Exploring models for the length of stay distribution

Theory and Practice of Mathematics and Computer Science , 2021

It is important to fit count data with suitable model(s), models such as Poisson Regression, Quassi Poisson, Negative Binomial, to mention but a few have been adopted by researchers to fit zero truncated count data in the past. In recent times, dedicated models for fitting zero truncated count data have been developed, and they are considered sufficient. This study proposed Bayesian multi-level Poisson and Bayesian multi-level Geometric model, Bayesian Monte Carlo Markov Chain Generalized linear Mixed Models (MCMCglmms) of zero truncated Poisson and MCMCglmms Poisson regression model to fit health count data that is truncated at zero. Data of visits to visit to the doctor of patients under National Health Insurance Scheme in Nigeria was obtained and used to fit the models. Suitable model selection criteria were used to determine preferred models for fitting zero truncated data. Results obtained showed that Bayesian multi-level Poisson outperformed Bayesian multi-level Poisson Geometric model; also MCMCglmms of zero truncated Poisson outperformed MCMCglmms Poisson.

BMC Medical Research Methodology

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 Informat...

Working Papers, 2003

I develop a new approach for sample selection problems that allows parametric forms of any type to be chosen for both for the selection and the observed variables. The Generalized Parametric Selection (GPS) model can incorporate both duration and count data models, unlike previous parametric models. MLE does not require numerical integration or simulation techniques, unlike previous models for count data. I discuss application to common duration models (exponential, Weibull, log-logistic) and count models (Poisson, negative binomial). I demonstrate the usefulness of the model with an application to the effects of insurance status and managed care on hospitalization duration data. The example indicates that the GPS model may be preferred even in cases for which other parametric approaches are available.

AStA Advances in Statistical Analysis, 2008

The least trimmed squares (LTS) estimator and the trimmed mean are two well known trimming based estimators. Both estimates are popular in practice, and they are implemented in standard statistical softwares. In this paper, we compare the asymptotic variances of these two estimators in a location model, when the two location estimates have the highest breakdown point (i.e., 50%). Some interesting results concerning the asymptotic variances of the two types of estimates are obtained for distributions with exponential and polynomial tails. We extend this comparison into regression problems. Some simulation studies are performed to compare the finite sample standard errors of these estimators in location and regression models. Two examples involving real data sets are presented to compare the bootstrap standard errors of the LTS estimator and the analogue of the trimmed mean in regression problems.

Australian & New Zealand Journal of Statistics, 2017

Statistical process monitoring (SPM) has been used extensively recently in order to assure the quality of the output of industrial processes. Techniques of SPM have been efficiently applied during the last two decades in non-industrial processes. A field of application with great interest is public health monitoring, where a pitfall with which we have to deal is the fact that available samples are not random in all cases. In the majority of cases, we monitor measurements derived from patient admissions to a hospital against control limits that were calculated using a sample of data taken from an epidemiological survey. In this work, we bridge the gap of a change in the sampling scheme from Phase I to Phase II, studying the case where the sampling during Phase II is biased. We present the appropriate methodology and then apply extensive numerical simulation in order to explore the performance of the proposed methodology, for measurements following various asymmetrical distributions. As the simulations show, the proposed methodology has a significantly better performance than the standard procedure.

Statistics in Medicine, 2008

Length of hospital stay (LOS) is an important indicator of the hospital activity and management of health care. The skewness exhibited by this variable poses problems in statistical modeling. The aim of this work is to model the variable LOS within diagnosis-related groups (DRG) through finite mixtures of distributions. A mixture of the union of Gamma, Weibull and Lognormal families is used in the model, instead of a mixture of a unique distribution family. Some theoretical questions regarding the model, such as the identifiability and study of asymptotic properties of ML estimators, are analyzed. The EM algorithm is proposed for performing these estimators. Finally, this new proposed model is illustrated by using data from different DRGs.

2019

It is important to fit count data with suitable model(s), models such as Poisson Regression, Quassi Poisson, Negative Binomial, to mention but a few have been adopted by researchers to fit zero truncated count data in the past. In recent times, dedicated models for fitting zero truncated count data have been developed, and they are considered sufficient. This study proposed Bayesian multi-level Poisson and Bayesian multi-level Geometric model, Bayesian Monte Carlo Markov Chain Generalized linear Mixed Models (MCMCglmms) of zero truncated Poisson and MCMCglmms Poisson regression model to fit health count data that is truncated at zero. Suitable model selection criteria were used to determine preferred models for fitting zero truncated data. Results obtained showed that Bayesian multi-level Poisson outperformed Bayesian multi-level Poisson Geometric model; also MCMCglmms of zero truncated Poisson outperformed MCMCglmms Poisson.

Journal of Advances in Mathematics, 2021

The aim of this study is to investigate the effect of different truncation combinations on the estimation of the normal distribution parameters. In addition, is to study methods used to estimate these parameters, including MLE, moments, and L-moment methods. On the other hand, the study discusses methods to estimate the mean and variance of the truncated normal distribution, which includes sampling from normal distribution, sampling from truncated normal distribution and censored sampling from normal distribution. We compare these methods based on the mean square errors, and the amount of bias. It turns out that the MLE method is the best method to estimate the mean and variance in most cases and the L-moment method has a performance in some cases.

Annals of the Institute of Statistical Mathematics, 1989

A sequential procedure is proposed for constructing a fixed-size confidence region for the parameters of a linear regression model. The procedure is based on certain regression analogues of trimmed means, as formulated by Welsh (1987,Ann. Statist.,15, 20–36), rather than least squares estimates. For error distributions with continuous, symmetric density and some moment higher than fourth finite, if the design points

2012

Distributions of healthcare utilization such as hospital length of stay and inpatient cost are generally right skewed. The extremes represent legitimate observations on patients who, because of the severity of their illness and need for medical intervention, have long in-stays and incur large costs. In this context we demonstrate the application of several parametric models for fitting heavy tailed data. Both maximum likelihood and Bayesian methods are used for estimation in certain Coxian phase-type models, mixtures of exponential distributions, and for comparison, the lognormal, loglogistic, Weibull, generalized gamma and generalized Pareto —including the standard Pareto and Burr distributions. We focus on the mean and percentiles of the response, and illustrate our methods with an empirical example on fitting models to hospital stays for acute myocardial infarction in the Nationwide Inpatient Sample of the Healthcare Utilization Project. A suite of SAS procedures is used in all ...