Artificial Intelligence-Assisted Learning for Nursing Drug Calculation

Medication administration errors are a significant challenge in nursing practice, particularly in high-acuity environments such as cardiovascular and critical care units. This study evaluates the effectiveness of an Artificial Intelligence-driven educational intervention designed to bridge the gap between theoretical knowledge and clinical application in drug calculations.

Study Design

This study employed a mixed-methods design comprising a quasi-experimental pretest-posttest approach with a control group, complemented by a qualitative focus group component. Participants were allocated to either an experimental group receiving Artificial Intelligence-assisted learning or a control group receiving traditional instruction

The Intervention (ٍStudy Group)

Participants in the experimental group used Artificial Intelligence-assisted learning software designed to enhance their educational experience through several advanced features. The software provides Adaptive Learning Paths, which adjust calculation complexity in accordance with the nurse's performance. Additionally, it offers Real-Time Feedback, ensuring immediate corrections and step-by-step guidance for complex drug dosing. Lastly, the software incorporates Artificial Intelligence-based Clinical Simulations that create high-pressure clinical decision-making scenarios for learners.

The Control Group

Participants in the control group received traditional teaching methods that encompassed standard lectures and paper-based practice sessions specifically aimed at drug calculation. This approach covered the same core curriculum as the experimental group but did not incorporate any Artificial Intelligence assistance.

The study evaluated three key areas before and after the intervention:

Nursing Knowledge, assessed using a standardized drug calculation examination.

Clinical Decision-Making, measured with a validated nursing decision-making scale.

Self-Efficacy, evaluated through a standardized self-efficacy scale to assess confidence in clinical calculations.

Data were analyzed using the Statistical Package for the Social Sciences to compare the mean scores between the experimental and control groups.

Qualitative Component (Focus Group Study)

To complement the quantitative findings, a qualitative focus group study was conducted with participants from the experimental group. The aim was to explore nurses' acceptance of the Artificial Intelligence platform, perceived usefulness, usability, visibility of learning progress, and overall opinions regarding the Artificial Intelligence-assisted teaching strategies.

Focus group discussions were audio-recorded, transcribed, and analyzed using thematic analysis to identify recurring patterns and themes related to user experience, perceived educational value, and readiness to integrate AI-based learning into clinical education. This qualitative component provided deeper insight into participants' attitudes toward AI integration in nursing education and enriched the interpretation of the quantitative outcomes.

• Artificial Intelligence-Assisted Drug Calculation Platform (Device): An innovative Artificial Intelligence software enhances nursing accuracy in drug calculations and clinical reasoning through scenario-based learning, providing real-time feedback and adaptive learning paths.
• Traditional Nursing Education (Other): Standard classroom-based instruction consists of theoretical lectures and paper-based practice focusing on medication dosage calculations.

Primary outcomes

• Nurses' Knowledge of Drug Calculation [Baseline (Pre-test) and 2 weeks post-intervention (Post-test)]: A 16-item assessment tool designed to evaluate the theoretical and practical knowledge of nurses regarding drug calculation principles (e.g., unit conversions, flow rate, and dose calculations). Each correct answer is scored "1" and each incorrect answer is scored "0". Scale Range: The total score ranges from a minimum of 0 to a maximum of 16. Interpretation: Higher scores indicate a better outcome (greater mastery of calculation principles). High (13-16): Competent level (> 80%). Moderate (10-12): Acceptable but incomplete knowledge (60%-80%). Low (0-9): Deficient understanding (< 60%).

Secondary outcomes

• Nurses' Drug Calculation Decision-Making Scale [Baseline (Pre-test) and 2 weeks post-intervention (Post-test)]: This scale is an 8-item self-report tool designed to assess the clinical judgment and confidence of nurses regarding medication dosage calculations. Each item is rated on a 5-point Likert scale ranging from 1 (Strongly Disagree) to 5 (Strongly Agree). Minimum and Maximum Values: The total score ranges from a minimum of 8 to a maximum of 40. Interpretation: Higher scores indicate a better outcome (greater clinical competence and safer decision-making). High (30-40): High competence. Moderate (19-29): Moderate ability; requires supervision. Low (8-18): Poor confidence and judgment.
• General Self-Efficacy Scale [Baseline (Pre-test) and 2 weeks post-intervention (Post-test)]: Description: A 10-item psychometric scale used to assess nurses' perceived confidence and self-belief in their ability to perform drug calculations and clinical tasks effectively under various conditions. Each item is scored on a 4-point Likert scale: 1 (Not at all true), 2 (Hardly true), 3 (Moderately true), and 4 (Exactly true). Scale Range: The total score ranges from a minimum of 10 to a maximum of 40. Interpretation: Higher scores indicate a better outcome (stronger perceived competence and higher self-efficacy). High (31-40): Strong perceived competence and self-belief. Moderate (21-30): Moderate confidence in managing demands. Low (10-20): Low belief in ability to cope with challenges.
• Nurses' Perception and Satisfaction with Artificial Intelligence-Assisted Learning (Qualitative) [2 weeks after the completion of the AI-assisted training]: Description: Assessment of participants' acceptance, perceived usefulness, usability, and satisfaction with the Artificial Intelligence-assisted learning platform. Data will be collected through focus group discussions. Method of Analysis: Results will be analyzed using Thematic Analysis to identify recurring patterns and themes. Unit of Measure: This is a qualitative outcome; results will be reported as narrative themes (for example: "Improved Calculation Confidence" or "User Interface Satisfaction").

Sex: All, ages: 18 Years to any, accepts healthy volunteers

Inclusion Criteria:

Nurses working in multiple clinical settings, including medical-surgical, cardiovascular, or critical care units..etc.

Nurses are responsible for medication administration and drug dosage calculations as part of their daily clinical duties.

Willingness to participate in the Artificial Intelligence-assisted learning program and sign the informed consent.

Exclusion Criteria:

Nurses who had recently received specific training in drug-calculation or had any prior exposure to AI-based educational tools (within the last 6 months)

• Results publication: Salem MFA, Sharaf AK, Abdelhafeez Abdelmonseif Younis A. Adaptive AI-based mobile simulation for drug calculation competency in nurses: a mixed-methods RCT. BMC Nurs. 2026 Sep 8;25(1):828. doi: 10.1186/s12912-026-05313-4. PMID 42717352