-level summary of the talk given by Prof. Elaine O. Nsoesie, Boston University at the PathCheck DICE Global Health Innovators seminar! Check out https://dice.pathcheck.org/talks.html to know more about previous talks!
Prof. Nsoesie mentioned various data sources and tools that capture both direct and indirect human interactions and how these could be used for various health monitoring use cases. She further alluded to her work on understanding the differential impact of various diseases and health policies through a racial lens.
Racial data tracker goals:
- Build a comprehensive resource of ratio data
- Make accessible data
- Improve the efficacy of research + highlight racial inequity
Research on surveillance tools to monitor disease patterns includes the collection of data at a timely and low-cost rate. The exposure of information on infectiousness spreads reduces data silos in individuals/communities
A few key tools and utilities:
- iwaspoisoned.com: pathogens have different incubation periods, which improves food quality indicators as fast-food chains
- Baidu, UptoDate: Search data for influenza in China to predict illness reoccurrences to predict influenza likelihoods
- Google: Analysis of counterfeited information and models for the spread of misinformation
- Instagram: Document discussions of food in Kenya, to indicate the prevalence of chronic disease, to implement interventions for healthy habits
- Yelp, Amazon: Reviewing foods and symptoms and health departments to track outbreaks of foodborne illnesses. Utilizing data sets from Amazon to build AI classifiers of food safety
- Mobile phone data: Attempt to record information in regions that don’t provide statistics, the effect of discussions between socio-economic groups on adherents to social distancing. Results show that people from high-income groups were already socially distant before the pandemic due to the flexibility of working from home. Lower-income groups could not immediately due to the labor demand likelihood of their work.
direct correlation between the provision of data and settlements of higher education/income
The study of demographics indicates gender groups can contrast in exercise dietary content when desegregated from social media. Eg. Females tend to appropriate towards yoga more and high caloric foods than males, which allure to cross fit and low caloric foods
Implications: Some groups are not represented in data groups, hence not service policies towards them, gaps of racial (BIPOC), gender, and income disparities.
- How can you reduce the data gap between extreme information silos and general trends?
Health departments generally acquire a report statement, though some communities do not report foodborne illnesses from culturally-induced /local limitations. One way to improve data reliance is by engaging with those communities to understand their communal behavior and intend to capture the solicited data.
- How can you reduce anomalous/ irrelevant data that might occur while analyzing Google Search Trends?
Combining the influx of arbitrary data from google, with electronic health records and clinician searches make it more efficient to deduce general trends and absolutism of validity. With online data, one of the limitations still is a lack of understanding of why each user searches intently for the specified branch of data.
Please check out dice.pathcheck.org/talks.html to watch this and previous talks and feel free to join the volunteer-driven Slack workspace of the PathCheck Foundation here: https://tiny.cc/pathcheckslack