A DELPHI EVALUATION OF THE FACTORS INFLUENCING LENGTH OF STAY IN AUSTRALIAN HOSPITALS JIANGUO XIAO 1 ,DAVID DOUGLAS 1 *, ANDY H. LEE 1 AND SIVA RAM VEMURI 2 1 Faculty of Science, Northern Territory University, Darwin 0909, NT, Australia 2 Faculty of Business, Northern Territory University, Darwin 0909, NT, Australia SUMMARY Using a modi®ed Delphi method, the factors in¯uencing length of inpatient stay (LOS) were explored. Row/column eects loglinear modelling was used to compare ratings between the ®rst and second rounds, and between the clinical and non-clinical groups. Rating scale modelling was used to classify and determine the relative importance of each factor. Six important and 48 signi®cant determinants of LOS were found, and four unimportant factors were identi®ed. The relationship among these factors and the implications of this study are explored. (&1997 by John Wiley & Sons, Ltd.) Int. J. Health Plann. Mgmt 12: 207±218, 1997 No. of Figures: 0, No. of Tables: 2, No. of References: 31 KEY WORDS: length of stay; in¯uencing factors; Delphi method INTRODUCTION Length of inpatient stay (LOS) is an important measure that is widely used in hospitals nowadays. It is one of the most useful hospital indicators as it re¯ects the eciency and performance of hospital activities and is therefore an essential component of hospital performance analysis. For example, LOS can be taken as a clinical outcome measure when comparing two or more dierent methods in various ®elds of surgery (Nuchtern et al., 1995; Engelman et al., 1994; Weingarten et al., 1994), obstetrics and gynaecology (Mazdisnian et al., 1995), paediatrics (Sutters et al., 1995) and emergency medicine (Sadaghdar et al., 1995). It is also used in the evaluation of nursing or pharmacy management (Redeker et al., 1994; Bjornson et al., 1993). However, despite its importance, there appears to be little understanding of the relative importance of the various factors in¯uencing LOS. One of the main reasons for this is the perceived tension between clinicians and administrators concerning various factors in¯uencing LOS. Therefore, there is not only a need to understand INTERNATIONAL JOURNAL OF HEALTH PLANNING AND MANAGEMENT, VOL. 12, 207±218 (1997) CCC 0749±6753/97/030207±12$17.50 &1997 by John Wiley & Sons, Ltd. *Correspondence to: David Douglas, Faculty of Science, Northern Territory University, Darwin 0909, NT, Australia.