Assessing Data Quality in Manual Entry of Ventilator Settings

The influence of methods for record keeping on the documentation of vital signs was assessed for the Anesthesia Information Management System (AIMS) NarkoData. We compared manually entered blood-pressure readings with automatically collected data. These data were stored in a data-base and subsequently evaluated and analyzed. The data sets were split into two groups, "manual" and "automatic". We evaluated the effect of automatic data collection on the incidence of corrected data, data validity and data variation. Blood-pressure readings of 37,726 data sets were analyzed. We could assess that the method of documentation did influence the data quality. It could not be assessed whether the incorrectness of data during automatic data gathering was caused by artefacts or by the anesthesiologist.

Influence of the method of data collection on the documentation of blood-pressure readings with an Anesthesia Information management system (AIMS)

Objective : The aim of this study was to test the hypothesis whether it is easily possible to transfer and apply the methods of data extraction and analysis of a performed study to a data pool of a different medical centre using the same type of Anaesthesia Information Management System (AIMS). For this purpose the objective of a study in cardiac anaesthesia, investigated at the University Hospital Giessen, was applied to the data pool of the Heart Centre Siegburg. Research design and methods : The Giessen study evaluated factors related to the use of positive inotropic drugs (PIDs) in adults undergoing elective cardiac surgery with cardiopulmonary bypass (CPB). The same objective and methods were applied to data of 1672 patients of the Heart Centre Siegburg. In both centres anaesthetic procedures were recorded with the AIMS NarkoData. Existing database queries were adapted according to the Siegburg database configuration for detection of patients having received PIDs during or after weaning from CPB. Res...

Potential benefits of Anaesthesia Information Management Systems for multicentre data evaluation: risk calculation of inotropic support in patients undergoing cardiac surgery.

https://www.anesthesiology.theclinics.com/article/S1932-2275(08)00057-8/pdf

The Value Proposition of Anesthesia Information Management Systems

Introduction. We evaluated whether automated anesthesia information systems can be used to calculate reference limits (population-based “normal values”) for vital signs. We considered four populations of women undergoing cesarean section: healthy under spinal anesthesia, healthy under general anesthesia, pre-eclamptic/eclamptic under spinal anesthesia, and pre-eclamptic/eclamptic under general anesthesia. Methods. Reference limits were calculated for each of the study populations by determination of percentiles for: minimum heart rate, maximum heart rate, minimum arterial oxyhemoglobin saturation (SaO2), minimum mean arterial pressure (MAP), maximum MAP, decrease in MAP, and increase in MAP. Results.There was one adverse anesthetic outcome among the 1,300 women in the study; the woman sustained a post-dural puncture headache. The 5th percentiles of SaO2 were at least 95% saturation under spinal versus90% under general. Under spinal anesthesia, 95th percentiles for decreases in MAP from baseline were 63 mmHg for healthy and 75 mmHg for pre-eclamptic/eclamptic women. Under general anesthesia, the 95th percentiles for maximum MAP were 161 and 177 mmHg, respectively. Two women of the 1,300 patients experienced simultaneously a minimum SaO2 <92% and minimum MAP <50 mmHg. Discussion. Automated anesthesia information systems can be used to determine reference limits for vital signs during anesthesia. Reference limits may play a role in malpractice cases when an expert claims that care by an anesthesiologist was sub-standard as shown by vital signs that were not maintained within the normal range during the critical portions of an anesthetic. Automated anesthesia information systems may enhance expert witnesses’ clinical judgment.

Norman J. Gould

Papers

Elusive artifact and cost issues with computerized patient records for anesthesia (CPRA)