Evaluating a medication error prevention training for nursing students using Kirkpatrick’s model Authors: Nayereh Baghcheghi, Hamid Reza Koohestani Published: 2025 ABSTRACT BackgroundMedication administration is a critical aspect of nursing, and insufficient competence among students can cause errors and compromise patient safety. This study aimed to design, implement, and evaluate an educational program for preventing medication errors in nursing students.MethodsIn this quasi-experimental study with a two-group pretest-posttest design, 69 nursing students were selected through convenience sampling and randomly assigned to intervention (n = 35) or control (n = 34) groups. The program, based on Kern’s six-step model, was implemented in the intervention group and its effectiveness evaluated using Kirkpatrick’s four-level model: reaction (satisfaction), learning (knowledge and attitude), behavior (performance), and results (error incidence). Data on knowledge, attitude, and performance were collected before and after the intervention with validated questionnaires and analyzed using SPSS 24.Results80% of participants reported high satisfaction with the program. No significant difference was observed between groups at pre-test (p > 0.05). However, Post-test 1 (immediately after) and Post-test 2 (four months later) showed significant differences in knowledge and attitude (p < 0.001). Performance also improved significantly after four months (p < 0.001). During follow-up, the control group reported one actual and two near-miss errors, while the intervention group reported only two near-miss errors.ConclusionThe educational program had a significant and lasting impact on improving nursing students’ knowledge, attitude, and performance in preventing medication errors. The high level of participant satisfaction indicates a favorable reception of the program. It is recommended that such structured educational programs be incorporated into nursing curricula.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12909-025-08350-w. Background Patient safety is a foundational principle of healthcare [1], and medication errors represent one of its most significant and persistent threats, jeopardizing patient well-being and imposing substantial costs on healthcare systems [2–4]. The prevalence of medication errors is high in many countries and occurs frequently among healthcare professionals, including physicians, pharmacists, and nurses [5]. Nurses play a pivotal role in preventing medication errors, yet many new graduates lack confidence in their ability to prevent such errors due to limited knowledge and insufficient practical trainin [6]. While most medication errors are preventable, nurses spend approximately 40% of their work time on medication administration, giving them a key role in error reduction [7]. Globally, patient safety has been increasingly emphasized in nursing education curricula. Nurses are expected to develop adequate competencies in medication error prevention to minimize risks and ensure the safe administration of medications. Nevertheless, research shows that nursing students remain vulnerable to medication errors, especially during clinical training [8, 9]. Evidence from a scoping review indicates that medication errors by nursing students in clinical settings occur at a rate of 1.1% to 6%. Key contributors include knowledge gaps, training deficiencies, and systemic factors. Enhancing education, supervision, and clinical protocols is crucial to reducing these errors and improving patient safety [10]. To achieve competence in safe medication administration, undergraduate nursing students require sufficient clinical training [10]. Evidence suggests that many undergraduate nursing curricula fail to adequately prepare students regarding factors contributing to medication errors and strategies for their prevention [11]. While educating nursing students to identify high-risk medication situations is crucial, and approaches such as simulation-based training have shown promising results in improving attitudes toward error reporting [11, 12], robust empirical evidence specifically evaluating the effectiveness of these educational interventions on practical performance, actual reduction of medication errors, and particularly the sustainability of educational effects over time among nursing students remains limited. This lack of compelling evidence highlights a distinct research gap. Therefore, there is a clear need to design and evaluate structured educational programs that not only enhance knowledge but also focus on practical behaviors related to error prevention and the long-term sustainability of learning outcomes. The present study aims to address this gap by designing, implementing, and evaluating a comprehensive educational program for medication error prevention among nursing students, with particular emphasis on the sustainability of its educational effects. Methods This study employed a two-group, pre-test–post-test quasi-experimental design conducted in 2024. Although the target population comprised all undergraduate nursing students, random sampling from the entire student body was not feasible due to educational constraints and varying learning needs. Thus, two pre-existing classes of 3rd- and 5th-semester students, most relevant to the educational intervention, were selected. Within each class, students were randomly assigned to intervention and control groups to ensure comparability. Accordingly, due to the non-random selection of participants (despite random assignment), the study was considered quasi-experimental. This aspect represents the quasi-experimental nature of our sampling approach. The curriculum development process followed Kern et al.’s six-step model [13], as outlined below: Step 1: problem identification and general needs assessment This step involves determining the gap between current and desired outcomes at a system level and identifying the general health/educational problem that the curriculum should address. In this phase, medication errors among nursing students were identified as a critical healthcare problem. Their prevalence, potential causes, and consequences were examined at local, national, and international levels. To inform program development, the researchers conducted interviews with nursing faculty members (N = 8) and reviewed existing literature and public health data. Step 2: assessment of target learners’ needs This step entails assessing the specific learning needs, characteristics, and contextual constraints of the intended learners to tailor learning objectives and content. The specific educational needs of nursing students were explored through interviews with current students (N = 15) and recent graduates (N = 9). Step 3: defining goals and objectives This step translates identified needs into broad goals and specific, measurable learning objectives that the curriculum will achieve. Based on the needs assessment, the overarching goal was: to design and implement an educational program for the prevention medication errors among nursing students, and to evaluate it using Kirkpatrick’s model. The specific objectives were to enhance students’ knowledge, attitudes, and practices regarding medication errors and to assess learners’ immediate reactions to the program (Kirkpatrick’s first level). Step 4: selection of educational strategies In this step, appropriate instructional methods and learning activities are selected to meet the stated objectives, taking into account adult learning principles and contextual feasibility. The instructional design integrated lectures, scenario-based learning, and group discussions. Step 5: implementation This step describes how the curriculum will be delivered: target groups, instructors, timetable, resources, and logistical arrangements needed for real-world rollout. A total of 78 undergraduate nursing students from the 3rd and 5th semesters were invited to participate in the study. Of these, 69 students met the inclusion criteria and completed both pre- and post-test assessments (experimental group = 35, control group = 34). The sample size for this study was determined based on practical considerations and alignment with similar interventional studies in nursing education. This sample size is consistent with those employed in comparable quasi-experimental studies in this field. For example, recent studies in nursing education have employed similar sample sizes, such as 63 participants [12], and 52 participants [14]. Students were randomly assigned to either the intervention or control group to ensure comparability. Participants were randomly assigned to either the intervention or control group using simple randomization. A random number table was used to ensure equal probability of allocation. The assignment process was conducted by an independent researcher who was not involved in the intervention or assessment procedures to minimize allocation bias. The intervention consisted of a one-day, 5-hour workshop, which employed a combination of lectures, group discussions, and scenario-based learning to enhance engagement and practical application of medication administration principles. The workshop was facilitated by the primary author, an associate professor of nursing with 20 years of teaching experience, who guided participants through structured scenarios, reflective exercises, and problem-solving activities relevant to real clinical situations (Fig. 1).Fig. 1Study flowchart of participant recruitment, intervention, and assessment Study flowchart of participant recruitment, intervention, and assessment Inclusion criteria were: (1) enrollment in the 3rd or 5th semester of the undergraduate nursing program (2), prior clinical experience in medication administration, and (3) willingness to participate in both the workshop and study assessments. Exclusion criteria included: (1) unwillingness to attend the workshop or participate in assessments, and (2) incomplete pre- or post-test data. The control group did not receive any specific intervention and continued with their regular nursing curriculum. Step 6: Evaluation and feedback This step involves selecting evaluation methods to determine whether objectives were met (reaction, learning, behavior, results) and using feedback to refine the curriculum. Program evaluation was conducted according to Kirkpatrick’s four-level model, which provides a systematic framework for assessing training effectiveness at multiple outcomes. Each level was clearly defined and measured using specific tools and procedures as follows: Reaction level: Reaction level: Reaction level: This level evaluates participants’ immediate satisfaction and perceived relevance of the educational experience. In this study, students’ satisfaction with the educational program was assessed immediately after the intervention using a 6-item, 5-point Likert scale satisfaction questionnaire, focusing on clarity, usefulness, and applicability of the training content. 2.Learning level: 2.Learning level: Learning level: This level measures the extent to which participants have acquired the intended knowledge, skills, and attitudes. Learning outcomes were assessed using two instruments: a 22-item knowledge test (including 12 questions on medication principles and 10 on drug calculations) and a 23-item attitude questionnaire. Both instruments were administered before the intervention, immediately after, and four months later to evaluate both short-term and retained learning. 3.Behavior level: 3.Behavior level: Behavior level: This level reflects the degree to which participants apply what they learned in real or simulated practice settings. In this study, behavioral change was evaluated using a 14-item, 5-point Likert scale performance questionnaire that assessed students’ adherence to safe medication practices. The questionnaire was administered before and four months after the intervention, allowing sufficient time for behavioral adaptation to occur. 4.Results level: 4.Results level: Results level: This level represents the final outcomes or impact of the educational intervention on clinical practice and patient safety. In this study, results were operationalized as the frequency and nature of medication errors reported by participants through four self-report items. These reports were reviewed and verified by clinical instructors who supervised students during their clinical rotations to ensure the accuracy and reliability of data. After the students completed the drug dose calculation questions, their responses were scored by one of the authors using a predetermined answer key and standard calculation procedures. The other questionnaires were self-reported and completed by the students. Since no comprehensive questionnaire existed that addressed the specific requirements of our educational context, aligned with the Kirkpatrick model, and provided an integrated assessment of all dimensions of medication error prevention, the researchers utilized a researcher-developed questionnaire for this study (supplementary file1). The questionnaire was developed through a literature review, with its validity and reliability rigorously tested. To determine content validity, the input of 10 faculty members and experts in the field was solicited. Content validity assessment encompasses both qualitative and quantitative approaches, both of which were utilized in this study. In the qualitative phase, specialists were requested to provide feedback regarding grammatical correctness, the use of appropriate and accurate wording, proper item placement, and the suitability and correctness of the scoring system. In the quantitative phase, to ensure the selection of the most significant and accurate content for the questionnaires, the Content Validity Ratio (CVR) and the Content Validity Index (CVI) were employed, respectively. For determining the CVR, and in accordance with Lawshe’s table, the minimum acceptable CVR value for a panel of 10 experts was set at a cut-off point of 0.62. Based on this criterion, one item from the knowledge questionnaire and two items from the performance questionnaire obtained CVR values below the cut-off point and were consequently removed. Finally, the training satisfaction questionnaire (with 6 items), the knowledge questionnaire (with 22 items including 12 questions on medication principles and 10 on drug calculations), the attitude questionnaire (with 23 items), and the performance evaluation questionnaire (with 14 items) proceeded to the CVI determination stage. For the CVI in this study, the cut-off point was set at 0.8. The Kappa statistic, which indicates the level of agreement between raters beyond that expected by chance, was used. On this basis, all items demonstrated CVI values higher than 0.8 (with an excellent Kappa coefficient), and consequently, no items were removed at this stage. To assess reliability, Cronbach’s alpha was used for three questionnaires (satisfaction, attitude, and performance), yielding acceptable scores of 0.83, 0.79, and 0.86, respectively. The Kuder-Richardson 20 (KR-20) coefficient was used for knowledge assessments, also showing high reliability with scores of 0.88 for the medication principles questionnaire and 0.89 for the drug calculations test. Data analysis Data were analyzed using SPSS version 24. Descriptive statistics (mean, standard deviation, and frequency) were used to summarize participants’ characteristics and study variables. The normality of data distribution was assessed using the Kolmogorov–Smirnov test, and homogeneity of variances was verified using Levene’s test. A two-way repeated-measures ANOVA was applied as the primary statistical method to examine the main effects of Time and Group, as well as the Group × Time interaction effect across knowledge, drug calculation, attitude, and performance domains. In cases where data did not meet normality assumptions, the nonparametric Friedman test was used. Post-hoc analyses with Bonferroni adjustment were conducted to explore significant main or interaction effects. Effect sizes were reported using partial eta squared (ηp²) for ANOVA results and Cohen’s d for pairwise comparisons. All tests were two-tailed, with the significance level set at p < 0.05. Results A total of 69 nursing students participated in this study. Approximately 54% of the participants (n = 37) were female, and about 52% (n = 36) were fifth-semester students, while the remaining participants were third-semester students. The mean age of the participants was 22.11 ± 1.14 years. There were no statistically significant differences between the intervention and control groups in terms of gender, semester, or age (p > 0.05). Reaction level evaluation results The results regarding participants’ satisfaction with the educational program were presented in Fig. 2. Fig. 2Participants’ satisfaction level with the educational program Participants’ satisfaction level with the educational program As shown in Figs. 2 and 80% of the students expressed a very high level of satisfaction with the conducted educational program. Learning level evaluation results Knowledge Medication principles To examine the effects of the intervention over time, a two-way repeated-measures ANOVA was conducted with Group (Intervention vs. Control) as the between-subjects factor and Time (Pre-test, Post-test 1, Post-test 2) as the within-subjects factor. All assumptions for the repeated-measures ANOVA, including normality (Kolmogorov-Smirnov test, p > 0.05), sphericity (Mauchly’s test, p > 0.05), and homogeneity of variances (Levene’s test, p > 0.05), were met. The results revealed a non-significant main effect of Group at the pre-test stage (p > 0.05), confirming baseline homogeneity. The analysis showed a significant main effect of Time, (F = 44.6, p < 0.001, ηp² = 0.39). More importantly, there was a significant Group × Time interaction with a large effect size, (F = 35.78, p < 0.001, ηp² = 0.34). Post-hoc pairwise comparisons with Bonferroni correction were conducted to decompose the significant interaction. As illustrated in Fig. 3, simple effects analysis showed that while the two groups did not differ at pre-test (p > 0.05), the intervention group demonstrated a significant and large improvement in scores at post-test 1 compared to the control group (p < 0.001(. This significant difference was maintained at post-test 2 (p < 0.001), indicating a sustained effect of the intervention.Fig. 3Mean knowledge domain scores in the control and intervention groups Mean knowledge domain scores in the control and intervention groups Table 1 presents the proportion of correct responses and the mean (± SD) of knowledge scores across the pre-test, post-test 1, and post-test 2 in the intervention group. The Kolmogorov–Smirnov test indicated non-normal data distribution (p < 0.05); therefore, the Friedman test was used as a non-parametric alternative to repeated-measures ANOVA. The results showed significant differences among the three stages (χ² = 57.08, p < 0.001, Kendall’s W = 0.81), indicating a large effect size.Table 1Absolute and relative frequency distribution of correct responses regarding knowledge in the intervention groupNo.Question (Correct Answer)Pre-testPost-test 1Post-test 2n%n%n%1Which of the following is not one of the 8 rights? (Right diagnosis)2674.3351003497.12In which situations should the label on the drug container be checked against the medication Kardex? (When removing the drug from the shelf, during preparation, before returning the container to the shelf, or before administering to the patient)1542.9351003394.33What is the most appropriate way to confirm patient identity? (Asking the patient’s name and checking it against the wristband)3497.135100351004Which is the appropriate time for the nurse to ensure that the patient has no drug allergy? (Before prescribing or administering the first dose)1028.63497.13497.15Which of the following should be checked during drug preparation? (Expiration date and clarity of the drug)3085.73497.13394.36When should drug administration be documented by the nurse? (Immediately after administration)14403497.13497.17What is one of the main purposes of accurate and complete drug documentation? (Enabling follow-up in case of adverse drug reactions)2674.33291.43188.68Monitoring and documenting drug side effects in patients reflects which principle of drug administration? (Right response)2777.1351003394.39The first step after a medication error is to record the error (False)1542.83497.13510010All tablets can be crushed (False)2057.1351003510011A nurse should not administer medication prepared by a colleague (True)3085.7351003497.112If a drug is recorded at the wrong time, the entry should be erased (False)2057.13510035100Total score(Mean ± SD)7.71 ± 2.511.82 ± 0.3811.6 ± 0.6 Table 1Absolute and relative frequency distribution of correct responses regarding knowledge in the intervention groupNo.Question (Correct Answer)Pre-testPost-test 1Post-test 2n%n%n%1Which of the following is not one of the 8 rights? (Right diagnosis)2674.3351003497.12In which situations should the label on the drug container be checked against the medication Kardex? (When removing the drug from the shelf, during preparation, before returning the container to the shelf, or before administering to the patient)1542.9351003394.33What is the most appropriate way to confirm patient identity? (Asking the patient’s name and checking it against the wristband)3497.135100351004Which is the appropriate time for the nurse to ensure that the patient has no drug allergy? (Before prescribing or administering the first dose)1028.63497.13497.15Which of the following should be checked during drug preparation? (Expiration date and clarity of the drug)3085.73497.13394.36When should drug administration be documented by the nurse? (Immediately after administration)14403497.13497.17What is one of the main purposes of accurate and complete drug documentation? (Enabling follow-up in case of adverse drug reactions)2674.33291.43188.68Monitoring and documenting drug side effects in patients reflects which principle of drug administration? (Right response)2777.1351003394.39The first step after a medication error is to record the error (False)1542.83497.13510010All tablets can be crushed (False)2057.1351003510011A nurse should not administer medication prepared by a colleague (True)3085.7351003497.112If a drug is recorded at the wrong time, the entry should be erased (False)2057.13510035100Total score(Mean ± SD)7.71 ± 2.511.82 ± 0.3811.6 ± 0.6 Absolute and relative frequency distribution of correct responses regarding knowledge in the intervention group Post-hoc Kruskal–Wallis tests with Bonferroni adjustment revealed that both post-test scores were significantly higher than pre-test scores (p < 0.05), while the difference between the two post-tests was not significant. These findings confirm that the educational intervention effectively improved participants’ knowledge and that this improvement was maintained over time. Drug calculations A two-way repeated-measures ANOVA was conducted to examine the effects of the educational intervention on drug calculation scores over time, with Group (Intervention vs. Control) as the between-subjects factor and Time (Pre-test, Post-test 1, Post-test 2) as the within-subjects factor. All statistical assumptions were met, including normality (Kolmogorov-Smirnov test, p > 0.05), sphericity (Mauchly’s test, p > 0.05), and homogeneity of variances (Levene’s test, p > 0.05). The analysis revealed a significant main effect of Time (F = 80.93, p < 0.001, ηp² = 0.71) More importantly, there was a significant Group × Time interaction,) F = 74.66, p < 0.001, ηp² = 0.69(with a large effect size, indicating that the pattern of score changes over time differed significantly between the intervention and control groups. Post-hoc pairwise comparisons with Bonferroni correction were conducted to decompose the significant interaction. Consistent with baseline homogeneity, no significant difference was found between groups at pre-test (p > 0.05). At post-test 1, the intervention group demonstrated significantly higher scores compared to the control group (p < 0.001). This significant difference was maintained at post-test 2 (p < 0.001). These results confirm that the educational intervention substantially enhanced students’ drug calculation performance, and this improvement was sustained over time (Fig. 4).Fig. 4Mean drug calculation scores in the control and intervention groups Mean drug calculation scores in the control and intervention groups Attitudes To evaluate the intervention’s impact on nursing students’ attitudes over time, a two-way repeated-measures ANOVA was conducted with Group (Intervention vs. Control) as the between-subjects factor and Time (Pre-test, Post-test 1, Post-test 2) as the within-subjects factor. All statistical assumptions were verified and met. The analysis yielded a significant main effect of Time with a substantial effect size,)F = 7.6, p < 0.001, ηp² = 0.1(. Crucially for our research objective, a significant Group × Time interaction with a medium effect size was observed,)F = 5.6, p < 0.001, ηp² =0.07(, demonstrating that the evolution of attitudes over time followed distinctly different trajectories between the two groups. Post-hoc analysis with Bonferroni adjustment confirmed that while no baseline differences existed between groups (p > 0.05), following the intervention, the intervention group showed significantly more positive attitudes at post-test 1 (p < 0.001). This attitudinal improvement remained stable at post-test 2 (p < 0.001). In summary, the educational intervention effectively enhanced nursing students’ attitudes, with substantial and sustained effects observed throughout the study period (Fig. 5).Fig. 5Mean attitude scores in the control and intervention groups Mean attitude scores in the control and intervention groups Table 2 presents the mean and standard deviation of attitude scores across the pre-test, post-test 1, and post-test 2. The Kolmogorov–Smirnov test confirmed the normal distribution of data (p > 0.05). A repeated measures ANOVA was conducted to compare attitude scores across the three stages, after confirming the assumption of sphericity using Mauchly’s test (p > 0.05). Results indicated a significant difference in attitude scores among the three time points (F = 8.25, p < 0.001, partial η² = 0.33), representing a large effect size. Bonferroni-adjusted pairwise comparisons revealed that mean scores at both post-test phases were significantly higher than pre-test scores (p < 0.05), with no significant difference between post-test 1 and post-test 2. These results confirm that the educational intervention effectively improved students’ attitudes toward medication error prevention and that this positive effect was sustained over time.Table 2Mean (± SD) attitude scores of nursing students in the intervention groupNo.ItemPre-test (Mean ± SD)Post-test 1 (Mean ± SD)Post-test 2 (Mean ± SD)1I prefer not to report a medication error because I am worried about being blamed by instructors or managers.2.97 ± 0.784.08 ± 0.814.02 ± 0.852I avoid reporting errors because I fear being labeled as a troublesome student.3.02 ± 0.563.62 ± 0.913.68 ± 0.863I refrain from reporting medication errors due to fear of the instructor’s reaction.2.80 ± 0.673.71 ± 0.893.82 ± 0.924I am concerned that reporting an error will make classmates or staff perceive me as incompetent.2.91 ± 0.563.65 ± 0.763.74 ± 0.745I fear punitive consequences such as grade reduction or loss of clinical placement.3.03 ± 0.893.31 ± 0.583.45 ± 0.656I am afraid of possible legal prosecution after reporting an error.2.94 ± 0.633.25 ± 0.613.37 ± 0.647I am worried about malpractice-related lawsuits.2.97 ± 0.613.54 ± 0.853.54 ± 0.748I am concerned about public disclosure of errors through the media.3.31 ± 0.523.60 ± 0.733.51 ± 0.749I believe reporting errors may create a negative perception of me among patients and their families.2.97 ± 0.563.40 ± 0.853.34 ± 0.9010I think reporting medication errors may cause educational problems for me.2.71 ± 0.753.45 ± 0.653.34 ± 0.5911I believe reporting medication errors can damage my dignity and reputation.2.94 ± 0.633.48 ± 0.703.57 ± 0.6912I think reporting medication errors may endanger my future career.2.77 ± 0.493.85 ± 0.943.77 ± 0.9713I believe some nursing students do not report their errors for fear of punishment.3.22 ± 0.543.50 ± 0.733.45 ± 0.7414I believe cultural differences in defining errors lead to underreporting by some students.3.34 ± 0.633.48 ± 0.783.37 ± 1.0015I believe punishment should be proportional to the consequences of the error and whether it is voluntarily reported.3.34 ± 0.723.65 ± 0.833.74 ± 0.8516I believe in some situations medication errors should be ignored.3.31 ± 0.673.45 ± 0.653.54 ± 0.7417I believe most medication errors are unavoidable and inevitable.3.05 ± 0.763.54 ± 0.783.65 ± 0.7218I believe specialized education can reduce medication errors.3.14 ± 0.733.60 ± 0.843.68 ± 0.7519I believe further research and innovative drug delivery methods are essential to enhance skills.3.11 ± 0.713.31 ± 0.833.42 ± 0.8520I believe barcode technology on medication labels can reduce medication errors.3.17 ± 0.743.11 ± 0.633.17 ± 0.6121I believe automated medication dispensing technology can reduce medication errors.3.11 ± 0.793.31 ± 0.753.37 ± 0.8422I believe error reporting is necessary even if no harm occurs to the patient.3.08 ± 0.743.40 ± 0.733.45 ± 0.7023I believe patients have the right to be informed of medication errors, regardless of harm.2.68 ± 0.632.88 ± 0.632.71 ± 0.62Total scoremean ± SD70.02 ± 2.6080.37 ± 3.9780.80 ± 4.26 Table 2Mean (± SD) attitude scores of nursing students in the intervention groupNo.ItemPre-test (Mean ± SD)Post-test 1 (Mean ± SD)Post-test 2 (Mean ± SD)1I prefer not to report a medication error because I am worried about being blamed by instructors or managers.2.97 ± 0.784.08 ± 0.814.02 ± 0.852I avoid reporting errors because I fear being labeled as a troublesome student.3.02 ± 0.563.62 ± 0.913.68 ± 0.863I refrain from reporting medication errors due to fear of the instructor’s reaction.2.80 ± 0.673.71 ± 0.893.82 ± 0.924I am concerned that reporting an error will make classmates or staff perceive me as incompetent.2.91 ± 0.563.65 ± 0.763.74 ± 0.745I fear punitive consequences such as grade reduction or loss of clinical placement.3.03 ± 0.893.31 ± 0.583.45 ± 0.656I am afraid of possible legal prosecution after reporting an error.2.94 ± 0.633.25 ± 0.613.37 ± 0.647I am worried about malpractice-related lawsuits.2.97 ± 0.613.54 ± 0.853.54 ± 0.748I am concerned about public disclosure of errors through the media.3.31 ± 0.523.60 ± 0.733.51 ± 0.749I believe reporting errors may create a negative perception of me among patients and their families.2.97 ± 0.563.40 ± 0.853.34 ± 0.9010I think reporting medication errors may cause educational problems for me.2.71 ± 0.753.45 ± 0.653.34 ± 0.5911I believe reporting medication errors can damage my dignity and reputation.2.94 ± 0.633.48 ± 0.703.57 ± 0.6912I think reporting medication errors may endanger my future career.2.77 ± 0.493.85 ± 0.943.77 ± 0.9713I believe some nursing students do not report their errors for fear of punishment.3.22 ± 0.543.50 ± 0.733.45 ± 0.7414I believe cultural differences in defining errors lead to underreporting by some students.3.34 ± 0.633.48 ± 0.783.37 ± 1.0015I believe punishment should be proportional to the consequences of the error and whether it is voluntarily reported.3.34 ± 0.723.65 ± 0.833.74 ± 0.8516I believe in some situations medication errors should be ignored.3.31 ± 0.673.45 ± 0.653.54 ± 0.7417I believe most medication errors are unavoidable and inevitable.3.05 ± 0.763.54 ± 0.783.65 ± 0.7218I believe specialized education can reduce medication errors.3.14 ± 0.733.60 ± 0.843.68 ± 0.7519I believe further research and innovative drug delivery methods are essential to enhance skills.3.11 ± 0.713.31 ± 0.833.42 ± 0.8520I believe barcode technology on medication labels can reduce medication errors.3.17 ± 0.743.11 ± 0.633.17 ± 0.6121I believe automated medication dispensing technology can reduce medication errors.3.11 ± 0.793.31 ± 0.753.37 ± 0.8422I believe error reporting is necessary even if no harm occurs to the patient.3.08 ± 0.743.40 ± 0.733.45 ± 0.7023I believe patients have the right to be informed of medication errors, regardless of harm.2.68 ± 0.632.88 ± 0.632.71 ± 0.62Total scoremean ± SD70.02 ± 2.6080.37 ± 3.9780.80 ± 4.26 Mean (± SD) attitude scores of nursing students in the intervention group A two-way repeated-measures ANOVA was conducted to evaluate the impact of the educational intervention on nursing students’ performance scores over time, with Group (Intervention vs. Control) as the between-subjects factor and Time (pre-test vs. post-test) as the within-subjects factor. The assumption of sphericity was met, as indicated by Mauchly’s test. The analysis revealed a significant main effect of Time,)F = 317.51, p < 0.001, partial η² = 0.83(, indicating that performance scores changed significantly from pre-test to post-test across both groups. More importantly, a significant Group × Time interaction was found with a large effect size, (F = 251.54, p < 0.001, partial η² = 0.79), demonstrating that the pattern of change over time differed substantially between the intervention and control groups. Post-hoc analyses with Bonferroni adjustment confirmed that while the two groups did not differ at pre-test (p > 0.05), the intervention group achieved significantly higher performance scores at post-test compared to the control group (p < 0.001). These results demonstrate the strong and meaningful impact of the educational intervention on improving students’ performance (Table 3).Table 3Mean (± SD) performance scores in the control and intervention groupsTestControl group (Mean ± SD)Intervention group (Mean ± SD)Pre-test42.47 ± 4.3241.88 ± 4.03Post-test43.44 ± 4.9058.57 ± 3.35 Table 3Mean (± SD) performance scores in the control and intervention groupsTestControl group (Mean ± SD)Intervention group (Mean ± SD)Pre-test42.47 ± 4.3241.88 ± 4.03Post-test43.44 ± 4.9058.57 ± 3.35 Mean (± SD) performance scores in the control and intervention groups As shown in Table 4, the within-group comparison using a paired t-test indicated a significant increase in mean performance scores from pre-test to post-test (p < 0.001). Test assumptions of normality (Kolmogorov–Smirnov test) were satisfied (p > 0.05). The effect size was large (Cohen’s d = 4.09), confirming a substantial improvement in performance following the intervention.Table 4Mean (± SD) performance scores of participants in the intervention groupNo.ItemPre-test (Mean ± SD)Post-test (Mean ± SD)1Before administering medication, I confirm the patient’s identity by checking the wristband and asking their name.3.05 ± 0.994.28 ± 0.512I prepare medications for more than one patient at the same time.2.65 ± 1.134.34 ± 0.763I label the medication container with the patient’s name and room number.2.51 ± 1.144.11 ± 0.714I carry the medication administration record with me while preparing medications.3.26 ± 0.914.42 ± 0.795I label syringes and infusions with the drug name, patient name, and room number.3.08 ± 1.174.37 ± 0.646I administer medication prepared by another nurse or student.3.05 ± 0.993.97 ± 0.897I double-check dose calculations (especially for pediatric patients).2.77 ± 1.054.02 ± 0.858Before administering high-risk medications, I check the normal dosage range.2.20 ± 1.023.80 ± 0.869Before administering medication, I review the patient’s allergies.2.34 ± 1.023.71 ± 0.7510I prepare and administer medications under direct supervision of the instructor.2.71 ± 1.224.02 ± 0.7411If I suspect a medication error, I immediately report it to the instructor.3.74 ± 0.914.48 ± 0.5612I verify unclear or illegible medication orders before implementation.3.80 ± 0.834.60 ± 0.4913I prepare medications in a calm environment without distractions.3.17 ± 1.224.60 ± 0.4914After administration, I dispose of remaining medications according to protocol.3.14 ± 1.064.11 ± 0.67Total scoremean ± SD41.88 ± 4.0358.57 ± 3.35 Table 4Mean (± SD) performance scores of participants in the intervention groupNo.ItemPre-test (Mean ± SD)Post-test (Mean ± SD)1Before administering medication, I confirm the patient’s identity by checking the wristband and asking their name.3.05 ± 0.994.28 ± 0.512I prepare medications for more than one patient at the same time.2.65 ± 1.134.34 ± 0.763I label the medication container with the patient’s name and room number.2.51 ± 1.144.11 ± 0.714I carry the medication administration record with me while preparing medications.3.26 ± 0.914.42 ± 0.795I label syringes and infusions with the drug name, patient name, and room number.3.08 ± 1.174.37 ± 0.646I administer medication prepared by another nurse or student.3.05 ± 0.993.97 ± 0.897I double-check dose calculations (especially for pediatric patients).2.77 ± 1.054.02 ± 0.858Before administering high-risk medications, I check the normal dosage range.2.20 ± 1.023.80 ± 0.869Before administering medication, I review the patient’s allergies.2.34 ± 1.023.71 ± 0.7510I prepare and administer medications under direct supervision of the instructor.2.71 ± 1.224.02 ± 0.7411If I suspect a medication error, I immediately report it to the instructor.3.74 ± 0.914.48 ± 0.5612I verify unclear or illegible medication orders before implementation.3.80 ± 0.834.60 ± 0.4913I prepare medications in a calm environment without distractions.3.17 ± 1.224.60 ± 0.4914After administration, I dispose of remaining medications according to protocol.3.14 ± 1.064.11 ± 0.67Total scoremean ± SD41.88 ± 4.0358.57 ± 3.35 Mean (± SD) performance scores of participants in the intervention group Results-level evaluation After the 4-month follow-up period, the self-reported data indicated that the control group reported one medication error and two near-miss incidents, whereas the intervention group reported only one near-miss medication error. The limited number of reported events necessitates cautious interpretation of these findings. Nevertheless, the lower incidence of reported errors in the intervention group suggests a potential positive effect of the training on medication safety awareness and practices, though the self-reported nature of the data should be considered when generalizing these results. To enhance the reliability and accuracy of the self-reported data, all reports were cross-verified by the clinical instructors who had directly supervised the students during their clinical rotations. This verification process involved reviewing the students’ clinical records and comparing the self-reports with the instructors’ own observational assessments. Discussion The primary objective of the present study was to design, implement an educational intervention, and evaluate its effectiveness in the domains of knowledge, attitude, and performance of nursing students regarding the prevention of medication errors. The obtained results indicate the significant effectiveness of the intervention in improving these three components. The effectiveness of the educational program can be interpreted through the lens of several educational theories. According to Kolb’s experiential learning theory, meaningful learning occurs when learners actively engage in experiences, reflect on them, conceptualize lessons learned, and apply them to new situations [15]. In this study, scenario-based discussions and problem-solving activities provided such experiential opportunities, allowing students to connect theoretical knowledge about medication safety with real clinical contexts. This active participation likely promoted deeper understanding, long-term retention, and the ability to transfer learning to practice. Moreover, the results can also be explained through Bandura’s self-efficacy theory, which emphasizes that individuals’ belief in their capability to perform specific tasks significantly influences their behavior [16]. The structured and supportive learning environment in our program helped increase students’ confidence in medication administration and error prevention. As their perceived competence grew, so did their willingness to report and prevent medication errors. This aligns with the observed improvements in attitude and performance, suggesting that the intervention not only enhanced knowledge but also empowered students to act safely and responsibly. The high level of student satisfaction with the educational course reflects the quality of the content and delivery method, as well as the positive psychological and professional impact of the educational intervention on participants. Such satisfaction can enhance students’ motivation to engage in future courses, improve their readiness to adopt patient safety concepts, and strengthen their sense of professional responsibility—factors that ultimately contribute to fostering a culture of safety within healthcare settings. Previous studies, such as Deswal et al. (2024), have also reported high student satisfaction with medication error prevention courses, confirming the quality and practical relevance of these programs [17]. This alignment of findings underscores the importance of learner satisfaction as a key indicator of educational success and as a driver for continuous improvement in educational processes. Furthermore, high student satisfaction can encourage policymakers and educational managers to institutionalize such training programs formally within nursing curricula. The significant improvement in knowledge scores observed in the intervention group, sustained both immediately after the training and during follow-up, highlights the effectiveness of structured educational programs in enhancing nursing students’ understanding of medication error prevention principles. This sustained increase suggests that the training facilitated meaningful and lasting learning rather than a temporary knowledge boost. Such outcomes reinforce the value of targeted educational interventions in nursing education, aligning with prior research that emphasizes the role of focused training in enhancing nurses’ competence in medication management and safety [14]. At post-test 2, the mean knowledge score of the intervention group slightly declined, whereas the control group showed a small increase; however, these changes were not statistically significant. The observed pattern in the intervention group may be explained by partial knowledge decay over time, the absence of reinforcement or review sessions, or possible testing fatigue. Conversely, the minor increase in the control group might be attributed to informal exposure to related content during clinical practice or peer discussions. Despite these slight fluctuations, the consistently higher scores of the intervention group across both post-tests confirm the effectiveness and relative sustainability of the educational intervention. Accurate calculation of intravenous drug flow rates is crucial for safe patient care, and weaknesses in the calculation skills of nursing students can be associated with serious clinical outcome [18]. The observed low pre-test scores in both groups highlight a critical gap in foundational drug calculation competencies, consistent with previous studies indicating that many nursing students struggle with this skill. For instance, the results of the research by Elonen et al. (2022) showed that although all students (99%) correctly performed calculations related to tablets, about 29% of students did not answer drug calculations related to liquid medications correctly [19]. Encouragingly, the educational intervention in this study markedly enhanced students’ drug calculation abilities, as reflected by the significant post-test improvements in the intervention group compared to controls. This confirms the effectiveness of targeted training in addressing core skill deficits. These findings are consistent with the results of previous research. For example, the research by Sayadi et al. (2021) showed that a drug dosage calculation training program for nursing students can lead to reduced math anxiety and improved drug dosage calculation skills [20]. The results of the study by Karabağ Aydin and Dinç (2017) also showed that web-based training on drug calculations has a positive effect on the drug calculation skills of nursing students [21]. Furthermore, another study found that a clinical skills workshop led to an increase in the mean drug dosage calculation scores of nursing students [22]. These cumulative findings emphasize the necessity of incorporating focused and practical drug calculation training within nursing education to mitigate medication errors and enhance patient safety. The findings of the present study, which indicate the persistence of the intervention’s effect until the second post-test, support the view that active and practical learning has greater effectiveness and retention. The present findings show that with proper intervention design, not only an immediate and short-term effect but also a lasting effect can be expected. The sustainability of the intervention’s effect is a significant result that has not been examined in previous studies, and this highlights the importance of the continuity and quality of educational programs. Overall, the increase in the scores of the intervention group students in the present study indicates that proper educational planning can play an important role in preventing this type of error. Attitude plays a key role in professional behaviors, and by understanding individuals’ attitudes, predicting and controlling their behaviors becomes easier [23]. Given that the attitude toward error reporting is considered one of the key factors in preventing the recurrence of medication errors, it can be concluded that educational interventions—particularly those addressing psychological and cultural dimensions—play an important role in fostering a culture of medication error prevention [24]. The results of a systematic review showed that only 39% of nurses with medication errors reported their errors [25]. One of the prominent results of this study was the improvement in the attitude of students in the intervention group towards reporting and dealing with medication errors. While no significant changes were observed in the control group. This finding is consistent with previous research which has shown that training can lead to an improvement in the attitude of nursing students towards preventing medication errors [12, 26]. Also, the results of the study by Salehi et al. (2023) showed that active training and learning can enhance error reporting in nurses [27]. The improvement in students’ attitude towards error reporting indicates the success of the intervention in reducing the “blame culture.” In many educational and therapeutic environments, fear of punishment and judgment is a serious barrier to error reporting [24]. Training can reduce these mental barriers. Many negative attitudes such as fear of blame, fear of damage to future careers, or feelings of incompetence among peers decreased after the intervention. This indicates that the education not only transferred knowledge but also led to a deep change in beliefs and feelings. The persistence of the positive effects of the intervention in the second post-test phase (several months later) suggests that effective educational programs can contribute to long-term attitude correction, which is a key goal in professional education. The improvement in students’ clinical performance in the intervention group indicates that education grounded in active and contextualized learning can effectively translate theoretical knowledge into safe clinical practices. This shift from “knowing” to “doing” reflects the essential goal of competency-based nursing education, where learning outcomes are measured not only by cognitive gains but also by behavioral changes. The enhanced adherence to patient identity verification and medication preparation principles demonstrates how structured educational interventions can directly influence patient safety practices. The absence of significant improvement in the control group suggests that experiential learning and reflective exercises, rather than routine experience alone, are critical for shaping safe medication practices. These findings align with previous research emphasizing that case-based and scenario-driven training strengthens students’ ability to apply theoretical knowledge in real clinical contexts [28]. Consequently, integrating such interactive approaches into nursing curricula may help bridge the persistent gap between knowledge acquisition and clinical performance in medication error prevention. Although data related to the actual occurrence of errors after 4 months require cautious interpretation due to the low number of cases, nevertheless, the number of error occurrences in the intervention group was lower than in the control group. This outcome reinforces the notion that effective learning in medication error prevention extends from knowledge acquisition to practical application in real clinical settings. This finding aligns with studies that have found interprofessional training on medication errors effective in reducing error rates among nurses and physicians [29]. Implications for education and practice The findings of this study highlight the importance of structured, needs-based educational interventions in improving nursing students’ knowledge, attitudes, and clinical performance regarding medication safety. Incorporating such programs into nursing curricula and in-service training can help foster safer medication practices and reduce preventable errors. Educational managers and policymakers are encouraged to integrate medication safety modules as a standard component of nursing education and continuous professional development. Furthermore, this study provides a practical, evidence-based framework that can inform the design of future educational programs and support curriculum development focused on patient safety competencies. Limitations This study has several limitations that should be considered when interpreting the findings. First, the study was conducted in a single nursing school with a relatively limited sample size, which may restrict the generalizability of the results. Second, the follow-up period was limited to four months; therefore, the long-term sustainability of the educational effects remains unclear. Third, the collected data were based on self-reported measures, which may be subject to response bias. It is recommended that future studies be conducted with larger sample sizes, across multiple institutions, with longer follow-up periods, and include performance evaluations in real clinical settings to confirm and expand upon the findings of this study. Future research could also explore implementing the program in digital or simulation-based learning environments and evaluate the long-term retention of learning outcomes. Conclusion The present study demonstrated that a structured needs-based educational program can lead to significant improvements in nursing students’ knowledge, attitudes, and performance regarding medication error prevention. The persistence of the intervention’s effects during the four-month follow-up period indicates relative sustainability of learning outcomes. The program’s effectiveness in enhancing drug calculation skills and fostering positive attitudinal changes toward error reporting underscores the value of implementing active, scenario-based learning methods in nursing education. Given the satisfactory results and positive student reception, incorporating such structured educational programs into nursing curricula could represent an effective step toward enhancing patient safety and institutionalizing a culture of medication error prevention. Future studies with longer follow-up periods across diverse educational settings could more precisely examine the role of such interventions in reducing medication errors in actual clinical environments. Supplementary Information Supplementary Material 1. Supplementary Material 1.