Donor behavior in healthcare crowdfunding success: A theory-driven approach using business intelligence and supervised machine learning in Pakistan

Document Type

Article

Publication Date

1-1-2026

Abstract

Medical crowdfunding has emerged as a crucial tool for bridging healthcare financing gaps in resource-constrained settings such as Pakistan. Despite its growing importance, the specific factors driving campaign success remain underexplored. This study adopts a theory-driven framework, grounded in signaling theory, to investigate the key determinants of successful campaigns on Transparent Hands, Pakistan's leading medical crowdfunding platform. Utilizing data from 1019 campaigns conducted between 2020 and 2023, we apply a multi-methodological approach. This includes statistical regression, business intelligence techniques, and a suite of supervised machine learning algorithms namely support vector regression, random forest, extreme gradient boosting, logistic regression, deep learning using convolutional neural networks, classification tree, and bootstrap aggregating ensemble. Our aim was to evaluate how campaign attributes, such as funds raised, patient demographics (gender and age), and type of medical procedure, influence donor engagement and the number of patients served. Results demonstrate that fundraising amount and patient age significantly shape donor behavior, while gender and procedure type affect both donor participation and service outreach. Significantly, the support vector regression and random forest models exhibited superior predictive accuracy, providing actionable insights for platform optimization. This study contributes substantively to the application of signaling theory in digital health financing and proposes data-driven strategies to enhance the effectiveness of health financing through medical crowdfunding.

Keywords

Business intelligence, Machine learning, Medical crowdfunding, Signaling theory

Publication Title

Engineering Applications of Artificial Intelligence

ISSN

09521976

DOI

10.1016/j.engappai.2025.113061

Volume

163

Share

COinS