RSNA 2007 

Abstract Archives of the RSNA, 2007


SSJ04-02

Patients Imaged Promptly during Admission to an Urban Medical Center Have Reduced Length of Hospital Stay

Scientific Papers

Presented on November 27, 2007
Presented as part of SSJ04: Health Services, Policy, and Research (EBM, Guidelines, and Outcomes)

 Trainee Research Prize - Resident

Participants

Juan Carlos Batlle MD, Presenter: Nothing to Disclose
Peter F. Hahn MD, PhD, Abstract Co-Author: Nothing to Disclose
Bert-Jan Nagtegaal, Abstract Co-Author: Nothing to Disclose
James Hunter Thrall MD, Abstract Co-Author: Medical Advisory Board, Bracco Group Board of Directors, E-Z-EM, Inc
Susanna In-Sun Lee MD, PhD, Abstract Co-Author: Nothing to Disclose

PURPOSE

To relate prompt use of advanced imaging (CT, MRI and nuclear scintigraphy) to length of hospital stay.

METHOD AND MATERIALS

We cross-matched all 33,226 admissions to an urban tertiary care hospital in 2005 with all 584,901 radiological studies performed in 2005. 10,005 (30.1%) admissions included at least one advanced imaging study during the period from one day prior to admission (day -1) through discharge. We identified the date of service (DOS) of the first advanced imaging study performed, and calculated the length of stay (LOS) and residual length of stay (LOS minus DOS) of each admission. LOS and residual LOS were correlated with DOS by linear regression. For admissions of at least 3 days' duration, mean LOS of patients with early imaging (day -1 or 0) was compared by t-test to those with later imaging (day 1 or 2).

RESULTS

Linear correlation of LOS vs. DOS demonstrates a slope of 1.27 (p<0.001). Linear correlation of residual LOS vs. DOS demonstrates a slope = 0.27 (p<0.001). Mean LOS was significantly shorter for those imaged on day -1 or 0 vs. day 1 or 2 for all admissions (8.6 vs. 9.0 days, p=0.015), and for the following specific ICD9 subgroups: all admissions excluding cardiovascular (CV) ICD9 codes (7.8 vs. 8.9 days, p<0.001), all admissions excluding CV and neurological ICD9 codes (8.5 vs. 9.3 days, p=0.004), all GI and GU ICD9 codes (7.5 vs. 8.8 days, p=0.007), and all acute and subacute GI and GU ICD9 codes (7.5 vs. 8.7 days, p=0.014). Mean LOS was significantly shorter for the following specific modality subgroups: all those first imaged with abdominopelvic CT (8.4 vs. 9.7, p=0.003), with any CT (8.8 vs. 9.6 days, p=0.001), with neurological MRI (7.6 vs. 8.7 days, p=0.030), and with any MRI (7.8 vs. 9.0 days, p=0.008).

CONCLUSION

Prompt imaging with CT or MRI, particularly on the day before or the day of admission, significantly shortens length of stay of inpatients. This applies to all patients as a whole and particularly to those with GI or GU diagnoses and those requiring neurological MRI.

CLINICAL RELEVANCE/APPLICATION

Prompt advanced imaging is associated with shorter hospital stays; this suggests clinical practice algorithms can be developed to lower the overall cost of inpatient care.

Cite This Abstract

Batlle, J, Hahn, P, Nagtegaal, B, Thrall, J, Lee, S, Patients Imaged Promptly during Admission to an Urban Medical Center Have Reduced Length of Hospital Stay.  Radiological Society of North America 2007 Scientific Assembly and Annual Meeting, November 25 - November 30, 2007 ,Chicago IL. http://archive.rsna.org/2007/5013515.html