Intensive care unit admission with community-acquired pneumonia

Adam S. Vohra, Hyo Jung Tak, Maulin B. Shah, David O. Meltzer, Gregory W. Ruhnke

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Background: There has been a dramatic increase in the use of intensive care units (ICUs) over the past 25 years. Greater use of validated measures of illness severity may better inform ICU admission decisions in patients with community-acquired pneumonia. This article examined predictors of ICU admission and hospitalization costs, including the pneumonia severity index (PSI) and CURB-65 (confusion, uremia, respiratory rate, blood pressure, age $65 years) scores. Methods: The study identified 422 patients hospitalized for communityacquired pneumonia, ascertaining patient characteristics by chart review and extraction of administrative data. Multivariate logistic regression was performed to quantify the association of the PSI, CURB-65 and comorbidities with ICU admission. The predictors of cost were estimated using a generalized linear model. Results: Compared to 194 general medicine patients, certain clinical and radiographic findings were more common among 228 ICU patients. Compared to PSI reference group I/II/III, ICU admission was strongly associated with risk class IV (odds ratio [OR], 3.06; 95% confidence interval [CI], 1.63- 5.72) and V (OR, 4.84; CI, 2.44-9.62), and also CURB-65 3 (OR, 2.90; CI, 1.51-5.56). The relative increase in mortality among PSI risk class V (compared to IV) patients was 2.68 times higher in general medicine, compared with the ICU. Among ICU admissions, risk class V was associated with an additional cost of 14,548 (95% CI, 4,232 to 24,864). Conclusions: Illness severity and chronic pulmonary disease are strong predictors of ICU admission. More extensive use of the PSI may optimize site-of-care decisions, thereby minimizing mortality and unnecessary resource utilization.

Original languageEnglish (US)
Pages (from-to)380-386
Number of pages7
JournalAmerican Journal of the Medical Sciences
Volume350
Issue number5
DOIs
StatePublished - Jan 1 2015

Fingerprint

Intensive Care Units
Pneumonia
Confidence Intervals
Odds Ratio
Costs and Cost Analysis
Medicine
Mortality
Uremia
Respiratory Rate
Lung Diseases
Comorbidity
Linear Models
Hospitalization
Chronic Disease
Logistic Models
Blood Pressure

Keywords

  • Community-acquired pneumonia
  • Hospitalization costs
  • Intensive care unit
  • Pneumonia severity index
  • Severity of illness

ASJC Scopus subject areas

  • Medicine(all)

Cite this

Intensive care unit admission with community-acquired pneumonia. / Vohra, Adam S.; Tak, Hyo Jung; Shah, Maulin B.; Meltzer, David O.; Ruhnke, Gregory W.

In: American Journal of the Medical Sciences, Vol. 350, No. 5, 01.01.2015, p. 380-386.

Research output: Contribution to journalArticle

Vohra, Adam S. ; Tak, Hyo Jung ; Shah, Maulin B. ; Meltzer, David O. ; Ruhnke, Gregory W. / Intensive care unit admission with community-acquired pneumonia. In: American Journal of the Medical Sciences. 2015 ; Vol. 350, No. 5. pp. 380-386.
@article{90f0db94254a4e37b2ab4536e4a7365f,
title = "Intensive care unit admission with community-acquired pneumonia",
abstract = "Background: There has been a dramatic increase in the use of intensive care units (ICUs) over the past 25 years. Greater use of validated measures of illness severity may better inform ICU admission decisions in patients with community-acquired pneumonia. This article examined predictors of ICU admission and hospitalization costs, including the pneumonia severity index (PSI) and CURB-65 (confusion, uremia, respiratory rate, blood pressure, age $65 years) scores. Methods: The study identified 422 patients hospitalized for communityacquired pneumonia, ascertaining patient characteristics by chart review and extraction of administrative data. Multivariate logistic regression was performed to quantify the association of the PSI, CURB-65 and comorbidities with ICU admission. The predictors of cost were estimated using a generalized linear model. Results: Compared to 194 general medicine patients, certain clinical and radiographic findings were more common among 228 ICU patients. Compared to PSI reference group I/II/III, ICU admission was strongly associated with risk class IV (odds ratio [OR], 3.06; 95{\%} confidence interval [CI], 1.63- 5.72) and V (OR, 4.84; CI, 2.44-9.62), and also CURB-65 3 (OR, 2.90; CI, 1.51-5.56). The relative increase in mortality among PSI risk class V (compared to IV) patients was 2.68 times higher in general medicine, compared with the ICU. Among ICU admissions, risk class V was associated with an additional cost of 14,548 (95{\%} CI, 4,232 to 24,864). Conclusions: Illness severity and chronic pulmonary disease are strong predictors of ICU admission. More extensive use of the PSI may optimize site-of-care decisions, thereby minimizing mortality and unnecessary resource utilization.",
keywords = "Community-acquired pneumonia, Hospitalization costs, Intensive care unit, Pneumonia severity index, Severity of illness",
author = "Vohra, {Adam S.} and Tak, {Hyo Jung} and Shah, {Maulin B.} and Meltzer, {David O.} and Ruhnke, {Gregory W.}",
year = "2015",
month = "1",
day = "1",
doi = "10.1097/MAJ.0000000000000568",
language = "English (US)",
volume = "350",
pages = "380--386",
journal = "American Journal of the Medical Sciences",
issn = "0002-9629",
publisher = "Lippincott Williams and Wilkins",
number = "5",

}

TY - JOUR

T1 - Intensive care unit admission with community-acquired pneumonia

AU - Vohra, Adam S.

AU - Tak, Hyo Jung

AU - Shah, Maulin B.

AU - Meltzer, David O.

AU - Ruhnke, Gregory W.

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Background: There has been a dramatic increase in the use of intensive care units (ICUs) over the past 25 years. Greater use of validated measures of illness severity may better inform ICU admission decisions in patients with community-acquired pneumonia. This article examined predictors of ICU admission and hospitalization costs, including the pneumonia severity index (PSI) and CURB-65 (confusion, uremia, respiratory rate, blood pressure, age $65 years) scores. Methods: The study identified 422 patients hospitalized for communityacquired pneumonia, ascertaining patient characteristics by chart review and extraction of administrative data. Multivariate logistic regression was performed to quantify the association of the PSI, CURB-65 and comorbidities with ICU admission. The predictors of cost were estimated using a generalized linear model. Results: Compared to 194 general medicine patients, certain clinical and radiographic findings were more common among 228 ICU patients. Compared to PSI reference group I/II/III, ICU admission was strongly associated with risk class IV (odds ratio [OR], 3.06; 95% confidence interval [CI], 1.63- 5.72) and V (OR, 4.84; CI, 2.44-9.62), and also CURB-65 3 (OR, 2.90; CI, 1.51-5.56). The relative increase in mortality among PSI risk class V (compared to IV) patients was 2.68 times higher in general medicine, compared with the ICU. Among ICU admissions, risk class V was associated with an additional cost of 14,548 (95% CI, 4,232 to 24,864). Conclusions: Illness severity and chronic pulmonary disease are strong predictors of ICU admission. More extensive use of the PSI may optimize site-of-care decisions, thereby minimizing mortality and unnecessary resource utilization.

AB - Background: There has been a dramatic increase in the use of intensive care units (ICUs) over the past 25 years. Greater use of validated measures of illness severity may better inform ICU admission decisions in patients with community-acquired pneumonia. This article examined predictors of ICU admission and hospitalization costs, including the pneumonia severity index (PSI) and CURB-65 (confusion, uremia, respiratory rate, blood pressure, age $65 years) scores. Methods: The study identified 422 patients hospitalized for communityacquired pneumonia, ascertaining patient characteristics by chart review and extraction of administrative data. Multivariate logistic regression was performed to quantify the association of the PSI, CURB-65 and comorbidities with ICU admission. The predictors of cost were estimated using a generalized linear model. Results: Compared to 194 general medicine patients, certain clinical and radiographic findings were more common among 228 ICU patients. Compared to PSI reference group I/II/III, ICU admission was strongly associated with risk class IV (odds ratio [OR], 3.06; 95% confidence interval [CI], 1.63- 5.72) and V (OR, 4.84; CI, 2.44-9.62), and also CURB-65 3 (OR, 2.90; CI, 1.51-5.56). The relative increase in mortality among PSI risk class V (compared to IV) patients was 2.68 times higher in general medicine, compared with the ICU. Among ICU admissions, risk class V was associated with an additional cost of 14,548 (95% CI, 4,232 to 24,864). Conclusions: Illness severity and chronic pulmonary disease are strong predictors of ICU admission. More extensive use of the PSI may optimize site-of-care decisions, thereby minimizing mortality and unnecessary resource utilization.

KW - Community-acquired pneumonia

KW - Hospitalization costs

KW - Intensive care unit

KW - Pneumonia severity index

KW - Severity of illness

UR - http://www.scopus.com/inward/record.url?scp=84946543145&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84946543145&partnerID=8YFLogxK

U2 - 10.1097/MAJ.0000000000000568

DO - 10.1097/MAJ.0000000000000568

M3 - Article

VL - 350

SP - 380

EP - 386

JO - American Journal of the Medical Sciences

JF - American Journal of the Medical Sciences

SN - 0002-9629

IS - 5

ER -