0
Correspondence |

Response FREE TO VIEW

Bekele Afessa, MD, FCCP; Ognjen Gajic, MD, FCCP
Author and Funding Information

From the Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Mayo Clinic College of Medicine.

Correspondence to: Bekele Afessa, MD, FCCP, Division of Pulmonary and Critical Care, Mayo Clinic, 200 First St SW, Rochester, MN 55905; e-mail: afessa.bekele@mayo.edu


Financial/nonfinancial disclosures: The authors have reported to CHEST that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestpubs.org/site/misc/reprints.xhtml).


© 2010 American College of Chest Physicians


Chest. 2010;137(6):1488-1489. doi:10.1378/chest.10-0138
Text Size: A A A
Published online

To the Editor:

We thank Drs Desai and El Solh for their letter about our article in CHEST (December 2009)1 addressing the association between ICU admission during morning round time and mortality. In our study, we created a multiple logistic regression model by entering admission source, Acute Physiology and Chronic Health Evaluation (APACHE) III-predicted mortality rate, and intensity of treatment. Drs Desai and El Solh highlight that such an approach may lead to colinearity, resulting in imprecise estimates of regression coefficients and reduced tests of significance. In addition to several other variables, the APACHE III prognostic model includes the patient’s location before ICU admission as one of the predictor variables.2 The intensity of treatment is also greatly influenced by the severity of illness, measured by the APACHE III prognostic model.3 We share the concern raised by Drs Desai and El Solh. In a previous study, we tried to customize the APACHE III prognostic model to our patient population.4 In that study, the addition of the ICU admission source and intensity of treatment improved the mortality prediction, leading us to include them in our recent study in CHEST. Because of the colinearity concern raised in the letter, we performed a multiple logistic regression analysis without including the intensity of treatment and the source of admission as predictor variables. The observed association between admission during rounding time and risk of death remained significant, with an odds ratio (95% CI) of 1.644 (range, 1.466-1.842) and P < .001.

We agree with Drs Desai and El Solh that the use of multiple logistic regression analysis does not compensate for the lack of randomization. Although propensity score may have partly reduced the bias in our study, it would not have solved the limitations imposed by the lack of randomization. Propensity score analysis does not balance unobserved covariates,5 and although intuitively appealing, there are unanswered questions related to its advantage over the conventional logistic regression analysis in estimating unbiased treatment effects.6

We agree with Drs Desai and El Solh that the nurse-to-patient ratio and variations of other support staff may have impacts on the outcome of the critically ill. Our ICU nurse-to-patient ratio is not affected by round time and remains the same at all times. The same applies to respiratory therapists, who are present in the ICU at all times. Pharmacists are present in the ICU between 7:00 am and 11:00 pm and in the hospital at all times. Therefore, we do not think the overall nonphysician staffing by itself is likely to explain the observed association between admission during ICU rounds and mortality in our study.

Afessa B, Gajic O, Morales IJ, Keegan MT, Peters SG, Hubmayr RD. Association between ICU admission during morning rounds and mortality. Chest. 2009;1366:1489-1495. [CrossRef] [PubMed]
 
Knaus WA, Wagner DP, Draper EA, et al. The APACHE III prognostic system. Risk prediction of hospital mortality for critically ill hospitalized adults. Chest. 1991;1006:1619-1636. [CrossRef] [PubMed]
 
Zimmerman JE, Wagner DP, Knaus WA, Williams JF, Kolakowski D, Draper EA. The use of risk predictions to identify candidates for intermediate care units. Implications for intensive care utilization and cost. Chest. 1995;1082:490-499. [CrossRef] [PubMed]
 
Finkielman JD, Morales J, Peters SG, et al. Mortality rate and length of stay of patients admitted to the intensive care unit in July. Crit Care Med. 2004;325:1161-1165. [CrossRef] [PubMed]
 
Joffe MM, Rosenbaum PR. Invited commentary: propensity scores. Am J Epidemiol. 1999;1504:327-333. [CrossRef] [PubMed]
 
Weitzen S, Lapane KL, Toledano AY, Hume AL, Mor V. Principles for modeling propensity scores in medical research: a systematic literature review. Pharmacoepidemiol Drug Saf. 2004;1312:841-853. [CrossRef] [PubMed]
 

Figures

Tables

References

Afessa B, Gajic O, Morales IJ, Keegan MT, Peters SG, Hubmayr RD. Association between ICU admission during morning rounds and mortality. Chest. 2009;1366:1489-1495. [CrossRef] [PubMed]
 
Knaus WA, Wagner DP, Draper EA, et al. The APACHE III prognostic system. Risk prediction of hospital mortality for critically ill hospitalized adults. Chest. 1991;1006:1619-1636. [CrossRef] [PubMed]
 
Zimmerman JE, Wagner DP, Knaus WA, Williams JF, Kolakowski D, Draper EA. The use of risk predictions to identify candidates for intermediate care units. Implications for intensive care utilization and cost. Chest. 1995;1082:490-499. [CrossRef] [PubMed]
 
Finkielman JD, Morales J, Peters SG, et al. Mortality rate and length of stay of patients admitted to the intensive care unit in July. Crit Care Med. 2004;325:1161-1165. [CrossRef] [PubMed]
 
Joffe MM, Rosenbaum PR. Invited commentary: propensity scores. Am J Epidemiol. 1999;1504:327-333. [CrossRef] [PubMed]
 
Weitzen S, Lapane KL, Toledano AY, Hume AL, Mor V. Principles for modeling propensity scores in medical research: a systematic literature review. Pharmacoepidemiol Drug Saf. 2004;1312:841-853. [CrossRef] [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543