Methods for estimating costs of healthcare-associated Infections

Abstract

Introduction: Healthcare-associated infections (HCAIs) cause significant clinical and economic burdens worldwide, making accurate cost estimation critical for effective prevention and resource allocation. Several methodologies exist to estimate the attributable costs and hospital length of stay due to HCAIs, each with its own strengths and limitations. Case studies are simple and inexpensive but subjective and less reliable, while unmatched case-control studies provide quick estimates but may overestimate costs by not controlling for confounders. Matched case-control studies offer improved accuracy by pairing infected patients with similar controls, although they face challenges related to matching complexity and time-dependent biases. Regression analysis allows for multivariable adjustment and reduces selection bias by including nearly all patients, but it requires detailed clinical data and expert analysis. Multistate models provide the most comprehensive insight by modeling patient transitions between health states over time, effectively addressing time-dependent bias, yet they require extensive data collection and complex modeling efforts. In conclusion, the choice of methodology depends on research objectives, data availability, and desired precision, and often a combination of approaches delivers the most accurate and actionable information for healthcare policy and infection control strategies.