Intelligent, adaptive, and interactive technologies powered by Artificial Intelligence (AI) greatly enhance the power and precision of technologies that transform the way health professionals prepare for and respond to complex clinical situations. Challenges in traditional teaching methods used in nursing mean learners are presented with insufficient opportunities to engage with diverse and changing clinical environments. ¹ This means nursing learners, in particular, do not achieve the clinical judgment and decision-making capabilities that are essential to ensure patient safety. The global community of educators expresses concern around

clinical judgment and decision-making, which encompasses the nursing process in reasoning and recognizing clinical cues, as well as arriving at informed decisions. ² Failure to do so has been shown to lead to delayed recognition of the deterioration of patients and to prioritize care inadequately. There is a strong focus then on developing the clinical judgment and decision making of nursing learners through educational innovations that create repeated practice, provide feedback, and present learners with a variety of challenging, realistic, and clinically authentic situations. It has been shown that certain technologies enabled by AI have the capacity to support the development of clinical reasoning, but their utility depends on how they are incorporated within the framework of other teaching and learning practices. ³

Advancements in clinical complexity and technology along with changes in the demands on the healthcare workforce are raising concerns about the readiness of nursing school graduates. Nurses need to be able to manage and act upon ambiguous information to make decisions quickly in the evolving and fast-paced nature of modern healthcare. ⁴ The nurses of today need skills and knowledge completely different from those of the past. Although competency-based nursing education becomes more common, many undergraduates of nursing schools are still unable to apply education theories to the practice of making clinical judgments. ⁵ Traditional teaching does not promote application of education to the practice of nursing in an ambiguous context. The limitations of case studies, such as variability, meaningful feedback, and time availability, are driving interest in the use of AI technology to aid reflection and to design clinical problems. AI case studies provide flexibility to incorporate computational pedagogy of adaptive case-based education and frameworks to provide contextual reasoning support, activity, and reflection-centered learning. ⁶

Research on AI in nursing and health education has grown globally, especially with the introduction of generative AI, intelligent tutoring systems, virtual simulations, and AI-enacted clinical case scenarios. Some findings suggest that AI-based learning environments engage students and improve their confidence, knowledge application, and clinical decision-making skills. However, the contemporary literature contains a fundamental debate on whether AI meaningfully supports the development of advanced cognitive skills, or, at most, simply enhances the efficiency of the task. ⁷ It is noted that the use of AI has improved clinical reasoning and decision-making in some empirical studies, while other studies have noted problematic dependence on AI, diminished critical thinking, and ethical issues, and have expressed concerns that students are not adequately evaluating information generated by AI. Recently, systematic reviews have documented a positive and fragmented view of the use of AI in education. There is not enough rigor and consistency to draw a definitive conclusion about the impact of different methods of assessment. It should not be assumed that AI fully replaces more advanced reasoning, and therefore the clinician's professional judgement, in the educational process. ⁸

AI-assisted case-based learning has some alignment with various existing educational frameworks from a theoretical standpoint. This includes constructivist learning theory, experiential learning theory, and cognitive apprenticeship theory. ⁹ According to constructivism, knowledge is constructed rather than simply provided to learners, which is illustrated in case-based learning with the presentation of clinical scenarios that require analysis, interpretation, and a course of action. Experiential learning theory is concerned with the development of professional competence with many clinical experiences, reflection, and feedback; AI-enhanced case-based learning adds to this by enabling students to work in active case scenarios and receive instantaneous responses, adjust their reasoning, and even test a hypothesis. Lastly, in cognitive apprenticeship theory, the development of a skill is aided by the presence of a guide and support, which AI may help with in providing case-based prompts, explanations, and feedback for answering clinical problems. Although AI-based systems help with some aspects of clinical problem-solving, there are theoretical limitations to the extent to which learning may substitute human clinical mentorship. ¹⁰

Challenges facing nursing education in South Asia, particularly in Pakistan, underscore the importance of the clinical reasoning and decision-making of nursing students. Restricted faculty, large student numbers, limited access to simulation, and variety in the availability of clinical opportunities negatively impact nursing students' exposure to different levels of patient complexity and the opportunity to make decisions under supervision.  ⁸ The disruptive events of the last two years have radically altered the education landscape throughout the world, and particularly so in South Asia and Pakistan. Traditional classroom teaching, clinical demonstrations, and discussions led by the instructor are the dominant forms of teaching in higher education in Pakistan, particularly in nursing, and this is the case even with limited published data on the use of advanced pedagogical approaches and Artificial Intelligence. The lack of contextual data means that there is uncertainty about whether the use of Artificial Intelligence would support and enhance existing learning approaches and improve the acquisition of competencies in the nursing profession in Pakistan. ¹¹

Analysis of existing literature shows important gaps in knowledge. For instance, in literature on AI in nursing education, the majority of studies have focused on perceptions, attitudes, academic outcomes and technological readiness, rather than objective assessment of clinical reasoning and decision making. Additionally, existing studies have focused on AI tools in an uncritical manner and have failed to differentiate AI-assisted case-based learning. ¹² Finally, the majority of studies have been conducted in technologically advanced education systems, thus curtailing the scope for application of research findings in developing countries, which lag behind in educational resources, digital networks, and faculty. Furthermore, due to small sample sizes, non-random allocation of study groups, and reliance on self-assessment, the effects of AI on education cannot be examined critically. These gaps in the literature beckon well-designed quasi-experimental studies to examine the effects of AI-based case learning on the development of clinical reasoning and decision making of nursing students. ¹³

This study is justified by the need for creative educational approaches that are grounded in data to prepare nursing students for advanced clinical practice. Case-based learning assisted by AI may provide important data on how traditional methods of teaching in undergraduate nursing programs can be improved and how important professional skills may be protected. Using AI-based case-based learning within a quasi-experimental study may help answer the riddle of how to use advanced technologies to enhance the current evidence base surrounding education. The results of this study may describe how AI-based learning technology may be integrated into nursing education; however, traditions of clinical judgment, ethical thinking, and the art of nursing may be preserved. ¹⁴

This research study implemented a quasi-experimental design to determine the influence of AI-based case-based learning on clinical reasoning and decision-making for undergraduate nursing students. The study was conducted in three nursing colleges in Lahore, Pakistan. Undergraduate nursing students in the nursing programs who had some knowledge of the theoretical aspect of the discipline as well as some clinical experience formed the study population. Sampling was based on the results of the G*Power calculator considering the objectives of the study, the power of the study, the significance level, and the effect size. A sampling size of 60 participants were selected. The sampling method of choice for this study was simple random sampling. This method ensures that all students have an opportunity for sampling and bias is as minimal as possible. Participants for the study were elected nursing students, while absent students as well as those that provided incomplete information were excluded. Data Collection Procedure Following approval of the ethics proposal and consent from the administration of the nursing colleges, data collection commenced. The researcher explained the study's goal and objectives to the participants. Written informed consent was obtained with reassurance to the students that their information would be kept confidential, participants had the right to refuse to participate, and that refusal would not affect their studies in any way. Before the AI-based case-based learning (AI-CBL) intervention, the control group participants were given a pre-assessment using a clinical reasoning and decision-making assessment tool. After the assessment, the participants engaged in guided analysis and problem-solving tasks through AI-CBL activities using provided clinical scenarios. After the intervention, participants completed a post-assessment using the same tool to evaluate their clinical reasoning and decision-making skills. The completed surveys were reviewed for completeness prior to coding and preparation for analysis. Data Analysis Procedure Data were entered and analyzed using Statistical Package for the Social Sciences (SPSS), Version 27. Demographic characteristics and study variables were analyzed using descriptive statistics. For clinical reasoning and decision-making scores, means and standard deviations were determined. For the analysis, conclusive research relies on the comparison of pre-intervention and post-intervention data. Based on data type and distribution, appropriate inferential statistical analyses were performed. Using a paired sample t-test, pre-test and post-test scores were analyzed for significance. Statistical significance was considered p <0.05. Results were interpreted to establish if AI-CBL improved clinical reasoning and decision-making skills of the undergraduate nursing students.

Demographic Analysis

The demographic findings indicated that 60 undergraduate nursing students involved in this study had a median age of 21 to 23 years (50.0%), a median age of 18 to 20 years (36.7%), and a median age of 24 years or older (13.3%). The female students constituted the majority of the sample (70.0%), with the male students accounting for the remaining 30.0% of the sample. At the level of academic year, the sample had a dominance of students in the second year (30.0%), followed by the second with 26.7% students level, and first year having 25.0% and fourth year students having 18.3% sample of participants. Participants were sampled from three nursing colleges in Lahore in a relatively well distributed manner, that is, College A (33.3%), College B (35.0%), and College C (31.7%). This shows that the study findings are applicable to the target population.

Table 1. Demographic Characteristics of Undergraduate Nursing Students (n=60)

Post-intervention clinical reasoning scores for the AI-based case-based learning intervention showed a substantial improvement for undergraduate students. The mean clinical reasoning score for the pre-intervention group was 58.42 ± 8.16, and for the post-intervention group it was 74.65 ± 7.92. The mean improvement for clinical reasoning was 16.23. The results of the paired sample t-test indicated that clinical reasoning scores post-intervention were significantly improved from the clinical reasoning scores pre-intervention (t = 12.48, p < 0.001). The results of the study indicated that AI-supported case-based learning improves students’ ability to analyze clinical information, identify important clues about patients, apply nursing knowledge and make better clinical judgments. The post-intervention improvement for the AI-based case-based learning was significant, and thus, it presents an opportunity to focus on using this approach to practice to enhance the cognitive aspects of professional nursing practice.

Table 2. Comparison of Clinical Reasoning Skills Before and After AI-Assisted Case-Based Learning Intervention (n=60)

Results indicated a marked improvement in students’ decision-making skills after the AI-assisted case-based learning activity. The mean decision-making score was 56.87 ± 9.24 prior to the learning activity, and was 72.93 ± 8.35 post learning activity score, reflecting an improvement by 16.06 points. Statistically, it was confirmed that this improvement was highly significant, t (11.76), p < 0.001. AI-based medical cases not only allowed students to practice evaluating a given clinical situation, but also allowed students to practice considering different alternative interventions and selecting appropriate nursing actions, with immediate feedback. AI-based case-based learning activities help graduate nursing students build their decision-making skills; this is evident through the observed improvement in the decision-making skills of nursing undergraduates.

Table 3. Comparison of Decision-Making Skills Before and After AI-Assisted Case-Based Learning Intervention (n=60)

Analysis of demographic factors showed that academic year and college affiliation had a clear association with progress made in clinical reasoning and decision making. There was a statistically significant relationship for the academic year (F=4.26, p=0.009), showing that students in one academic year had more progress than students in the other academic years. Similar findings were noted for the three participating colleges (F=3.74, p=0.030). Consideration of the effects of age (F=2.31, p=0.108) and gender (t=1.84, p=0.071), however, did not yield substantial findings. This shows that theatz responding to AI-based learning interventions in a case-based context may be more influenced by educational exposure and institutional factors as opposed to demographic factors.

Table 4. Association Between Demographic Characteristics and Improvement in Clinical Reasoning and Decision-Making Skills (n=60)

The use of AI-supported case-based learning improved nursing students' clinical reasoning skills, as measured by post-intervention clinical reasoning scores. Using AI to develop case-based learning environments likely improves students’ ability to analyze the information, recognize the important clues of the patient, and develop evidence-based judgments. Similar results in clinical reasoning and

decision-making were documented by AI-driven case learning. This is attributed to interactive case learning and the provision of feedback by the AI in the learning process.¹⁵The use of AI in learning platforms presents nursing students with many different and complex case-based learning scenarios which means students have to go beyond what is taught by the common frameworks of traditional simple case-based learning.

The use of case-based learning significantly improved undergraduate students' clinical reasoning skills, which supports the use of case-based learning in nursing education. The use of case-based learning encourages students to apply the taught concepts to practice, thus requiring students to exercise their thinking skills in a case rather than rote memorization. Many studies show that clinical reasoning for nursing students is improved by case-based learning.¹⁶ The use of case-based learning by Liu et al. on a learning platform showed that nursing students' clinical reasoning, critical thinking, and self-efficacy were enhanced, while self-directed learning was improved when compared to traditional learning methods. These findings provide evidence of the use of AI to support case-based learning in nursing education.

This study found that participants who engaged in case-based learning with AI showed statistically meaningful improvement in decision-making. Post-intervention, participants scored higher in decision-making. This AI-driven case-based learning method assisted participants in evaluating patients, considering alternative interventions, and selecting the most appropriate nursing actions. AI-based case-based learning has been shown to positively influence nursing students’ clinical judgement and the process of clinical decision-making. However, Shin et al. found that while AI-based case-based learning improved participants’ learning, it also created doubts on the technology’s reliability, and educators argued the need to interpret AI-based case-based learning output. The participants of the current study showed that AI should not replace clinical judgement, but should be an aid that complements, supports, and enhances reasoning processes. ¹⁷

AI-based case-based learning, within the context of this study, revealed a positive result and demonstrated the growing importance of technology-supported teaching methods and frameworks to support the education of nursing students in complex clinical environments. AI systems can offer constructive feedback, create clinical scenarios, and support repetition of learning tasks. This may improve the learners’ confidence and cognitive involvement. Recent studies have shown that when combined with structured case-based learning and reflected practice, generative AI may enhance components of clinical reasoning (critical thinking and self-efficacy) and teaching approaches. ¹⁸

The correlation of the academic year to progress in clinical reasoning and decision-making skills indicates that different education backgrounds may empower students to engage with AI-based learning. Some students may be able to apply their existing knowledge to more complex clinical cases and adapt accordingly to AI-centered feedback. Variations in the three colleges indicate that a particular institution’s learning environment, teaching methods, and the state of its technology may impact the outcome of an intervention. AI-based learning interventions implement education practices that go beyond technology. Curriculum, faculty engagement, and students’ digital skills are factors that influence the effectiveness of these interventions. The implications of these findings are that in order to implement AI within nursing education, strategies must be devised in order to do so with intentionality. ¹⁹

This research study provides an important framework for nursing educators and curriculum developers. Within the displaced nature of traditional nursing education and the lack of clinical context, learning opportunities, and the ability for students to practice clinical reasoning and decision-making, AI-based technologies support clinically safe and interactive learning environments for students to engage with clinical problems and receive immediate feedback on their clinical decisions prior to practice. Previous research focuses on AI-based clinical case teaching and has shown a positive influence on clinical reasoning of nursing students and the application of learned knowledge within the context of clinical practice, coupled with sustained learning satisfaction. This research has added to the existing literature and supports the integrative use of AI-assisted learning within nursing curricula. ²⁰

Although the results are insightful, they have their constraints. Generalization is limited by the quasi-experimental design, the short sample size, and the chosen nursing colleges in Lahore. More extensive studies are needed to evaluate the sustained impact of AI on nursing competencies with large samples, diverse geographical areas, randomized control, and longer intervals of time. Research also needs to evaluate students’ AI literacy, their ethical concerns, the AI technology acceptance, and the long-term implications of AI on independent, clinical judgment. It is perceived that the use of AI in education is promising, but successful integration will depend on the balance between the use of AI to support clinical judgment and the AI not being a substitute for clinical judgment. ²¹

The outcome of this study indicated that AI-supported case-based learning was effective for developing clinical reasoning and making clinical decisions for students at the undergraduate level of nursing. The participants of this study were able to evaluate the information of a clinical case, analyze the information of a case, develop a nursing diagnosis, and develop a plan after participating in some AI-enhanced learning activities. The use of AI in case-based learning provided an environment to nursing students for developing skills of advanced clinical decision making through the use of AI in case-based learning with realistic clinical situations, instant feedback, and opportunities for reflection. It exhibits the potential to fill the gaps of traditional nursing education of limited exposure to actual clinical experiences with insufficient opportunities to solve clinical problems repeatedly.

Recommendations

Based on findings from the present study, some recommendations are presented on improving nursing education through the strategic integration of AI-based case-based learning. Nursing universities are advised to incorporate AI-based case-based learning in their nursing undergraduate programs to enhance students’ clinical judgment and critical thinking and facilitate their decision-making. In this regard, faculty are advised to undergo appropriate training on the ethical incorporation of AI tools in instructional design and offer appropriate guidance to students. Nursing educators need to view AI-based supplementary learning tools in the context of traditional teaching modalities and as a strategic means to provide case-based learning and case-based reasoning substituting mentorship and human interaction. Further research is needed with greater sample sizes and more geographic diversity to include more nursing universities. Longitudinal research is needed to examine the impact and the length of time required for AI-based supplementary learning to improve students’ clinical practice and build their confidence to practice professionally. Future research may also focus on the students’ perceptions and their level of acceptance and concerns regarding AI ethics and integration of technology in nursing education. The responsible and ethical use of AI in nursing education is the core responsibility of educational policymakers and nursing regulatory bodies to ensure that AI applications in health care education support safe, evidence-based, and patient-centered practice.