Concept Architecture
Concept
Theoretically, Geometric Mean Length of Stay (Geometric Mean LOS) is a statistical measure that summarises the central tendency of hospital length of stay using the geometric mean rather than the arithmetic mean. It is particularly appropriate for positively skewed length-of-stay distributions, where a small number of patients with exceptionally long admissions can substantially inflate the arithmetic mean. The measure is founded on the theory of geometric averages and log-transformed data and is widely used in hospital reimbursement systems and health services research.
Mathematically, the Geometric Mean LOS is calculated as the nth root of the product of all individual lengths of stay or, equivalently, as the exponential of the arithmetic mean of the natural logarithms of the observed lengths of stay. This approach reduces the influence of extreme values while preserving the multiplicative characteristics of skewed healthcare utilisation data.
In practice, Geometric Mean LOS is used in Diagnosis-Related Group (DRG) payment systems, hospital benchmarking and health economic analyses. It provides a stable estimate of expected resource utilisation for patient groups, supports reimbursement calculations and facilitates comparisons of hospital efficiency across providers and healthcare systems.
Purpose
Used to summarise typical hospital length of stay for skewed data, support DRG reimbursement, benchmark healthcare providers and inform health economic analyses of inpatient resource utilisation.
Mathematical Formulae
Primary Formula
Geometric Mean LOS = (x? ? x? ? ? ? x?)^(1/n)
Equivalent logarithmic form
Geometric Mean LOS = exp[(�ln(x?)) � n]
where:
- x? = individual length of stay
- n = number of observations
Supporting Formulae
Arithmetic Mean LOS
Mean LOS = �x? � n
Length of Stay
LOS = Discharge Date ? Admission Date
Related Mathematical Methods
- Geometric mean
- Logarithmic transformation
- Arithmetic mean
- Median
- Length of stay analysis
- Diagnosis-Related Group analysis
Example
Five patients have hospital stays of:
2, 3, 4, 5 and 20 days.
Geometric Mean LOS
= (2 ? 3 ? 4 ? 5 ? 20)^(1/5)
= 1,200^(1/5)
� 4.13 days
Although the arithmetic mean is 6.8 days, the Geometric Mean LOS of 4.13 days better reflects the typical stay because it is less influenced by the single prolonged admission.
Excel Implementation
| Function | Example Formula | Health Economics Application |
|---|---|---|
| GEOMEAN | =GEOMEAN(A2:A101) | Calculate the Geometric Mean LOS. |
| AVERAGE | =AVERAGE(A2:A101) | Compare arithmetic and geometric mean LOS. |
| MEDIAN | =MEDIAN(A2:A101) | Compare geometric mean with the median LOS. |
| LN | =LN(A2) | Calculate logarithms for manual geometric mean calculations. |
| EXP | =EXP(AVERAGE(B2:B101)) | Calculate the geometric mean from log-transformed LOS values. |
VBA (Optional)
Automate calculation of Geometric Mean LOS by DRG, hospital, specialty and reporting period while generating reimbursement reports, benchmarking dashboards and healthcare utilisation analyses.
Sources
- Centers for Medicare & Medicaid Services. MS-DRG Definitions Manual.
- Fetter RB, Shin Y, Freeman JL, Averill RF, Thompson JD. Case Mix Definition by Diagnosis-Related Groups.
- OECD. Health at a Glance.
- Hosmer DW, Lemeshow S, Sturdivant RX. Applied Logistic Regression.
- Drummond MF, Sculpher MJ, Claxton K, Stoddart GL, Torrance GW. Methods for the Economic Evaluation of Health Care Programmes.
- Briggs A, Claxton K, Sculpher M. Decision Modelling for Health Economic Evaluation.
Related Concepts (2)
Library
Publications
1
Productivity Growth in the English National Health Service from 1998/1999 to 2013/2014 — Bojke, Castelli, Grašič, Howdon & Street, Vol. 26, No. 5 ed., 2017 (Health Economics)
The York Centre for Health Economics measurement of NHS productivity growth as a chained index of outputs over inputs across 15 years, the standard methodological reference for English NHS productivity analysis.
Journal ArticleView source →
Frequently Asked Questions (6)
What is geometric mean length of stay?
A statistical measure of typical diagnosis-related group length of stay, using geometric rather than arithmetic mean to reduce outlier influence.
Source: Fetter et al. 1980
What typical length of stay does geometric mean length of stay measure?
Geometric mean length of stay is a statistical measure of the typical length of stay for a diagnosis-related group. It measures the usual stay for the group, giving a benchmark for how long such patients tend to remain. It uses the geometric mean rather than the arithmetic mean to reduce the influence of outliers, so a few very long stays do not distort the figure. It applies to a diagnosis-related group, the category of clinically similar cases. It underlies excess days, the count of days a patient stays beyond this expected length. A robust benchmark stay for a case group is what it measures. Fetter and colleagues (1980) set this out.
Source: Fetter et al. 1980
What does geometric mean length of stay measure?
Geometric mean length of stay measures the typical diagnosis-related group length of stay, so it gauges the typical length of stay for a diagnosis-related group, using geometric rather than arithmetic mean to reduce outlier influence. This measurement of typical length of stay defines it. So geometric mean length of stay is a statistical measure of typical diagnosis-related group length of stay, using geometric rather than arithmetic mean to reduce outlier influence.
Source: Fetter et al. 1980
Why does geometric mean length of stay use the geometric mean?
Geometric mean length of stay uses the geometric rather than arithmetic mean to reduce outlier influence, so it lessens the effect of unusually long or short stays on the typical diagnosis-related group length of stay. This reduction of outlier influence defines why it is used. So geometric mean length of stay is a statistical measure of typical diagnosis-related group length of stay, using geometric rather than arithmetic mean to reduce outlier influence.
Source: Fetter et al. 1980
What does geometric mean length of stay apply to?
Geometric mean length of stay applies to a diagnosis-related group, so it is a statistical measure of the typical length of stay for such a group, using geometric rather than arithmetic mean to reduce outlier influence. This tie to a diagnosis-related group defines it. So geometric mean length of stay is a statistical measure of typical diagnosis-related group length of stay, using geometric rather than arithmetic mean to reduce outlier influence.
Source: Fetter et al. 1980
How does geometric mean length of stay relate to excess days?
Geometric mean length of stay relates to excess days as the expected stay against which excess is measured: geometric mean length of stay is a statistical measure of typical diagnosis-related group length of stay, and excess days are hospital days a patient spends beyond what would typically be expected for their diagnosis. So geometric mean length of stay provides the typical stay excess days are counted beyond, connected as the expected length and the days over it.
Source: Fetter et al. 1980
Trust Record
Verified by Dr Darrin Baines
British health economist
Professional identity: darrinbaines.org
Verification date: 30 Mar 2026
Content version: 1.0.0
Canonical Identity
- Persistent URI
- https://healtheconomics.wiki/concept/geometric-mean-los
- Term code
- HS-NHS-PM-010
Stable URI · Machine-readable · Resolvable · CC BY 4.0