Probabilistic classification of acute myocardial infarction from multiple cardiac markers

Logistic regression and Gaussian mixture model (GMM) classifiers have been trained to estimate the probability of acute myocardial infarction (AMI) in patients based upon the concentrations of a panel of cardiac markers. The panel consists of two new markers, fatty acid binding protein (FABP) and glycogen phosphorylase BB (GPBB), in addition to the traditional cardiac troponin I (cTnI), creatine kinase MB (CKMB) and myoglobin. The effect of using principal component analysis (PCA) and Fisher discriminant analysis (FDA) to preprocess the marker concentrations was also investigated. The need for classifiers to give an accurate estimate of the probability of AMI is argued and three categories of performance measure are described, namely discriminatory ability, sharpness, and reliability. Numerical performance measures for each category are given and applied. The optimum classifier, based solely upon the samples take on admission, was the logistic regression classifier using FDA preprocessing. This gave an accuracy of 0.85 (95% confidence interval: 0.78-0.91) and a normalised Brier score of 0.89. When samples at both admission and a further time, 1-6 h later, were included, the performance increased significantly, showing that logistic regression classifiers can indeed use the information from the five cardiac markers to accurately and reliably estimate the probability AMI. © Springer-Verlag London Limited 2008.

The Impact of Admissions for the Management of End-stage renal disease on hospital bed occupancy

BACKGROUND:

End-stage renal disease (ESRD) is increasingly prevalent but the inpatient costs associated with this condition are poorly defined due to limitations with data extraction and failure to differentiate between hospitalisation for renal and non-renal disease reasons. The impact of admissions primarily for the management of ESRD on hospital bed utilisation was assessed over a 5-year period in a large teaching hospital.

METHODS:

All admission episodes were reviewed and the ESRD group was identified by a primary International Classification of Diseases code for ESRD or a non-specific primary renal failure code with a secondary code for ESRD. The frequency and duration of hospitalisation and contribution to bed day occupancy of this group with ESRD was determined.

RESULTS:

There were 70,808 patients responsible for a total of 116,915 admissions and 919,212 bed days over the study period. Of these, 988 (1.4%) patients were admitted for the management of ESRD, accounting for 2,387 (2.0%) of admissions and utilisation of 23,011 (2.5%) bed days. After adjustment for age and gender, those admitted for ESRD management were significantly more likely to have a prolonged admission exceeding 30 days (odds ratio 1.46, 95% confidence interval 1.23-1.72, p < 0.001). When the admission was an emergency rather than an elective event, the patient was 4.6 times more likely to be hospitalised for over 30 days.

CONCLUSIONS:

Persons admitted for ESRD management are hospitalised more frequently and for longer than the overall inpatient population, occupying a substantial number of bed days.