Graphical abstract

Introduction

Background/rationale

Medication errors remain a critical threat to patient safety, particularly in hospital settings where patients are often clinically vulnerable [1]. Errors in drug preparation and dose calculation are especially concerning, as they can directly compromise treatment efficacy and increase the risk of adverse outcomes [1,2]. Nursing students are known to commit errors in drug preparation and calculation, and systematic reviews have reported prevalence rates ranging from 1.1% to 6% during clinical rotations [2].

In many healthcare systems, nurses are the primary professionals responsible for medication preparation, dose calculation, treatment administration, and monitoring patients for both therapeutic and adverse effects. This central role places nurses at the forefront of patient safety but simultaneously exposes them to a heightened risk of calculation errors. Contributing factors include heavy workloads, limited access to advanced technologies, and continued reliance on manual computation [1]. Strengthening drug-calculation competencies among nursing students is therefore essential to ensuring safe practice as they transition into the professional workforce [2].

Previous studies have demonstrated that nursing students frequently struggle with drug calculations, with errors influenced not only by mathematical ability but also by psychological factors such as math anxiety and self-efficacy [2,3]. These challenges are particularly pronounced in healthcare systems where nurses often lack access to electronic calculation aids or systematic double-checking mechanisms that are routinely available in high-income countries. Despite the importance of this issue, evidence remains limited regarding the readiness of Palestinian nursing students to perform accurate drug calculations and the factors that influence their competence.

Objectives

It was hypothesized that drug-calculation competence among nursing students would be positively associated with grade point average (GPA), prior training, confidence, and self-efficacy, and negatively associated with math anxiety. Accordingly, this study was undertaken to evaluate the ability of Palestinian nursing students to perform accurate drug calculations and to identify the academic and psychological factors influencing their performance. By addressing this critical competency, the study seeks to inform nursing education and support strategies aimed at reducing medication errors and enhancing patient safety within the Palestinian healthcare system.

Methods

Ethics statement

The study adhered to national and international ethical principles, including the Declaration of Helsinki, and was approved by the Institutional Review Board (IRB) of An-Najah National University (NNU-IRB: # March-17/205). Participation was voluntary, and written informed consent was obtained from all participants prior to enrollment.

Study design

This cross-sectional study was conducted among near-graduation nursing students (third and fourth academic years) in Palestine. Nursing education in the country has expanded substantially in recent decades, with accredited programs offered across universities and colleges. The Bachelor of Nursing degree comprises 137 credit hours completed over 4 years [4], integrating biomedical sciences, clinical training, and specialized courses in pharmacology, patient safety, and ethics [5]. Students receive structured training in drug calculations through pharmacology coursework and laboratory practice, including dosage computations, infusion and drip-rate calculations, and unit conversions (Supplement 1). Upon graduation, nurses are licensed to practice across all levels of healthcare and represent the largest professional group delivering direct patient care. The study adhered to the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) statement (https://www.strobe-statement.org).

Settings

The study was conducted across multiple accredited nursing schools in the West Bank, Palestine, to ensure representativeness and diversity of educational experiences. Students were recruited from An-Najah National University, the Arab American University, Al-Quds University, Birzeit University, Nablus University for Vocational and Technical Education, Palestine Ahliya University, Hebron University, Palestine Polytechnic University, Bethlehem University, and Zaytouna University of Science and Technology. Data collection took place during the 2024–2025 academic year, between September 2024 and May 2025.

Participants

The study population comprised nursing students enrolled in accredited programs in Palestine. Eligibility was restricted to students approaching graduation (third and fourth academic years), ensuring that participants had completed most coursework and clinical training and were representative of the cohort soon to enter professional practice. Recruitment from multiple institutions enhanced generalizability and captured the diversity of educational experiences nationwide. Participation was voluntary, and only students who provided written informed consent were included. Exclusion criteria were applied to preserve sample integrity and ensure that outcomes reflected student competencies rather than professional experience. Students who declined consent, had already graduated, were enrolled in postgraduate education, or had commenced nursing practice were excluded. These criteria minimized potential confounding and ensured the assessment focused on students at the threshold of graduation.

Variables

The primary outcome was drug-calculation competence. The main exposures/predictors were prespecified academic and psychological factors. Potential confounders and covariates were defined a priori as sex (male/female), age (years), academic year (third vs. fourth), and self-perceived learning style (auditory/visual/kinesthetic), based on their plausible associations with both academic or psychological predictors and calculation performance. No effect modifiers were prespecified, and no formal interaction terms were planned. Diagnostic criteria were not applicable because this was not a diagnostic study and no clinical diagnoses were used as study variables.

Data sources/measurement

Data were collected using a structured, self-administered questionnaire developed for this study following a comprehensive literature review on medication errors and drug-calculation competence. The instrument included demographic and academic variables (sex, age, academic year, GPA, preferred learning style, and self-rated mathematical performance), prior training in drug calculations, and self-rated knowledge and confidence in drug preparation. Psychological factors were assessed using the 9-item Abbreviated Math Anxiety Scale and the 10-item General Self-Efficacy Scale (Supplement 2) [6,7], both validated and administered in Arabic. Internal consistency in the current sample was high, with Cronbach’s α values of 0.87 and 0.84, respectively.

Drug-calculation competence was measured using 15 clinical scenarios requiring computation of doses, volumes, and infusion rates. Each correct response was awarded 1 point (raw score range, 0–15), with scores linearly transformed to a 0–100 scale. Scenarios were reviewed by subject-matter experts, piloted among 35 students, and demonstrated good reliability (Cronbach’s α=0.81). The main exposure variables included academic and psychological factors: GPA, self-rated mathematical performance (5-point Likert scale), receipt of dedicated drug-calculation training (yes/no), and self-rated knowledge and confidence in drug preparation (4-point Likert scale), as well as total scores from the Abbreviated Math Anxiety Scale (9 items; range, 9–45) and the General Self-Efficacy Scale (10 items; range, 10–40). Additional covariates included sex, age, academic year, and self-perceived learning style. The final questionnaire is provided in Supplement 3 (Dataset 1).

Bias

To minimize selection bias and enhance representativeness, near-graduation (third- and fourth-year) nursing students were recruited from multiple accredited nursing schools across the West Bank. To reduce information and measurement bias, data were collected using a structured questionnaire informed by prior literature and incorporating validated Arabic versions of the Abbreviated Math Anxiety Scale and the General Self-Efficacy Scale, with internal consistency assessed in the current sample. Drug-calculation competence was measured using expert-reviewed, piloted, and reliable scenario-based items.

Study size

The target population comprised approximately 200 nursing students in their third and fourth academic years at each of the 10 accredited universities in Palestine, yielding an estimated source population of 2,000 students. To ensure adequate power for correlation and regression analyses, the sample size was calculated using a standard formula for cross-sectional studies with a 95% confidence level, a 5% margin of error, and an assumed response distribution of 50%. A sample size calculator (www.raosoft.com) indicated a minimum required sample of 323 students. To enhance reliability and reduce the risk of type II error, oversampling was employed. A total of 702 students were invited, of whom 562 completed the study instrument, providing a sufficiently large sample to support multivariable analyses.

Statistical methods

Statistical analyses were performed using IBM SPSS Statistics ver. 22.0 (IBM Corp.). Continuous variables were assessed for normality using descriptive statistics, histograms, and the Shapiro-Wilk test. Associations between categorical characteristics and drug-calculation scores were examined using independent-samples t-tests or one-way analysis of variance, as appropriate. Pearson’s correlation coefficients were calculated for continuous variables. Variables demonstrating significant bivariate associations were entered into multiple linear regression models to identify independent predictors of performance. Regression diagnostics included assessment of collinearity, model fit, and residual distribution. Results were reported as unstandardized coefficients (B), standard errors (SE), standardized coefficients (β), t-statistics, and P-values, with collinearity evaluated using tolerance and variance inflation factor.

Missing data (7.6%) were handled using listwise deletion, and sensitivity analyses confirmed no meaningful differences between excluded and retained participants. Recruitment was stratified across institutions, with an average cluster size of approximately 56 students per university; however, analyses were conducted at the individual level to allow rigorous evaluation of demographic, academic, and psychological predictors. Institution-specific comparisons were not reported to avoid ethical concerns and unauthorized benchmarking. Statistical significance was defined as a 2-sided P-value <0.05.

Results

Participants

Of the 702 nursing students invited to participate, 562 completed the study instrument and were included in the final analysis (Fig. 1). Most respondents were female (71.2%; n=400), with a mean age of 23.0±1.5 years (Table 1). Self-perceived learning styles were distributed as auditory (32.0%; n=180), visual (32.2%; n=181), and kinesthetic (35.8%; n=201). The majority of students (68.7%; n=386) reported prior training in drug calculations. Self-rated knowledge and confidence in drug preparation varied considerably, ranging from 22.2% (n=125) at the lowest level to 23.8% (n=134) at the highest level (Table 1).

Math anxiety

Overall, the mean score on the Abbreviated Math Anxiety Scale was 26.7±4.3 (95% confidence interval [CI], 26.4–27.1). The most anxiety-provoking situations were mathematics examinations and unexpected “pop” quizzes, with approximately 20% of students reporting high levels of anxiety in these contexts (Table 2). Observing algebraic problem-solving and completing difficult homework assignments also elicited moderate to high anxiety among substantial proportions of respondents.

General self-efficacy

The mean score on the General Self-Efficacy Scale was 24.9±3.5 (95% CI, 24.6–25.1). Across most scale items, nearly half of respondents expressed strong confidence in their ability to solve difficult problems, achieve goals despite opposition, persist toward objectives, and cope effectively with unexpected challenges (Table 3).

Associations between student variables and calculation scores

The mean drug-calculation score was 79.8%±11.0% (95% CI, 78.9–80.7). Female students achieved significantly higher scores than male students (Table 4). GPA showed a positive association with calculation performance. Self-rated mathematical ability demonstrated a clear gradient, with higher self-assessments corresponding to better performance. Prior training in drug calculations and higher self-rated confidence in drug preparation were both associated with improved scores. In contrast, math anxiety was negatively correlated with calculation performance, whereas general self-efficacy showed a positive correlation.

Multivariate linear regression identified several independent predictors of drug-calculation performance. Female students scored significantly higher than male students, and GPA remained strongly associated with improved performance after adjustment (Table 5). Self-rated mathematical ability emerged as the most influential predictor. Prior training in drug calculations and greater confidence in drug preparation also contributed significantly to higher scores. Math anxiety was independently and negatively associated with performance, whereas general self-efficacy demonstrated a positive independent association.

Discussion

Key results

This cross-sectional study assessed drug-calculation competence among 562 Palestinian nursing students in their third and fourth academic years. Overall proficiency was moderate, with performance significantly influenced by sex, GPA, self-rated mathematical ability, prior training, confidence in drug preparation, math anxiety, and self-efficacy. Multivariate regression analysis confirmed self-rated mathematical ability as the strongest independent predictor, followed by GPA, self-efficacy, and prior training.

Interpretation

This study highlights the interplay between academic achievement, self-perceived mathematical competence, prior training, and psychological factors in shaping drug-calculation skills. Self-rated mathematical performance emerged as the strongest predictor, consistent with evidence suggesting that numeracy confidence, rather than raw computational ability alone, plays a decisive role in medication safety [8-10]. GPA was positively associated with performance, reinforcing prior findings that general academic success supports clinical competence [11]. Structured training also improved scores, aligning with intervention studies demonstrating that simulation-based learning and repeated practice enhance calculation accuracy and reduce medication errors [10].

Psychological factors were also critical. Math anxiety was negatively correlated with performance, echoing previous reports that anxiety impairs engagement, working memory, and calculation accuracy [12,13]. In contrast, self-efficacy was positively associated with performance, consistent with evidence that confidence promotes persistence, attention, and precision during complex tasks [14].

Taken together, these findings indicate that drug-calculation competence is a multidimensional construct shaped by academic achievement, targeted training, and psychological readiness. Integrated educational strategies that build technical skills while reducing anxiety and strengthening self-efficacy may therefore be essential for preparing nursing students for safe medication administration.

Comparison with previous studies

The findings of this study align with international evidence showing that drug-calculation competence among nursing students is heterogeneous and influenced by numeracy confidence, academic performance, and targeted training [8-10,14]. Multicenter studies have reported substantial variability in calculation accuracy across institutions, with foundational mathematical skills and curricular emphasis contributing to performance differences, consistent with the observed gradients by GPA and self-rated mathematical ability [8,15]. European data further highlight error clustering around unit conversions and infusion-rate calculations, underscoring the need for structured and repeated practice within pharmacology and clinical modules [8].

Intervention studies similarly confirm that focused training improves performance, mirroring the regression results observed for dedicated drug-calculation instruction in this study [10]. Randomized and quasi-experimental trials of simulation-based training have demonstrated significant gains in both accuracy and confidence, particularly when learning is iterative and integrated longitudinally across curricula [9,15].

Psychological mechanisms have also been well documented. Math anxiety is known to impair working memory and calculation reliability, while self-efficacy predicts persistence and effective error recovery [9]. Interventions that reduce anxiety and foster self-efficacy—through supportive assessment, feedback, resilience-building, and mastery experiences—have consistently been shown to improve numeracy outcomes [8,10,12-14]. Collectively, the global literature supports a multidimensional model in which mathematical skills and academic achievement provide the foundation, targeted training builds proficiency, and psychological readiness governs reliability under clinical pressure. The present findings are consistent with this model and add important context from Palestine, where calculation accuracy has heightened implications for patient safety.

Limitations

This study has several limitations. First, its cross-sectional design precludes causal inference; longitudinal research is needed to determine whether interventions targeting math anxiety or self-efficacy lead to sustained improvements in calculation performance. Second, reliance on self-reported measures (e.g., mathematical performance, confidence, and learning style) introduces potential recall and social desirability bias, and subjective ratings may not fully reflect actual competence. Third, drug-calculation skills were assessed using 15 scenario-based items administered under classroom conditions, which may not capture the full complexity of real-world clinical tasks or the pressures of patient care. Fourth, the study was conducted exclusively among nursing students in Palestine, which may limit generalizability to settings with different curricula, clinical exposure, or cultural contexts. Fifth, unmeasured confounders, such as prior clinical experience or institutional teaching quality, may have influenced outcomes. Finally, competence was evaluated through educational assessment rather than direct clinical observation, which may overestimate performance compared with real-world practice.

Generalizability

This study highlights implications for nursing education and practice beyond the local context. Although conducted in Palestine, the identified predictors of drug-calculation performance—academic achievement, self-rated mathematical competence, prior training, math anxiety, and self-efficacy—mirror findings from diverse international settings, supporting generalizability to other low- and middle-income countries.

Curricular reforms should prioritize structured, repeated training in drug calculations through scenario-based exercises and simulation. In parallel, educational interventions designed to reduce math anxiety and foster self-efficacy may further enhance performance. At the policy level, accreditation bodies and nursing schools should recognize numeracy competence as a core educational outcome, ensuring that graduates are adequately prepared to deliver safe medication care, reduce errors, and strengthen patient safety.

Suggestions

Future research should examine the longitudinal development of drug-calculation competence and rigorously test interventions such as structured training, simulation, and resilience-building programs aimed at reducing math anxiety. Cross-cultural studies may clarify whether observed predictors are universal or context specific. For nursing education, repeated scenario-based training and strategies that explicitly foster self-efficacy appear essential, particularly in healthcare systems where reliance on manual calculations increases patient-safety risks. Supportive learning environments, mentorship, and targeted workshops may further strengthen both competence and confidence.

Conclusion

This study demonstrated that drug-calculation competence among Palestinian nursing students is influenced by academic achievement, prior training, confidence, math anxiety, and self-efficacy. Self-rated mathematical ability emerged as the strongest predictor of performance. These findings underscore the need for nursing curricula that combine structured technical training with strategies to reduce math anxiety and strengthen self-efficacy, thereby ensuring that graduates are adequately prepared to administer medications safely.

Supplementary materials

Fig. 1.

Flow diagram of the participant recruitment process.

References

• 1. Kuitunen S, Niittynen I, Airaksinen M, Holmstrom AR. Systemic causes of in-hospital intravenous medication errors: a systematic review. J Patient Saf 2021;17:e1660-e1668. https://doi.org/10.1097/PTS.0000000000000632 ArticlePubMed
• 2. Stolic S, Ng L, Southern J, Sheridan G. Medication errors by nursing students on clinical practice: an integrative review. Nurse Educ Today 2022;112:105325. https://doi.org/10.1016/j.nedt.2022.105325 ArticlePubMed
• 3. Belli M, Uslu N. Investigation of the relationship between mathematics anxiety of nursing students and their self-efficacy in drug administration in children. Nurse Educ Pract 2025;83:104254. https://doi.org/10.1016/j.nepr.2025.104254 ArticlePubMed
• 4. Shawahna R. Agreement of Palestinian nursing students with recommendations to eliminate epilepsy stigma and change perception of the general public about epilepsy: a cross-sectional study. Epilepsy Behav 2020;109:107126. https://doi.org/10.1016/j.yebeh.2020.107126 ArticlePubMed
• 5. Shawahna R. Using a mixed method to develop consensus-based aims, contents, intended learning outcomes, teaching, and evaluation methods for a course on epilepsy for postgraduate or continuing education in community health nursing programs. BMC Med Educ 2021;21:572. https://doi.org/10.1186/s12909-021-03001-2 ArticlePubMedPMC
• 6. Hopko DR, Mahadevan R, Bare RL, Hunt MK. The Abbreviated Math Anxiety Scale (AMAS): construction, validity, and reliability. Assessment 2003;10:178-182. https://doi.org/10.1177/1073191103010002008 ArticlePubMed
• 7. Jerusalem M, Schwarzer R. Die Entwicklung des Selbstkonzepts in schulischen Bezugsgruppen: Eine dynamische Mehrebenenanalyse [The development of self-concept in school reference groups: a dynamic multilevel analysis]. In: Luer G, editor. Bericht über den 33. Kongress der Deutschen Gesellschaft für Psychologie in Mainz 1982 [Report on the 33rd Congress of the German Psychological Society in Mainz 1982] [Internet]. Hogrefe; 1983 [cited 2025 Nov 10]. p. 639-642. Available from: https://www.psychauthors.de/psychauthors/index.php?wahl=forschung&uwahl=psychauthors&uuwahl=p00298RS
• 8. Elonen I, Salminen L, Brasaite-Abrome I, Fuster P, Kukkonen P, Leino-Kilpi H, Loyttyniemi E, Noonan B, Stubner J, Svavarsdóttir MH, Thorsteinsson H, Koskinen S. Medication calculation skills of graduating nursing students within European context. J Clin Nurs 2022;31:548-558. https://doi.org/10.1111/jocn.15908 ArticlePubMed
• 9. McMullan M, Jones R, Lea S. Math anxiety, self-efficacy, and ability in British undergraduate nursing students. Res Nurs Health 2012;35:178-186. https://doi.org/10.1002/nur.21460 ArticlePubMed
• 10. Pol-Castaneda S, Carrero-Planells A, Moreno-Mulet C. Use of simulation to improve nursing students’ medication administration competence: a mixed-method study. BMC Nurs 2022;21:117. https://doi.org/10.1186/s12912-022-00897-z ArticlePubMedPMC
• 11. Nemati-Vakilabad R, Hashemian M, Mojebi MR, Mirzaei A. Assessing self-perceptions related to safe medication management among nursing students: an analytical cross-sectional study in Iran. BMC Nurs 2025;24:207. https://doi.org/10.1186/s12912-025-02859-7 ArticlePubMedPMC
• 12. McKenna L, Johnston J, Cross R, Austerberry J, Mathew T, McKenzie G. Mathematics anxiety and associated interventions in nursing: a scoping review. Nurse Educ Today 2022;112:105335. https://doi.org/10.1016/j.nedt.2022.105335 ArticlePubMed
• 13. Austerberry J, Mathew T, Johnston J, Murphy M, Safari K, Murray C, McKenna L. Mathematics anxiety and undergraduate nursing students: a mixed methods study. Nurs Open 2024;11:e70082. https://doi.org/10.1002/nop2.70082 ArticlePubMedPMC
• 14. Owegi R, Burdick K, Cannon E, McQuiston L, Arvin S. Medication math dosage assessment anxiety in undergraduate nursing students: a systematic review. J Prof Nurs 2021;37:735-740. https://doi.org/10.1016/j.profnurs.2021.05.003 ArticlePubMed
• 15. Gregory L, Villarosa AR, Ramjan LM, Hughes M, O’Reilly R, Stunden A, Daly M, Raymond D, Fatayer M, Salamonson Y. The influence of mathematics self-efficacy on numeracy performance in first-year nursing students: a quasi-experimental study. J Clin Nurs 2019;28:3651-3659. https://doi.org/10.1111/jocn.14963 ArticlePubMed

Figure & Data

References

Citations

Citations to this article as recorded by