# Using Data Science to Quantify the Impact of Misinformation, Mistrust, and Other Key Psychosocial Factors on Vaccine Hesitancy Among Vulnerable People Experiencing Psychopathology

> **NIH NIH K01** · UNIVERSITY OF MINNESOTA · 2024 · $187,261

## Abstract

People with severe mental illness (SMI) are ~3x more likely to contract and ~7x more likely to die from vaccine-
preventable disease. And yet, vaccine uptake is significantly lower among people with SMI, likely due to
greater vaccine delay/refusal despite availability (“hesitancy”). Critical knowledge gaps – including extremely
limited knowledge of whether and how unique features of SMI modulate hesitancy – impede selection of
interventions on vaccination for people with SMI. This K01’s purpose is to determine unique reasons for
hesitancy among people with SMI (Aim 1), reveal causal pathways to hesitancy in people with SMI and
forecast which variables in these pathways are likely the best intervention targets (Aim 2), and examine the
impact of education about herd immunity – which improves vaccine intentions in the general population – on
vaccine intentions and uptake in people with SMI (Aim 3).
In Study 1, we will recruit people with SMI and two comparison groups: people with depression/anxiety and
healthy people. Participants will complete measures of SARS-CoV-2 and influenza vaccine intentions/uptake,
of factors that may be uniquely relevant to hesitancy in people with SMI, and of variables implicated in general
population models of vaccine hesitancy. We will compare groups’ willingness to vaccinate, their level of
concern about vaccines’ dangerousness, and their endorsement of anti-vaccine information (Aim 1). We will
use active-learning causal discovery algorithms and intervention calculus to reveal causal pathways to
hesitancy in people with SMI, quantify the lower-bound total causal effect of variables in these pathways on
hesitancy, and prescribe future experiments to resolve any uncertainties in these causal pathways (Aim 2). In
Study 2, we will use established natural language processing tools to analyze rates of and attitudes to anti-
vaccine misinformation in Tweets from Study 1 participants. This analysis will complement our laboratory
assessments of anti-vaccine misinformation endorsement for Aim 1. In Study 3, we will recruit a subset of
participants from Study 1’s healthy and SMI groups. We will compare the effects of education about herd
immunity across these groups, with the goal of identifying differential efficacy and examining the need to tailor
existing interventions to people with SMI (Aim 3).
Our rigorous characterization of unique aspects of vaccine hesitancy among people with SMI, along with
causal pathways we reveal, our corresponding predictions about the impact of manipulating variables in these
pathways, and our quantification of herd immunity education’s effect on vaccine decisions, will provide critical
road-maps for empirically-informed interventions on hesitancy in people with mental illness. These road-maps
will move us closer to a world where effective, evidence-based interventions on hesitancy are deployed to
protect people with SMI from vaccine-preventable disease.

## Key facts

- **NIH application ID:** 10885682
- **Project number:** 1K01MH132899-01A1
- **Recipient organization:** UNIVERSITY OF MINNESOTA
- **Principal Investigator:** Michael V Bronstein
- **Activity code:** K01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2024
- **Award amount:** $187,261
- **Award type:** 1
- **Project period:** 2024-06-01 → 2028-05-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10885682

## Citation

> US National Institutes of Health, RePORTER application 10885682, Using Data Science to Quantify the Impact of Misinformation, Mistrust, and Other Key Psychosocial Factors on Vaccine Hesitancy Among Vulnerable People Experiencing Psychopathology (1K01MH132899-01A1). Retrieved via AI Analytics 2026-05-22 from https://api.ai-analytics.org/grant/nih/10885682. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*