Improved hidden Markov model for nosocomial infections.

Area of Science:

• Epidemiology
• Biostatistics
• Mathematical Modeling

Background:

• Accurate parameter estimation is crucial for understanding and controlling disease transmission within hospitals.
• Standard hidden Markov models (HMMs) often rely on simplifying assumptions that may not reflect real-world hospital dynamics.
• Model misspecification can lead to flawed insights and ineffective intervention strategies.

Purpose of the Study:

• To propose a novel hidden Markov model (HMM) for parameter estimation in hospital transmission models.
• To evaluate the impact of commonly used simplifying assumptions in standard HMMs.
• To demonstrate the superiority of an alternative HMM in providing accurate parameter estimates.

Main Methods:

• Developed a novel HMM that avoids restrictive assumptions of fixed patient counts and binomially distributed detections.
• Compared the performance of the novel HMM against a standard HMM using simulated hospital transmission data.
• Analyzed parameter estimation accuracy and model misspecification under different scenarios.

Main Results:

• Common simplifying assumptions in standard HMMs lead to significant model misspecification.
• The proposed alternative HMM demonstrates robust performance, yielding accurate parameter estimates.
• Simulated data analysis confirmed the detrimental effects of standard model assumptions on estimation quality.

Conclusions:

• The novel HMM offers a more accurate approach to parameter estimation in hospital transmission modeling.
• Avoiding common simplifying assumptions is essential for reliable epidemiological modeling.
• This work provides a foundation for improved data analysis in infectious disease epidemiology.