Dataset Information

Taking a machine learning approach to optimize prediction of vaccine hesitancy in high income countries.

ABSTRACT: Understanding factors driving vaccine hesitancy is crucial to vaccination success. We surveyed adults (N = 2510) from February to March 2021 across five sites (Australia = 502, Germany = 516, Hong Kong = 445, UK = 512, USA = 535) using a cross-sectional design and stratified quota sampling for age, sex, and education. We assessed willingness to take a vaccine and a comprehensive set of putative predictors. Predictive power was analysed with a machine learning algorithm. Only 57.4% of the participants indicated that they would definitely or probably get vaccinated. A parsimonious machine learning model could identify vaccine hesitancy with high accuracy (i.e. 82% sensitivity and 79-82% specificity) using 12 variables only. The most relevant predictors were vaccination conspiracy beliefs, various paranoid concerns related to the pandemic, a general conspiracy mentality, COVID anxiety, high perceived risk of infection, low perceived social rank, lower age, lower income, and higher population density. Campaigns seeking to increase vaccine uptake need to take mistrust as the main driver of vaccine hesitancy into account.

SUBMITTER: Lincoln TM

PROVIDER: S-EPMC8827083 | biostudies-literature | 2022 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Taking a machine learning approach to optimize prediction of vaccine hesitancy in high income countries.

Lincoln Tania M TM Schlier Björn B Strakeljahn Felix F Gaudiano Brandon A BA So Suzanne H SH Kingston Jessica J Morris Eric M J EMJ Ellett Lyn L

Scientific reports 20220208 1

Understanding factors driving vaccine hesitancy is crucial to vaccination success. We surveyed adults (N = 2510) from February to March 2021 across five sites (Australia = 502, Germany = 516, Hong Kong = 445, UK = 512, USA = 535) using a cross-sectional design and stratified quota sampling for age, sex, and education. We assessed willingness to take a vaccine and a comprehensive set of putative predictors. Predictive power was analysed with a machine learning algorithm. Only 57.4% of the partici ...[more]

PMID: 35136120

Similar Datasets

Project description:Vaccine hesitancy forms a critical barrier to the uptake of COVID-19 vaccine in high-income countries or regions. This review aims to summarize rates of COVID-19 hesitancy and its determinants in high-income countries or regions. A scoping review was conducted in Medline®, Embase®, CINAHL®, and Scopus® and was reported in accordance with the PRISMA-SCr checklist. The search was current as of March 2021. Studies which evaluated COVID-19 vaccine hesitancy and its determinants in high-income countries (US$12,536 or more GNI per capita in 2019) were included. Studies conducted in low, lower-middle, and upper-middle income countries or regions were excluded. Factors associated with vaccine hesitancy were grouped into four themes (vaccine specific, individual, group, or contextual related factors). Of 2237 articles retrieved, 97 articles were included in this review. Most studies were conducted in U.S. (n = 39) and Italy (n = 9). The rates of vaccine hesitancy across high-income countries or regions ranged from 7-77.9%. 46 studies (47.4%) had rates of 30% and more. Younger age, females, not being of white ethnicity and lower education were common contextual factors associated with increased vaccine hesitancy. Lack of recent history of influenza vaccination, lower self-perceived risk of contracting COVID-19, lesser fear of COVID-19, believing that COVID-19 is not severe and not having chronic medical conditions were most frequently studied individual/group factors associated with increased vaccine hesitancy. Common vaccine-specific factors associated with increased vaccine hesitancy included beliefs that vaccine are not safe/effective and increased concerns about rapid development of COVID-19 vaccines. Given the heterogeneity in vaccine hesitancy definitions used across studies, there is a need for standardization in its assessment. This review has summarized COVID-19 vaccine hesitancy determinants that national policymakers can use when formulating health policies related to COVID-19 vaccination.

Project description:We study the impact of public health messages on intentions to vaccinate and vaccination uptakes, especially among hesitant groups. We performed an experiment comparing the effects of egoistic and altruistic messages on COVID-19 vaccine intentions and behaviour. We administered different messages at random in a survey of 6379 adults in December 2020, following up with participants in the nationally representative survey Citizens' Attitudes Under COVID-19 Project covering nine high-income countries (Australia, Austria, France, Germany, Italy, New Zealand, Sweden, the UK and the USA). Four alternative interventions were tested, based on narratives of (1) self-protection, (2) protecting others, (3) reducing health risks and (4) economic protection. We measure vaccination intentions in the December 2020 survey and elicit actual vaccination behaviour by respondents in the June/July 2021 survey. Messages conveying self-protection had no effect on vaccine intentions but altruistic messages, emphasising protecting other individuals (0.022, 95% CI -0.004 to 0.048), population health (0.030, 95% CI 0.003 to 0.056) and the economy (0.038, 95% CI 0.013 to 0.064) had substantially stronger effects. These effects were stronger in countries experiencing high COVID-19 mortality (Austria, France, Italy, Sweden, the UK and the USA), where health risks may have been more salient, but weaker and, in several cases, not significant where mortality was low (Australia, Germany and New Zealand). On follow-up at 6 months, these brief communication interventions corresponded to substantially higher vaccination uptake. Our experiments found that commonly employed narratives around self-protection had no effect. However, altruistic messages about protecting individuals, population health and the economy had substantially positive and enduring effects on increasing vaccination intentions. Our results can help structure communication campaigns during pandemics and are likely to generalise to other vaccine-preventable epidemics.

Dataset Information

Taking a machine learning approach to optimize prediction of vaccine hesitancy in high income countries.

Publications

Taking a machine learning approach to optimize prediction of vaccine hesitancy in high income countries.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets