JMIR Publications

ABSTRACT

Background:

Information technology can improve patient care plans to achieve high-quality care. However, few information systems can strengthen decision-making competency in health care environments. The most suitable and reliable inference engines for clinical decision support systems in nursing clinical practice have rarely been explored.

Objective:

This study examined the nursing diagnostic accuracy of a knowledge-based clinical decision support system (KBCDSS) based on clinical diagnostic validity (CDV) and the Bayesian decision models for nursing students to use during a psychiatric–mental health nursing practicum.

Methods:

A single-blinded, nonequivalent control group pretest–posttest design was adopted. Participants were 607 nursing students. In the quasi-experimental design, 2 intervention groups using a KBCDSS with the CDV (n=203) or the Bayesian decision (n=198) inference engine and a control group (n=206) used the psychiatric care plan system without guidance indicators to support decision-making.

Results:

Positive predictive value (PPV) and sensitivity indicated that decision-making competency was the highest for those in the CDV group followed by the Bayesian decision (BADE) and control groups. Relative to the control group, both the CDV and BADE groups exhibited significantly higher 3Q model questionnaire and modified TAM3 scores. In terms of perceived usefulness and behavioral intention, the CDV group had significantly higher 3Q model and modified TAM3 scores than did the BADE group, which had significantly higher scores than the control group did.

Conclusions:

KBCDSSs can be adopted to provide empirical patient-oriented information and help nursing students rapidly manage patient information and formulate patient-centered care plans.