Big Data for Reproductive Health (2020-2021)
Increasing access to family planning is a global health priority that improves women’s health, economic status and quality of life. Despite this, two-thirds of women in low-income countries who wish to prevent pregnancy discontinue contraceptive use within a year. Ensuring women continue using the contraceptive that allows them to meet their reproductive needs requires strong health systems and well-informed advocates and policymakers.
The main source of data on contraception discontinuation are multicountry household surveys administered every five years. Such contraceptive data are accurate, but not current. Complementary, more up-to-date sources of data are needed in this field. There is a precedent for using internet search and social media data to track flu trends, vaccine hesitancy and cervical cancer screenings, but little is known about the usefulness of social media data to track contraceptive use.
This project aims to examine social media data to analyze prospective trends in contraceptive use in the US and an African country. The Duke team will partner with IntraHealth’s digital health team to conduct landscape analysis on social media, contraceptive use and public health. Team members will investigate whether social media data sources, such as Google Trends and Twitter, can be used to measure contraceptive use and its discontinuation, design public health interventions and conduct disease surveillance.
In past years, this Bass Connections project team created a suite of online data visualization tools that allow stakeholders, advocates and policymakers to easily see trends in existing contraceptive data. Team members used machine learning to build a precision medicine tool that will help health workers determine the most effective contraceptive options for their populations. However, they had to rely on retrospective survey data, which can be outdated and difficult to analyze.
In 2020-2021, team members will use the existing research tools to conduct analysis of contraceptive use in the US and Kenya (or another African country) as revealed by social media data and develop a current data source to share with policymakers, stakeholders and advocates.
Policy brief for stakeholders; participation in conferences; published manuscript; proposals for grant funding
Fall 2020 – Spring 2021
Fall 2020: Conduct landscape analysis; review how social media has shaped public health; design study; assess which African country will be suitable for the in-depth study; employ global digital health networks to assist in analysis and selection process; conduct focus groups around social media in the US and the selected country; selected team members travel to Kenya to conduct focus groups and validate what team has learned
Spring 2021: Implement the project; conduct analysis of contraceptive use in the selected country and the US; begin accessing, sorting, organizing and analyzing the social media data for the countries of interest; travel to Washington, DC, and possibly to conferences
See earlier related team, Big Data for Reproductive Health (2019-2020).
Image from “Using a chord diagram to visualize dynamics in contraceptive use: Bringing data into practice,” by Amy Finnegan, Saumya S. Sao, Megan J. Huchko
- Amy Finnegan, IntraHealth International
- Megan Huchko, School of Medicine-Obstetrics and Gynecology
/graduate Team Members
Kelly Hunter, Public Policy Studies-PHD
/yfaculty/staff Team Members
Amy Herring, Arts & Sciences-Statistical Science
/zcommunity Team Members