FM sonography in diffuse liver disease: prospective assessment and blinded analysis

FM sonography - a signal-processing technique that uses frequency and phase information as well as amplitude data - shows promise in evaluation of patients with diffuse liver disease. In a prospective blinded review of 37 patients with biopsy-proved liver disease and 42 healthy volunteers, FM sonography was clearly superior to traditional amplitude-based (AM) sonography in distinguishing healthy from diseased subjects. Statistically significant differences were seen in accuracy (FM, 98.7%; AM, 84.8%), sensitivity (FM, 97.3%; AM, 70.3%), and negative predictive value (FM, 97.7%; AM, 78.8%). Our data also suggest that current FM sonographic techniques cannot differentiate among histologic findings associated with different hepatic parenchymal abnormalities. It is unclear, therefore, whether FM imaging can reduce the numbers of patients who require biopsy for diagnosis or the frequency of biopsy procedures in patients with known disease.     

Rates of in-hospital arrests, deaths and intensive care admissions: the effect of a medical emergency team

OBJECTIVES: To evaluate the effectiveness of a medical emergency team (MET) in reducing the rates of selected adverse events. DESIGN: Cohort comparison study after casemix adjustment. PATIENTS AND SETTING: All adult (> or = 14 years) patients admitted to three Australian public hospitals from 8 July to 31 December 1996. INTERVENTION STUDIED: At Hospital 1, a medical emergency team (MET) could be called for abnormal physiological parameters or staff concern. Hospitals 2 and 3 had conventional cardiac arrest teams. MAIN OUTCOME MEASURES: Casemix-adjusted rates of cardiac arrest, unanticipated admission to intensive care unit (ICU), death, and the subgroup of deaths where there was no pre-existing "do not resuscitate" (DNR) order documented. RESULTS: There were 1510 adverse events identified among 50 942 admissions. The rate of unanticipated ICU admissions was less at the intervention hospital in total (casemix-adjusted odds ratios: Hospital 1, 1.00; Hospital 2, 1.59 [95% CI, 1.24-2.04]; Hospital 3, 1.73 [95% CI, 1.37-2.16]). There was no significant difference in the rates of cardiac arrest or total deaths between the three hospitals. However, one of the hospitals with a conventional cardiac arrest team had a higher death rate among patients without a DNR order. CONCLUSIONS: The MET hospital had fewer unanticipated ICU/HDU admissions, with no increase in in-hospital arrest rate or total death rate. The non-DNR deaths were lower compared with one of the other hospitals; however, we did not adjust for DNR practices. We suggest that the MET concept is worthy of further study.     

Validation of a predictive model for asthma admission in children: how accurate is it for predicting admissions?

We studied 364 index presentations to the Emergency Department of a children's hospital with a diagnosis of asthma. The admission rate for this group of children was about 31%. We developed a parsimonious multiple logistic regression model to predict asthma hospital admission based on asthma severity indicators. We then evaluated the model's predictive ability using two methods of cross-validation, using the same sample that was used for the predictive model, and using data from a split sample. The logistic regression model had a predictive accuracy of 90% (95% confidence interval 85–95%). The sensitivity and specificity were 86% and 88%, respectively. Cross-validation models confirmed that the predictive ability of the model was stable. In studies with limited sample sizes, it is possible to validate a model without setting aside a split sample for cross-validation.