Evaluation of the Accuracy of Nursing Diagnoses Determined by Users of a Clinical Decision Support System

To analyze the accuracy of nursing diagnoses determined by users of a clinical decision support system (CDSS) and to identify the predictive factors of high/moderate diagnostic accuracy.

This is an exploratory-descriptive study carried out from September 2017 to January 2018. Participants were nurses, resident nurses, and senior year undergraduates. Two written case studies provided the participants with the clinical data to fill out the assessment forms in the CDSS. The accuracy of the selected diagnostic labels was determined by a panel of experts using the Diagnostic Accuracy Scale, Version 2. Descriptive statistics were used to describe the level of accuracy according to each group of participants. Analysis of variance was used to compare the mean percentages of accuracy categories across groups. A linear regression model was used to identify the predictors of diagnostic accuracy. The significance level was 5%. The study was approved by the Ethics Committee.

Fifteen undergraduates, 10 residents, and 22 nurses were enrolled in the study. Although resident nurses and students had selected predominantly highly accurate diagnoses (51.8 ± 19.1 and 48.9 ± 27.4, respectively), and nurses had selected mostly diagnoses of moderate accuracy (54.7 ± 14.7), there were no differences in the accuracy level of selected diagnoses across groups. According to the linear regression model, each diagnosis added by the participants decreased the diagnostic accuracy by 2.09% (p = .030), and no experience or a low level of experience using the system decreased such diagnostic accuracy by 5.41% (p = .022).

The CDSS contributes to decision making about diagnoses of less experienced people. Adding diagnoses not indicated by the CDSS and experience with the system are predictors of diagnostic accuracy.

In-service education regarding the use of CDSSs seems to be crucial to improve users' clinical judgment and decision making.