A comparison of statistical methods for modeling count data with an application to hospital length of stay

Abstract

Background Hospital length of stay (LOS) is a key indicator of hospital care management efficiency, cost of care, and hospital planning. Hospital LOS is often used as a measure of a post-medical procedure outcome, as a guide to the benefit of a treatment of interest, or as an important risk factor for adverse events. Therefore, understanding hospital LOS variability is always an important healthcare focus. Hospital LOS data can be treated as count data, with discrete and non-negative values, typically right skewed, and often exhibiting excessive zeros. In this study, we compared the performance of the Poisson, negative binomial (NB), zero-inflated Poisson (ZIP), and zero-inflated negative binomial (ZINB) regression models using simulated and empirical data. Methods Data were generated under different simulation scenarios with varying sample sizes, proportions of zeros, and levels of overdispersion. Analysis of hospital LOS was conducted using empirical data from the Medical Information Mart for Intensive Care database. Results Results showed that Poisson and ZIP models performed poorly in overdispersed data. ZIP outperformed the rest of the regression models when the overdispersion is due to zero-inflation only. NB and ZINB regression models faced substantial convergence issues when incorrectly used to model equidispersed data. NB model provided the best fit in overdispersed data and outperformed the ZINB model in many simulation scenarios with combinations of zero-inflation and overdispersion, regardless of the sample size. In the empirical data analysis, we demonstrated that fitting incorrect models to overdispersed data leaded to incorrect regression coefficients estimates and overstated significance of some of the predictors. Conclusions Based on this study, we recommend to the researchers that they consider the ZIP models for count data with zero-inflation only and NB models for overdispersed data or data with combinations of zero-inflation and overdispersion. If the researcher believes there are two different data generating mechanisms producing zeros, then the ZINB regression model may provide greater flexibility when modeling the zero-inflation and overdispersion.

Association between fat‐free mass and survival in critically ill patients with COVID‐19: A prospective cohort study

“…Body composition changed significantly in our patients with COVID-19 during the acute and late phases of critical illness. Our study results are consistent with previous studies in general ICU patients 6,28 and show that critically ill patients with COVID-19 also have a large loss of FFM and skeletal muscle mass during their ICU stay.…”

“…The association between body composition parameters and the outcomes ICU stay, LOS, and duration of mechanical ventilation was analyzed using negative binomial regression analyses because this method performed favorably for this data. 28 Changes in body composition parameters (ie, FM, FFM, skeletal muscle mass, FM index, FFM index, skeletal muscle mass index, total body water, intercellular water, extracellular water, and phase angle) between the acute and late phase were used as continuous determinants. Analyses were additionally corrected for the following confounders: sex, age, SOFA score at baseline, presence of any comorbidity (yes/no), BMI, and the time between measurements (days).…”

Long-term exposure to air pollution and severe COVID-19 in Catalonia: a population-based cohort study

“…We assessed the proportional hazards assumption of our models by visual inspection of score residuals plotted against event time. We fitted negative binomial regression models to estimate the association between the 2019 annual average air pollution and hospital LOS among those individuals that were hospitalized 38 . Measures of association for air pollutants were reported as hazard ratios (HR) or incidence rate ratios (IRR) per interquartile range (IQR) increase, with their 95% confidence intervals (CI).…”

Characterising Hospital Admission Patterns and Length of Stay in the Emergency Department at Mater Dei Hospital Malta

“…There are some recent reviews of machine learning and statistical methods for the hospital length of stay estimation. [43,48,51,62,68]. Keegan [18] argued in favour of evidence showing that bed occupancy rate is a reliable key performance indicator for hospitals' capability to provide good quality care to patients.…”