Home / Current Issue / Paper 1718571
Clustering of Individuals by Health Literacy and Risk Perception Patterns Using Unsupervised AI to Characterise Non-Prescription Medication Use and Care-Seeking Delays
Subject area: Science,Engineering and Technology · Area of research: Nursing
DOI: https://doi.org/10.64388/IREV9I12-1718571
Abstract
Non-prescription (over-the-counter) medication use is prevalent worldwide, yet inappropriate self medication contributes to delayed diagnosis, adverse drug events, and increased healthcare costs. Individual health literacy and perceived risk of self medication are key behavioral determinants, but their combined patterns remain poorly understood. This study applies unsupervised machine learning to cluster individuals based on their health literacy levels and risk perception profiles, and then characterizes each cluster’s non-prescription medication use and healthcare-seeking delays. A systematic review of 48 studies (2015–2025) examining health literacy, risk perception, self medication practices, and unsupervised clustering methods was conducted. Findings reveal four distinct clusters: (1) High literacy: Low risk perception (frequent inappropriate self medication, delayed care); (2) Low literacy: High risk perception (avoidant behaviour, unnecessary delays, and anxiety); (3) Moderate literacy: Balanced risk perception (appropriate short term use, timely care seeking); and (4) Low literacy: Low risk perception (high risk overuse, very long delays, and frequent adverse outcomes). Unsupervised AI techniques, particularly k-means, hierarchical clustering, and latent profile analysis, effectively identify these subgroups. The study demonstrates that clustering by behavioral phenotypes enables targeted interventions, such as literacy sensitive education, pharmacy based screening, and digital nudges, to reduce inappropriate self medication and delays in care. AI driven clustering offers a scalable approach for public health stratification and personalized risk communication.
Keywords
Health Literacy, Risk Perception, Non Prescription Medication, Self Medication, Care Seeking Delay, Unsupervised AI, Clustering, K Means, Latent Profile Analysis.
How to cite this paper
@article{1718571,
author = {Oluwadare Ibukun Jumoke, Cecilia Bukola Bello},
title = {Clustering of Individuals by Health Literacy and Risk Perception Patterns Using Unsupervised AI to Characterise Non-Prescription Medication Use and Care-Seeking Delays},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {9},
number = {12},
pages = {160-165},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1718571.pdf},
abstract = {Non-prescription (over-the-counter) medication use is prevalent worldwide, yet inappropriate self medication contributes to delayed diagnosis, adverse drug events, and increased healthcare costs. Individual health literacy and perceived risk of self medication are key behavioral determinants, but their combined patterns remain poorly understood. This study applies unsupervised machine learning to cluster individuals based on their health literacy levels and risk perception profiles, and then characterizes each cluster’s non-prescription medication use and healthcare-seeking delays. A systematic review of 48 studies (2015–2025) examining health literacy, risk perception, self medication practices, and unsupervised clustering methods was conducted. Findings reveal four distinct clusters: (1) High literacy: Low risk perception (frequent inappropriate self medication, delayed care); (2) Low literacy: High risk perception (avoidant behaviour, unnecessary delays, and anxiety); (3) Moderate literacy: Balanced risk perception (appropriate short term use, timely care seeking); and (4) Low literacy: Low risk perception (high risk overuse, very long delays, and frequent adverse outcomes). Unsupervised AI techniques, particularly k-means, hierarchical clustering, and latent profile analysis, effectively identify these subgroups. The study demonstrates that clustering by behavioral phenotypes enables targeted interventions, such as literacy sensitive education, pharmacy based screening, and digital nudges, to reduce inappropriate self medication and delays in care. AI driven clustering offers a scalable approach for public health stratification and personalized risk communication.},
keywords = {Health Literacy, Risk Perception, Non Prescription Medication, Self Medication, Care Seeking Delay, Unsupervised AI, Clustering, K Means, Latent Profile Analysis.},
month = {June},
doi = {https://doi.org/10.64388/IREV9I12-1718571}
}