Predicting Vaccine Hesitancy Using Machine Learning

Factors Influencing Vaccine Hesitancy Among Parents of Children Aged 0-48 Months: A Machine Learning Prediction

In recent years, emerging technologies such as the Internet of Things (IoT), Artificial Intelligence (AI), and Virtual Reality (VR) have rapidly become integrated into daily life. The widespread use of these applications has led to the accumulation of vast amounts of data, giving rise to what is commonly referred to as "Big Data." Due to the sheer volume, manual processing and analysis of these large datasets are not feasible. Therefore, software tools and libraries-such as Python and R libraries-have been developed to perform these analyses efficiently and to generate predictions for the future by leveraging historical data through Machine Learning (ML) algorithms.

The primary goal of machine learning algorithms is to discover patterns within existing data and use these patterns to make accurate predictions on new data. The use of machine learning in the field of healthcare has gained significant momentum in recent years. However, a review of the literature reveals that research specifically addressing childhood vaccine hesitancy remains limited.

This study aims to identify the factors contributing to vaccine hesitancy among parents of children aged 0-48 months and to develop a predictive model using machine learning techniques based on these factors. Such a model could help anticipate the likelihood of vaccine refusal among parents and thereby support the development of targeted public health strategies for at-risk populations.

Study Overview

• Status: Completed
• Conditions: Children; Machine Learning; Vaccine Hesitancy; Vaccine Refusal

• Study Type: Observational
• Enrollment (Actual): 461

Contacts and Locations

Study Locations

• Turkey (Türkiye)
  • Yalova, Turkey (Türkiye), 77200
    • Yalova University

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: Yes

• Sampling Method: Non-Probability Sample

Study Population

No sample size calculation was performed in this study. Between June 2025 and September 2025, all parents of children aged 0-4 years who met the inclusion criteria and could be reached with the questionnaires were included in the study. Snowball sampling method will be used to reach the parents. The study population consists of parents who have children aged 0-4 years, possess proficiency in understanding and reading the Turkish language, and can be accessed via social media platforms using technological devices.

Description

Inclusion Criteria:

Parents who meet the following criteria will be included in the study:

• Have proficiency in understanding and reading the Turkish language,
• Are able to use technological devices such as mobile phones or computers to access social media platforms (Participants used social media platforms such as Instagram and WhatsApp to communicate) via the internet,
• Are 18 years of age or older,
• Have children aged 0-4 years,
• Consent to participate in the study,
• Have children without any absolute contraindications to vaccination.

Exclusion Criteria:

• Parents of high-risk infants (such as those with a history of systemic illness, preterm labor, anomalies, etc.),
• Parents who are unwilling to participate in the study will be excluded.

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort

Parents of children aged 0-48 months; Parents who meet the following criteria will be included in the study: those who have proficiency in understanding and reading the Turkish language; are 18 years of age or older; have children aged 0-4 years; are able to use technological devices such as mobile phones or computers to access social media platforms (Participants used social media platforms such as Instagram and WhatsApp to communicate) via the internet; consent to participate in the study; and whose children do not have any absolute contraindications for vaccination.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Parental Vaccine Hesitancy Status	The Vaccine Hesitancy Scale is a measurement tool designed to assess individuals' vaccine hesitancy or opposition. It consists of 21 items across 4 subscales and uses a 5-point Likert scale format. The four subscales of the scale are: Vaccine Benefit and Protective Value, Vaccine Opposition, Solutions for Avoiding Vaccination, and Justification of Vaccine Hesitancy. Higher scores on the scale indicate greater levels of vaccine hesitancy or opposition.	Day 1 (Parents will be sent the questionnaire and asked to respond promptly.)

Collaborators and Investigators

• Sponsor: University of Yalova
• Investigators: Principal Investigator: EMEL AVÇİN, Assist Prof, University of Yalova

Study record dates

Study Major Dates

• Study Start (Actual): July 2, 2025
• Primary Completion (Actual): March 1, 2026
• Study Completion (Actual): June 1, 2026

Study Registration Dates

• First Submitted: May 17, 2025
• First Submitted That Met QC Criteria: May 23, 2025
• First Posted (Actual): May 25, 2025

Study Record Updates

• Last Update Posted (Actual): July 14, 2026
• Last Update Submitted That Met QC Criteria: July 12, 2026
• Last Verified: July 1, 2026

More Information

Terms related to this study

• Keywords: Children; Vaccines; Parent; Immunization; Machine learning; Hesitancy
• Additional Relevant MeSH Terms: Behavior; Treatment Adherence and Compliance; Health Behavior; Treatment Refusal; Vaccination Refusal; Vaccination Hesitancy
• Other Study ID Numbers: Emel Avcin MLT

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No