Understanding inappropriate hospital admissions of patients presenting to the Emergency Department

Authors

  • Roberta Siliquini Dipartimento di Sanità Pubblica e Microbiologia, Università degli Studi di Torino, Torino, Italy
  • Nadia Rocca Azienda Ospedaliera San Giovanni Battista, Torino, Italy
  • Giuseppe Rocca Azienda Ospedaliera San Giovanni Battista, Torino, Italy
  • Davide Minniti Dipartimento di Sanità Pubblica e Microbiologia, Università degli Studi di Torino, Torino, Italy
  • Maria Michela Gianino Dipartimento di Sanità Pubblica e Microbiologia, Università degli Studi di Torino, Torino, Italy
  • Giovanni Renga Dipartimento di Sanità Pubblica e Microbiologia, Università degli Studi di Torino, Torino, Italy

DOI:

https://doi.org/10.2427/5982

Keywords:

appropriateness of care, concurrent review, indicators of inappropriate use, mergency Departments, costs

Abstract

Objectives. To identify 1) the characteristics of patients receiving non acute (inappropriate) care and 2) the
variables associated to inappropriate hospital use, in order to 3) estimate the relevance of the problem and
to 4) focus future concurrent reviews and efforts to allocate patients to alternative health care settings.
Design. A prospective review of a random sample of adult patients who presented to the Emergency
Department of the Molinette Hospital. Patients were assessed at admission and on day 3, 5 and 8 using the
Appropriateness Evaluation Protocol (Italian validated version). Patients: 490 overall; 312 (64 %) medical
and 178 (36 %) surgical.
Outcome measures. Acute (appropriate) and non acute (inappropriate) admissions, Major Disease Category,
costs, mean weights of Diagnosis Related Groups, and length of stay (days).
Results. The proportion of patients requiring acute care declined rapidly from presentation (84.5%) to the
fifth day of admission (60.9%). Patients admitted during weekends showed a higher rate of inappropriate
stay on day 5 (P=0.04). The proportion of inappropriate admissions was higher for medical rather than
surgical patients (P=0.07) at presentation and at day 5 (P < 0.01). Traditional social-demographic variables
were not significant risk indicators for inappropriate admissions. The likelihood ratio for inappropriate
admission at presentation was significantly higher for minor illnesses and disturbances (P=0.03).
Inappropriate stay on day 5 was significantly associated with lower cost (P < 0.01), lower mean DRG weight
(P < 0.01) and shorter length of stay (P=0.05) for medical but not for surgical admissions.
Conclusions. Traditional epidemiological indicators are inadequate to target prospective concurrent reviews.
Qualitative studies focusing on patient physician dialogue in different situations and contexts could widen
our understanding of the problem and suggest new theoretical frameworks and theories to provide us with
more detailed explanations.

Downloads

Published

2005-06-30

Issue

Vol. 2 No. 2 (2005)

Section

Long Paper