Tailoring Outreach Strategies for COVID-19 Vaccination: Using Data Science to Address Vaccine Hesitancy

This entry is part 10 of 17 in the series July 2023

Methods from data science and behavioral science can be used by public health organizations to segment populations of interest and design more tailored interventions.

Vaccine hesitancy has become a significant challenge in recent years, particularly in the context of COVID-19 vaccination. Our new article, “Population Segmentation for COVID-19 Vaccine Outreach: A Clustering Analysis and Implementation in Missouri,” highlights the findings and key takeaways from a project done in the state of Missouri where public health experts utilized data science and behavioral science methods to develop tailored outreach strategies to encourage COVID-19 vaccine uptake. By identifying distinct population clusters based on shared characteristics and social determinants of health, the study aimed to create more nuanced and equitable interventions for different segments of the population.

Our Approach

There are varying levels of vaccine acceptance across different communities. Factors such as age, race, and political affiliation contribute to differences in vaccine uptake. Communities of color and younger individuals have been disproportionately affected by the virus and have shown lower vaccine acceptance rates. We support the need to move beyond broad outreach efforts and instead focus on targeted interventions tailored to specific cohorts.

The project utilized data from the ShowMeVax Immunization Information System, the American Community Survey (ACS), and HealthPrismTM. These sources provided information on vaccination rates, demographic characteristics, and social determinants of health. Regression techniques were employed to identify variables associated with vaccination, and dimensionality reduction and clustering models were used to group census tracts into community segments.

A machine learning technique known as cluster analysis was employed to identify subgroups within the population based on shared features. This approach is commonly used in the private sector for customer segmentation but has seen limited application in public health. By grouping Missouri’s census tracts into distinct clusters, the researchers gained a more nuanced understanding of the population and were able to tailor outreach strategies accordingly. The research team also utilized a vaccination model that considers motivational and practical factors influencing vaccination behavior. Motivational factors include perceived risk, trust, and safety concerns, while practical factors encompass convenience and cost. The team hypothesized primary barriers to vaccination for each segment and developed corresponding intervention recommendations.

What We Found

Ten distinct community segments were identified, each with unique characteristics and vaccine uptake patterns. Geographic patterns emerged, with some segments concentrated in urban areas and others in rural regions.

The community segments with the highest vaccination rates, relative to the state average and other segments with lower vaccination rates, typically had higher educational attainment and income levels, slightly more diversity across racial groups, and low unemployment, with residents generally more likely to work in white-collar professions, suggesting residents in these communities may be experiencing fewer practical barriers to vaccination. However, differences in racial distribution and religiosity also played a role in distinguishing segments. Outreach strategies centered around leveraging social norms and highlighting the higher rates of community vaccination.

Community segments with average to slightly below-average vaccination rates, ranging from 52% to 57%, experienced some of the largest increases in vaccine uptake in the 3 months preceding this analysis. These segments had a mix of income that skewed lower, had higher proportions of racial and ethnic minority populations, and were more likely to be uninsured and utilize community health clinics, hospitals, or emergency departments for health care. Unlike the first group of segments, residents in these segments appeared to face both practical and motivational barriers, including lack of transportation access, lack of health insurance, and household income below $40,000 per year. Outreach recommendations focused on making vaccinations accessible to address practical barriers while building trust for those facing motivational barriers.

The remaining segments had the lowest overall vaccination rates (44% and 49%, respectively) and showed minimal recent uptake over the preceding 3 months. These segments’ residents had relatively lower levels of education, were more likely to live in rural areas, and were more likely to support religious causes. Residents experienced several practical barriers, such as lack of reliable broadband access; however, their primary barriers to vaccination were hypothesized to be motivational, given trends of vaccine resistance observed in rural parts of the state and among populations with lower educational attainment. Outreach strategies prioritized training trusted community leaders to encourage vaccination and offering ample choices around vaccination type and timing.

Read the article

Our findings in “Population Segmentation for COVID-19 Vaccine Outreach: A Clustering Analysis and Implementation in Missouri,” highlight the significance of data-driven methods and tailored interventions in addressing vaccine hesitancy. Public health departments should consider utilizing cluster analysis to identify subgroups within their population and develop customized outreach strategies. Practical barriers, such as transportation access and health insurance status, should be addressed through targeted interventions, while motivational barriers can be tackled by leveraging trusted messengers and tailoring messaging to address specific concerns within each community segment.

You Might Also Like

Eleanor Chessen, MPP, is a Manager in Deloitte’s Customer Strategy & Applied Design practice. She brings 9+ years of experience working with government and public health clients on design-led innovation, and she is passionate about using behavioral science to help clients design policies, programs, and systems that are more human-centered.

Jeffrey Glenn, PhD, is a Specialist Leader in Deloitte’s AI and Data Engineering Practice. He brings 17+ years of experience in the healthcare and education sectors and has authored 20+ peer reviewed papers and book chapters. He has extensive knowledge of behavioral science, research methods, statistics and communicating analytic results.

Madelyn Ganser, MSc, is a Senior Consultant in Deloitte’s Customer Strategy & Applied Design practice where she applies her background in psychology and global health to addressing the social determinants of health and driving health equity with organizations in the public health and the health care industry.

Series Navigation<< Designing Data Dashboards to Build Community Engagement Around Young People’s Well-Being  Successes and Opportunities for Leadership in State Health Departments for Public Health Professionals of Color >>

Sign up for our newsletter!

To receive news and information about the Journal of Public Health Management and Practice and JPHMP Direct.

We don’t spam! Read our privacy policy for more info.