Health Belief Model Guided Mixed Methods Study of Caregivers Handling Pediatric Oral Liquid Medications in Malaysia.

INTRODUCTION/OBJECTIVE: At-home pediatric oral liquid medication errors are common, preventable patient-safety risks, yet remain incompletely characterized in Malaysia due to limited theory-driven mixed-methods research and rural-urban comparisons. This study examined caregivers' practices, knowledge gaps, and behavioral determinants related to safe and unsafe handling of pediatric oral liquid medications. METHODS: A sequential exploratory mixed-methods study was conducted among rural and urban caregivers in Malaysia. Focus Group Discussions (FGDs; n=40) explored caregiver perspectives; these findings informed the development of a validated cross-sectional survey (n=377) that classified knowledge and perception as good or poor using a prespecified 75% threshold. Theory-driven triangulation identified convergence, divergence, and contextual elaboration to develop a Health Belief Model-informed framework. RESULTS: Six FGD themes (dosing, administration, storage, reconstitution, disposal, caregiver expectations) revealed safety gaps and practical trade-offs. Knowledge and perception were significantly associated (p < 0.001, Cramer's V = 0.185). Tertiary education (vs primary/secondary education) (AOR = 3.68) and older age (AORs = 2.38-5.82) predicted better knowledge. Caring for ≥2 children was associated with better perception (AOR = 1.58), whereas Chinese caregivers had lower odds of good perception (AOR = 0.55, 95% CI: 0.31-0.98). Unsafe practices involved incorrect dosing and administration, inappropriate medication-use behaviors, inconsistent reconstitution, and improper storage and disposal. DISCUSSION: Higher knowledge was associated with better perceptions but did not consistently predict safer self-reported practices. Low confidence, shortcuts, limited prompts, and structural barriers shaped behaviors. These differences may reflect variations in educational exposure, caregiving experience, health literacy, and access to information. Geographic concentration in northern Malaysia and non-probability sampling limited generalizability. CONCLUSION: Tailored interventions incorporating dosing aids, management algorithms, visual tools, and low-cost digital reminders may improve safety; these strategies warrant formal evaluation.