- Meeting abstract
- Open Access
TISS 76 and TISS 28: outcome discrimination and correlation with length of ICU/hospital stay in 303 consecutive patients of a medical intensive care unit
Critical Care volume 4, Article number: P241 (2000)
TISS 76 (T76) and the simplified version TISS 28 (T28) quantify nursing workload and allow evaluation of severity of illness and, therefore, may be applied for outcome discrimination. The objective of this study was to compare T76 and T28 with regard to outcome discrimination in patients of a medical intensive care unit (ICU) and correlation with ICU length of stay (LOS) and hospital LOS.
All patients who stayed >24 h in ICU were included between 11/97 and 2/98. T76, T28 were collected daily. Discrimination power for survivors (S) and nonsurvivors (NS) was assessed by the area under the Receiver Operating Characteristic (AUROC) curve. Relation of both scores on day 1 as well as last measured score before ICU discharge (T76x, T28x) with ICU/hospital LOS and ICU/hospital mortality was assessed with Pearson's correlation and logistic regression.
303 patients (216 male [71.3%], 62± 12 years, ICU LOS 3.7± 4.7 days, hospital LOS 15.1± 13.2 days, SAPS II 26± 13) were studied. ICU mortality was 7.6%, hospital mortality was 14.5%. AUROC for T76 day 1 was 0.79± 0.04 for T28 0.76± 0.05. Correlation between T76/T28 day 1 and ICU/hospital LOS was only weak. T76x and T28x did not correlate with hospital LOS. Risk of death was associated with T76 (odds ratio 1.1, 95% CI 1.04 to 1.18) but not with T28 or T76x/T28x.
Outcome discrimination with T76 and T28 on day 1 was reliable. T76 was slightly superior to T28. Correlation with ICU/hospital LOS was only weak. Risk of death was significantly associated with T76 on day 1. Last measured score before ICU discharge did not correlate with hospital LOS or mortality.
About this article
Cite this article
Janssens, U., Graf, C., Graf, J. et al. TISS 76 and TISS 28: outcome discrimination and correlation with length of ICU/hospital stay in 303 consecutive patients of a medical intensive care unit. Crit Care 4, P241 (2000). https://doi.org/10.1186/cc960
- Public Health
- Logistic Regression
- Intensive Care Unit
- Receiver Operating Characteristic
- Operating Characteristic