Development and initial validation of a computerized adaptive test prototype for ECG interpretation using item...

Area of Science:

• Medical Education
• Cardiology
• Psychometrics

Background:

• Accurate electrocardiogram (ECG) interpretation is crucial for clinical decisions, but competency is often suboptimal among healthcare professionals.
• Computerized adaptive testing (CAT) offers a potential solution for rapid, individualized, and standardized assessment of ECG interpretation skills.
• Implementing CAT requires a well-calibrated item bank based on item response theory.

Purpose of the Study:

• To establish item parameters for an ECG interpretation item bank.
• To evaluate the latent structure of ECG interpretation competency using multidimensional item response theory (MIRT).
• To develop an initial CAT prototype for ECG competency assessment.

Main Methods:

• Analysis of responses from 535 healthcare professionals on a 50-item online ECG test.
• Exploratory and confirmatory MIRT analyses to examine dimensionality.
• Estimation of item parameters (difficulty and discrimination) and model fit analyses.

Main Results:

• Both unidimensional and two-factor models were plausible latent structures for ECG interpretation competency.
• 47 out of 50 items demonstrated strong discrimination, with appropriately distributed difficulty parameters.
• Model fit analyses confirmed the selected model's suitability for CAT development.

Conclusions:

• A unidimensional model was selected for CAT development due to its simplicity and suitability.
• The calibrated ECG item bank facilitated the creation of a preliminary CAT prototype.
• CAT shows feasibility as an efficient and standardized method for assessing ECG interpretation competency in healthcare professionals.