A dataset containing the SOFA scores (Vincent et al, 1996). for 520 patients, hospitalized in the intensive care unit (ICU) with Acute Lung Injury. Daily measurements are available for as long as each one remains in the ICU. This is an example of variable-domain functional data, as described by Gellar et al. (2014).
sofaA data frame with 520 rows (subjects) and 7 variables:
binary indicator that the subject died in the ICU
520 x 173 matrix in variable-domain format (a ragged array).
Each column represents an ICU day. Each row contains the SOFA scores for
a subject, one per day, for as long as the subject remained in the ICU.
The remaining cells of each row are padded with NAs. SOFA scores
range from 0 to 24, increasing with severity of organ failure. Missing
values during one's ICU stay have been imputed using LOCF.
Identical to the SOFA element, except that it contains
some missing values during one's hospitalization. These missing values
arise when a subject leaves the ICU temporarily, only to be re-admitted.
SOFA scores are not monitored outside the ICU.
ICU length of stay, i.e., the number of days the patient remained in the ICU prior to death or final discharge.
Patient age
Binary indicator for male gender
Charlson co-morbidity index, a measure of baseline health status (before hospitalization and ALI).
The data was collected as part of the Improving Care of ALI Patients (ICAP) study (Needham et al., 2006). If you use this dataset as an example in written work, please cite the study protocol.
Vincent, JL, Moreno, R, Takala, J, Willatts, S, De Mendonca, A, Bruining, H, Reinhart, CK, Suter, PM, Thijs, LG (1996). The SOFA ( Sepsis related Organ Failure Assessment) score to describe organ dysfunction/failure. Intensive Care Medicine, 22(7): 707-710.
Needham, D. M., Dennison, C. R., Dowdy, D. W., Mendez-Tellez, P. A., Ciesla, N., Desai, S. V., Sevransky, J., Shanholtz, C., Scharfstein, D., Herridge, M. S., and Pronovost, P. J. (2006). Study protocol: The Improving Care of Acute Lung Injury Patients (ICAP) study. Critical Care (London, England), 10(1), R9.
Gellar, Jonathan E., Elizabeth Colantuoni, Dale M. Needham, and Ciprian M. Crainiceanu. Variable-Domain Functional Regression for Modeling ICU Data. Journal of the American Statistical Association, 109(508):1425-1439, 2014.