Improving nested case-control studies to conduct a full competing-risks analysis for nosocomial infections

Area of Science:

• Epidemiology
• Biostatistics

Background:

• Competing risks are crucial when analyzing risk factors for nosocomial infections (NIs).
• Established methods for nested case-control designs require improvement for comprehensive risk analysis in hospital settings.

Purpose of the Study:

• To identify additional information provided by competing risks analysis in hospitals.
• To enhance established nested case-control methods for acquiring this information.

Main Methods:

• Utilized data from Spanish intensive care units and model simulations.
• Employed time-dynamic sampling for NIs to weight controls for risk-factor analysis of competing risks (death or discharge without infection).
• Extended methods for hazard rate and prediction analysis of NIs.

Main Results:

• Estimates from the extended method showed good agreement with full cohort data.
• The adapted nested case-control design successfully avoided competing risks bias.
• Reduced dataset results prevented common misinterpretations in competing-risks settings.

Conclusions:

• A nested case-control design can be adapted using routinely collected hospital data to avoid competing risks bias.
• This adapted method can reanalyze past studies to enhance their findings.
• The approach offers improved risk-factor analysis and prediction for nosocomial infections.