Length of stay

Length of stay (LOS) is a term to describe the duration of a single episode of hospitalization. Inpatient days are calculated by subtracting day of admission from day of discharge.

Analysis

A common statistic associated with length of stay is the average length of stay (ALOS), a mean calculated by dividing the sum of inpatient days by the number of patients admissions with the same diagnosis-related group classification. A variation in the calculation of ALOS can be to consider only length of stay during the period under analysis.

Length of stay is typically highly skewed, so statistical approaches taking that into account are recommended. While the mean length of stay is useful from the point of view of costs, it may be a poor statistic in terms of representing a typical length of stay; the median may be preferred.

It is useful to be able to predict an individual's expected length of stay or to model length of stay to determine factors that affect it. Various analyses have sought to model length of stay in different condition contexts. This has usually been done with regression models, but Markov chain methods have also been applied. Within regression approaches, linear, log-normal and logistic regression approaches have been applied, but have been criticised by other researchers. Carter & Potts (2014) instead recommend use of negative binomial regression.