Table 1: Number of patients by ICU hospital level and geographical locality.

Transcription

1 Web-based supporting materials for Evaluating the performance of Australian and New Zealand intensive care units in 2009 and 2010, by J. Kasza, J. L. Moran and P. J. Solomon Table 1: Number of patients by ICU hospital level and geographical locality. ICU geographical locality ICU level NT NSW ACT SA VIC WA NZ QLD TAS Total Rural Metropolitan Tertiary Private Total

2 var(log(smr)), assuming expected number of deaths fixed var(log(smr)) Figure 1: Estimated variances of the log-smrs of each ICU estimated under the assumption that the expected number of deaths is fixed, plotted against the estimated variances of the log-smrs of each ICU. The line is the y = x line, and indicates that variances of the log-smrs are underestimated when the expected number of deaths is assumed to be fixed. 2

3 3 Table 2: Stage 1 and Stage 2 model parameter estimates. All patients in the hematologic category were non-surgical, so no parameters were estimated for the interaction of hematologic with patient surgical status. Age and APACHE III have been scaled so that coefficients correspond to 10-year/unit increases. For example, for each 10 years that age increases, the log odds of in-hospital mortality increases by , and for each 10 units that APACHE III increases, the log odds of in-hospital mortality increases by For factor variables, the interpretation is the increase relative to the baseline level. For example, being an elective surgical patient, versus a non-surgical patient, decreases the log odds of in-hospital mortality by Log odds p-value 95% CI Log odds p-value 95% CI Age < < Age squared < < APACHE III score < < APACHE III score squared < < Age APACHE III score < < Gender Patient Category

4 4 Table 2 continued from previous page Gastrointestinal Metabolic < < Neurologic < < Respiratory < < Trauma < < Renal/ genitourinary < < Hematologic < < Patient surgical status Elective surgery < < Emergency surgery < < Ventilation < < ICU source Patient category APACHE III score Gastrointestinal APACHE III Metabolic APACHE III Neurologic APACHE III < < Respiratory APACHE III < < Trauma APACHE III < < Renal/ genitourinary APACHE III Hematologic APACHE III < < Patient category surgical status

5 5 Table 2 continued from previous page Gastrointestinal Elective surgery < < Gastrointestinal Emergency surgery Metabolic Elective surgery Metabolic Emergency surgery Neurologic Elective surgery < < Neurologic Emergency surgery < < Respiratory Elective surgery < < Respiratory Emergency surgery Trauma Elective surgery < < Trauma Emergency surgery Renal/ genitourinary Elective surgery Renal/ genitourinary Emergency surgery Hematologic Elective surgery (empty) (empty) Hematologic Emergency surgery (empty) (empty) Patient surgical status APACHE III score Elective surgery APACHE III Emergency surgery APACHE III Ventilation APACHE III score < < Ventilation Patient category Ventilation Gastrointestinal Ventilation Metabolic

6 6 Table 2 continued from previous page Ventilation Neurologic < < Ventilation Respiratory < < Ventilation Trauma Ventilation Renal/ genitourinary Ventilation Hematologic < < Annual volume Geographical locality Northern Territory Australian Capital Territory South Australia Victoria Western Australia New Zealand Queensland Tasmania Hospital level Rural Metropolitan Private Hospital level surgical status Rural Elective surgery

7 7 Table 2 continued from previous page Rural Emergency surgery Metropolitan Elective surgery Metropolitan Emergency surgery Private Elective surgery < < Private Emergency surgery < Hospital level ICU source Rural ICU source Metropolitan ICU source Private ICU source Intercept < < Random-effect parameters Estimate SE Estimate SE APACHE III coefficient variance Intercept variance Covariance of APACHE III and Intercept Separate intercepts and APACHE III coefficients for potentially unusual ICUs (Stage 2 model only) Intercepts APACHE III coefficients Log odds p-value 95% CI Log odds p-value 95% CI < < < < < <

8 Table 2 continued from previous page < < < < < < < < < < < < < < < < < <