Statistical Models of Outcome of Cancer Patients Admitted to Intensive Care Units

Stan Lemeshow

Ohio State University

Refreshments: 3:30 - 4:00 p.m. Friday, November 5, at 327 Yost
Talk: 4:00 - 5:00 p.m. Friday, November 5, at 327 Yost.
Intensive care in the United States is estimated to account for more than 1% of the GNP. Models for estimating the probability of survival for patients entering an intensive care unit (ICU) exist, but are not accurate for cancer patients because such patients have much higher mortality rates than the typical patient admitted to a general medical/surgical ICU. Logistic regression modeling was used to estimate the probability that a cancer patient admitted to an ICU would survive to hospital discharge. Analyses were based upon 1483 patients. The resulting model, calculated at admission to the ICU, contained 16 easily determined variables and demonstrated excellent performance in its ability to accurately estimate outcome. A separate validation data set confirmed the performance of this model.

Questions? Nidhan Choudhuri