Testing the reliability of a newly developed sedation scale

Abstract

A sedation tool for AICU (Adult Intensive Care Unit) patients is required to provide sound clinical care. Development of a new sedation scale was undertaken to overcome limitations of the available sedation scales. The goals of this research were to design a scale that possessed Content Validity with goal based parameters (with clearly defined parameter levels). In addition the tool was tested for reliability. To ensure the tool was reliable, the tool underwent interobserver reliability testing. An assessment tool composed of 9 goal based parameters was developed. A pilot study was conducted in 10 AICU patients to ensure that no gross problem areas were present. Following this, all minor alterations were completed. During the main trial phase, one parameter was assess in 211 AICU patients, while the remaining 8 parameters were assessed in 112 AICU patients. The developed tool doe possess Content Validity. The final results revealed that the scale as a whole "could" be deemed reliable based on the overall values for Percent Agreement (91.87%), Disagreement Rate (0.0813) and weighted Kappa (0.5137). Strigent reliability testing of each of the nine individual parameters revealed that 4 passed, 1 was borderline and 4 failed. Modification of the 5 substandard parameters is required in order to achieve complete reliability of the tool as a whole. Once this has been achieved, this tool, will have a significant impact on use of sedation in the critically ill patient.