How to cite item

Addressing missing data in health research: a narrative review of mechanisms, methods, and implications for healthcare quality and policy

   
@article{JHMHP10527,
	author = {Majed Al-Turbag},
	title = {Addressing missing data in health research: a narrative review of mechanisms, methods, and implications for healthcare quality and policy},
	journal = {Journal of Hospital Management and Health Policy},
	volume = {10},
	number = {0},
	year = {2026},
	keywords = {},
	abstract = {Background and Objective: Missing data are pervasive in healthcare research and routinely affect hospital performance indicators, patient safety metrics, clinical outcomes, and health policy evaluations. Despite extensive methodological literature, applied healthcare studies continue to rely on suboptimal or poorly reported approaches for handling missing data. This narrative review aims to synthesise missing data mechanisms and statistical handling methods through a healthcare systems and policy lens, highlighting their implications for hospital management and decision-making.Methods: A narrative review was conducted using PubMed, Scopus, and Web of Science to identify English-language literature on missing data mechanisms, prevention strategies, and analytical methods relevant to health research, hospital datasets, and clinical studies. Key methodological papers, reviews, and applied healthcare studies were synthesised narratively.Key Content and Findings: Rubin’s framework of missing completely at random (MCAR), missing at random (MAR), and missing not at random (MNAR) remains foundational for guiding analytical decisions. In healthcare settings, missing data frequently arise from patient non-response, loss to follow-up, electronic health record incompleteness, and administrative data linkage failures. Simple approaches such as case deletion and single imputation remain common but can distort estimates, reduce statistical power, and misinform quality assessments. Likelihood-based methods and multiple imputation (MI) generally provide more valid inference under MAR assumptions, while MNAR scenarios require explicit modelling or sensitivity analyses using pattern-mixture or selection models.Conclusions: Effective management of missing data in health research requires integration of prevention strategies with analytically appropriate methods aligned to the assumed missingness mechanism. Transparent reporting and sensitivity analysis are essential to support reliable hospital management decisions and health policy formulation.},
	issn = {2523-2533},	url = {https://jhmhp.amegroups.org/article/view/10527}
}