Abstract
Hospital Length of Stay (LoS) is a valid proxy to estimate the consumption of hospital resources. Average LoS, however, albeit easy to quantify and calculate, can be misleading if the underlying distribution is not symmetric. Therefore the average does not reflect the nature of such underlying distribution and may mask different effects. This paper uses routinely collected data of an Italian hospital patients from different departments over a period of 5 years. This will be the basis for a running example illustrating the alternative models of patients length of stay. The models includes a new density model called Hypergamma. The paper concludes by summarizing these various modelling techniques and highlighting the use of a risk measure in bed planning.
This is a preview of subscription content, access via your institution.
References
 1.
Artzner P, Delbaen F, Eber JM, Heath D (1999) Coherent measures of risk. Math. Finance 3:203228. Retrieved February 2011
 2.
Asmussen S, Nerman O, Olsson M (1996) Fitting phasetype distribution via EM algorithm. Scand J Statist 23:419–441
 3.
Bobbio A, Cumani (1992) Computer performance evaluation: 33–46
 4.
Cheridito P, Delbaen F, Kupper M (2005) Coherent and convex monetary risk measures for unbounded cádlág processes. Finance Stoch 9:1713–1732
 5.
Costa AX (2003) Mathematical modelling and simulation for planning critical care capacity. Anaesthesia 58:320327
 6.
Delbaen F (2002) Coherent measures of risk on general probability spaces. In: Advances in finance and stochastics. Essays in honour of dieter sondermann. SpringerVerlag
 7.
Fackrell M (2008) Modelling healthcare system with phasetype distributions Health Care Manage Sci
 8.
Faddy MJ, McClean SI (1999) Analysing data on length of stay of hospital patients using phasetype distribution. Appl Stoch Models Bus Ind 15:311–317
 9.
Fang KT, Yang Y (1994) Numbertheoretic methods in statistics. Chapman&Hall, London
 10.
Harper P, Shahani A (2002) Modelling for the planning and management of bed capacities in hospitals. J Opl Res Soc 53:1118
 11.
He Q, Zhang H (2005) A note on unicyclic representation of phasetype distributions. Stoch Models 21:465– 483
 12.
McClean SI, Millard PH (1993) Patterns of length of stay after admission in geriatric medicine: an event history approach. Stat 42:263–274
 13.
Millard PH (1994) Current measures and their defects. In: Modelling hospital resource use: a different approach to the planning and control of health care systems Millard P H, McClean S I (eds), Royal Society of Medicine, London
 14.
Neuts MF (1975) Probability distributions of phasetype. In: Liber amicorum Prof. Emeritus H. Florin, Department of Mathematics, University of Louvain, Louvain, pp 173206
 15.
Niederreiter H (1992) Random number generation and quasimonte carlo methods. SIAM, Philadelphia
 16.
O’Cinneide CA (1999) Phasetype distributions: open problems and few properties. Commun Stat Stoch Models 15:731–757
 17.
Ross SM Introduction to probability models, 5th edn. Academic, San Diego
 18.
Ross SM Stochastic process, Second Edition. Wiley
 19.
Vasilakis C, Marshall AH (2005) Modelling nationwide hospital length of stay: opening the black box. J Oper Res Soc 56:862869
Acknowledgments
This research is supported in part by the administrative division of the hospital Campus BioMedico.
Author information
Affiliations
Corresponding author
Appendix section
Appendix section
A function \(g:\mathbb {R}\rightarrow \mathbb {R}\) is a probability density function of a continuous random variable if

g is a nonnegative Lebesgueintegrable function;

\({\int }_{\infty }^{+\infty }\, g(x)\, dx=1\).
Then we have the following result
Proposition 1
Let g be a probability density function. For every α _{1} , α _{2} , …, α _{ n } ≥0 with α _{1} +α _{2} +⋯+α _{ n } =1, and ł _{1} , ł _{2} , …, ł _{ n } > 0, the function
is a probability density function.
Proof
It suffices to prove that \({\int }_{\mathbb {R}}\, h(\tau )\, d\tau =1\). In fact, we have
For any integral, in the sum, we apply the change of variable τ = −s/λ _{ i }. Thus
In the same way it is possible to evaluate the mean and the variance of a random variable distributed according to a Hypergamma density function:
Here the moment generating function is well defined on the interval t < 𝜃min(λ _{1}, λ _{2}, . . . , λ _{ n }).
Rights and permissions
About this article
Cite this article
Papi, M., Pontecorvi, L. & Setola, R. A new model for the length of stay of hospital patients. Health Care Manag Sci 19, 58–65 (2016). https://doi.org/10.1007/s1072901492889
Received:
Accepted:
Published:
Issue Date:
Keywords
 Length of stay
 Bed planning
 PH distribution
 Hospital resources
Mathematics Subject Classifications (2010)
 62H12
 60E60
 60J60