Comparison of Procedural Content Item Generator versus Interactive Tool for Clinical Skills Acquisition in Physiotherapy Students

Abstract

Case-based learning (CBL) is an educational method widely used in health professional training, emphasizing theoretical knowledge’s practical application. However, traditional CBL has many challenges related to the complexity and accessibility of case scenarios and its demands on educators to effectively facilitate and evaluate student discussions. Despite its popularity and proven benefits, the comparative effectiveness and efficiency of CBL against newer educational technologies still need to be explored. In this quasi-experimental study, third-year physiotherapy students at the University of Lleida used a computer program for procedural content generation (PCG) and interactive clinical cases (ICs), and assessed them using the Spanish physiotherapy self-efficacy scale and the System Usability Scale, and a satisfaction scale. The study showed a significant improvement in self-efficacy among participants after using PCG and ICs. The usability of the PCG was moderate, while the ICs scored higher and had greater user satisfaction. Satisfaction metrics revealed a strong preference among students for incorporating clinical cases into other subjects, with higher satisfaction levels observed in the IC group compared to the PCG group. The study demonstrated that PCG and ICs significantly enhance clinical skills acquisition in physiotherapy education by improving student performance and engagement.

1. Introduction

Case-based learning (CBL) is an instructional approach in which students are presented with real-life scenarios or cases they must examine and discuss, typically in groups. This student-centered method emphasizes the practical application of theoretical knowledge, particularly in health professional education. The teacher’s role as a facilitator guides discussions to ensure students address relevant aspects of the case, enhancing their learning experience. CBL has been widely adopted in various educational fields due to its effectiveness in contextualizing learning and integrating theoretical knowledge with practical application. Students report that CBL enhances their learning by allowing them to apply theory to practice, improving their critical thinking and problem-solving skills. Despite ongoing debates about its effectiveness compared to other teaching methods, CBL remains a popular and engaging educational tool [1,2,3].

Teachers must ensure that cases are complex enough to challenge students but still accessible for their level of knowledge and experience. Secondly, facilitating CBL requires a different pedagogical approach, where teachers act as guides rather than traditional lecturers, which requires them to develop new skills and adapt to a more student-centered course. This transition involves monitoring group discussions, providing feedback, and addressing misconceptions without dominating the learning process. Additionally, evaluating students’ performance in CBL can be more complex and subjective than traditional assessments, requiring detailed rubrics and potentially laborious analysis of students’ reasoning and decision-making processes. These factors contribute to the perception that CBL is a time-intensive teaching method for educators [4,5] despite being valued by both students and educators. Single-handedly generating a variety of lifelike case scenarios can demand considerable time and resources.

Digital technologies play a crucial role in the education of allied healthcare professionals, enhancing their engagement and understanding of concepts. They also contribute to developing critical thinking and clinical reasoning across various disciplines, such as dental technology, physiotherapy, pharmacy, nutrition, nursing, and laboratory medicine [6]. Using digital tools like video methods and e-learning complements traditional curricula in healthcare education by improving psychomotor skills and communication while enriching the overall learning experience for students [6,7]. Additionally, technologies such as social media platforms, eBooks, and virtual reality support students in achieving their learning goals; however, meaningful evaluations of these educational resources are essential [8]. Active learning strategies like problem-based learning are associated with favorable student satisfaction within healthcare education settings but depend on various factors, including facilitator involvement and tutorial structure [9]. Research suggests that integrating technology into nursing curricula enhances student experiences and better prepares them to meet technological demands in real-world healthcare environments [10,11].

Procedural content generation (PCG) is a technique of generating data algorithmically instead of manually. It usually involves a mix of human-generated content and algorithms combined with computer-generated randomness and processing power [12]. PCG can create diverse, realistic case scenarios that represent physiotherapy students’ challenges in clinical practice. However, despite potential benefits in creating case scenarios for educational purposes, its use in clinical case creation has not been studied in healthcare education or higher education. Automating case scenario with PCG can alleviate some time and resource burdens associated with traditional case-based learning preparation. Nonetheless, this approach’s specific implications and effectiveness in clinical skills acquisition for physiotherapy students require further research.

Moreover, using an interactive tool for clinical skills acquisition in physiotherapy has also gained popularity. This tool allows students to engage in interactive simulations, with virtual patients, and in other computer-based activities that simulate real-world clinical scenarios. These interactive tools provide a dynamic and engaging learning experience, allowing students to practice their clinical skills in a controlled and safe environment [13].

The main goal of this study was to compare the effectiveness and feasibility of using a procedural content item generator versus an interactive tool for the acquisition of clinical skills. Additionally, this research aims to evaluate the increase in self-efficacy among physiotherapy students, assess the usability of both tools, and determine the level of satisfaction with each tool. This multifaceted approach allows for a comprehensive understanding of how these educational technologies contribute to the development of clinical skills in the context of physiotherapy education.

2. Materials and Methods

2.1. Design

In this quasi-experimental study, we assessed the increase in self-efficacy and feasibility of self-generated clinical cases in physiotherapy education. Two clinical case models were developed: one for self-generated cases based on PCG and an interactive web-based clinical case (IC). The vice-dean of the Evaluation Commission has positively evaluated the study protocol for innovation and teaching improvement projects at the University of Lleida. This study followed the transparent reporting of evaluations with nonrandomized designs (TREND) [14].

2.2. Participants

Physiotherapy students from the University of Lleida were enrolled for this study. The participants were 3rd year students enrolled in the “Methods in Physiotherapy’’ course. The University of Lleida offers a degree in physiotherapy and three double degrees: Physiotherapy and Nutrition, Physiotherapy and Sports and Physical Activity Sciences, and Physiotherapy and Nursing. The demographics will be detailed later. The classes took place between September and December of 2022.

2.3. Instrument

We designed a computer program that facilitates learning through PCG and interactive clinical cases in physiotherapy.

A program was created using Twine© (version 2.5.1.0) software to facilitate the self-generation of interactive clinical cases based on the “Mechanical Diagnosis and Treatment” method syllabus. This course focuses on the treatment and diagnosis of low back pain conditions [15]. The interactive fiction aspect allowed for the random generation of pathologies, conditions, signs, and symptoms based on established scientific evidence.

The algorithm for PCG was designed in two parts: generating the clinical case and selecting the treatment. To generate the clinical case, it was randomly assigned to a category, with each category assigned a specific condition. Each condition included major and minor signs and symptoms, demographics, etc., chosen based on the existing medical literature and expert guidelines. A line of code was also added to prevent the absence of signs or symptoms. Additionally, each condition generated responses to appropriate and inappropriate treatments.

When the condition and treatment were presented to the student, it was in a mainly text-based scenario. The students had to analyze and conduct clinical tests based on the information provided. Upon completing their assessment of the virtual patient, the program prompted them to consider if the patient is suitable for treatment or requires referral to a specialist. If they referred the patient, they received feedback on their decision. However, if they opted to treat the patient, immediate feedback was given—incorrect choices resulted in instant feedback, while correct decisions led to selecting a treatment option.

The treatment selection process was designed as a decision tree, with each choice leading to a random result based on the conditions outlined in the case algorithm. After receiving the response regarding the treatment, the patient had to decide whether to continue with the current treatment or consider an alternative. The program would provide feedback after each case if incorrect choices were made.

Additionally, for the IC, an interactive web-based clinical case featuring a video scenario between a physiotherapist and a patient was designed to require student input at specific points in the recording regarding appropriate treatment methods such as functional bandages, stretching techniques, proprioceptive exercises, and vestibular maneuvers.

2.4. Assessment

The assessment was performed using a pre-test and post-test design. The student completed the following measures to evaluate the effectiveness of this approach in improving adherence, self-efficacy, knowledge acquisition, and student satisfaction compared to traditional interactive clinical cases. The questionnaires were provided digitally, and personal data generated a unique number to identify the student anonymously. The course development can be found in Figure 1, and the assessment is described here:

• We assessed self-efficacy using a Spanish version of the physiotherapy self-efficacy scale (PSE), which is being validated, based on the English version of the PSE [16]. Self-efficacy refers to an individual’s belief in their capacity to execute behaviors necessary to produce specific performance attainments. In the context of physiotherapy, self-efficacy is crucial as it can influence a patient’s motivation, adherence to treatment plans, and overall recovery outcomes. The PSE was answered at the beginning of the course (Before PSE) and after (After PSE) using the tools. We used only the results from the students who answered the PSE before and after the course.
• To analyze the usability and user experience of the interactive tool, we used the Spanish, validated version of the System Usability Scale (SUS) [17]. It encompasses several key aspects: ease of use, effectiveness, efficiency, satisfaction, and learnability. The SUS is a Likert questionnaire that provides a specific score. The mean score will indicate the level of usability of PCG and ICs. The SUS scoring is interpreted as a percentage of usability, where a score above 80 indicates excellent usability, whereas a score under 50 indicates poor usability. Usually, the SUS results are converted into three scales: an adjective scale, a grade scale, and an acceptability scale. The adjective scale consists of the adjective ratings “Worst imaginable”, “Poor”, “OK”, “Good”, ”Excellent”, and “Best imaginable”. A grade scale is suggested using the traditional school grading scale from F to A. The acceptability scale classifies average SUS scores as “Acceptable”, “Marginal”, or “Not acceptable”. A percentile curve contextualization graph is also usually presented [18]. This graph shows the calculated SUS study scores from the uploaded dataset on a percentile curve derived from over 5000 SUS questionnaires. The data for this curve were taken from Sauro et al. [19]. Our study assessed the SUS of PCG and ICs after the students had used it.
• The survey aims to assess the impact of educational activities in a physiotherapy curriculum on several vital outcomes (

  Table 1. Satisfaction survey questions.

  Statement
  1. To what extent would you like these types of clinical cases to be used in other subjects?
  2. The content of the activity allows me to establish effective communication with patients, family, social groups, and peers and to promote health education.
  3. The content of the activity allows me to assess the patient’s functional status, considering the physical, psychological, and social aspects
  4. I believe I have a good command of the theoretical material of the Methods in Physiotherapy course
  5. Carrying out the activity allows me a better understanding of the possibilities and limitations of interventions in physiotherapy
  6. The content of the activity helps me integrate and relate the knowledge acquired from this and other subjects by applying them to a real clinical case
  7. The knowledge acquired during the activity enables me to better identify and analyze the crucial elements to solve the problems of a real case
  8. The content of the activity helps me consolidate the knowledge acquired in the Methods in Physiotherapy course
  9. The knowledge acquired during the activity is relevant for the practice of physiotherapy

  ): the relevance of acquired knowledge to practice, consolidation of course content, enhancement of problem-solving skills in real-case scenarios, integration of interdisciplinary knowledge, understanding of physiotherapy interventions, mastery of theoretical material, communication, and health education skills, and ability to assess patients’ functional status comprehensively. Additionally, it seeks student feedback on the preference for using clinical case-based approaches across other subjects.

2.5. Statistical Analysis

The statistical analysis was conducted using IBM SPSS Statistics (Version 27), involving descriptive statistics frequency counts and the Student’s t-test to compare hypotheses. The Shapiro–Wilk test was used to analyze the degree of normality. Changes in the PSE were analyzed using a paired t-test to compare pre-test and post-test scores. The statistical analysis was conducted at a 95% confidence level. A p-value of less than 0.05 was considered statistically significant in all studies.

The analysis of the SUS used the online System Usability Scale Analysis Toolkit (https://sus.mixality.de, accessed on 22 July 2024).

The satisfaction questionnaire results were analyzed using a frequency analysis of Likert responses.

3. Results

Sixty-three students participated in the PSE questionnaire at the beginning of the course. However, only 36 answered the PSE questionnaire at the end of the course. The Shapiro–Wilk test indicates that the PSE answers before and after the course show no significant deviation from a normal distribution (W = 0.9759, p = 0.6075). For the PSE at the end of the course, data showed no significant deviation from normality (W = 0.9648, p = 0.3008).

The participation in this study is explained in

Table 2. Participation demography.

	Before PSE (n = 63)	After PSE (n = 36)	SUS (PCG) (n = 35)	SUS (IC) (n = 33)	Satisfaction (n = 63)
Gender
Male	26	16	20	18	31
Female	36	19	16	15	31
Nonbinary	1	1	0	0	1
Mean age (standard deviation)	21.30 (2.39)	21.03 (4.11)	22.93 (4.64)	22.22 (4.00)	21.97 (3.26)
Degree
Science of physical activities and sport + physiotherapy	25	8	8	6	14
Nursing + physiotherapy	17	14	17	8	21
Human nutrition + physiotherapy	13	10	4	13	11
Physiotherapy	8	4	8	6	17

PSE: physiotherapy self-efficacy; SUS: System Usability Scale; PCG: procedural content generation. IC: interactive case.

.

3.1. Self-Efficacy

The PSE scores were evaluated before and after a specified intervention to assess the impact on participants’ self-reported physiotherapy-related self-efficacy (Figure 2,

Table A1. Frequency of the Likert answers to the previous PSE.

Statement	1 *	2	3	4	5
I feel adequately prepared to undertake a caseload.	0	3	14	18	1
I feel that I am able to verbally communicate effectively and appropriately for a caseload.	0	1	14	14	7
I feel that I am able to communicate in writing effectively and appropriately for a caseload.	0	3	13	15	5
I feel that I am able to perform subjective assessments for a caseload.	1	3	19	10	3
I feel that I am able to perform objective assessment for a caseload.	0	3	14	17	2
I feel that I am able to interpret assessment findings appropriate for a caseload.	0	3	17	15	1
I feel that I am able to identify and prioritize patient’s problems for a caseload.	1	4	8	18	5
I feel that I am able to select appropriate short and long term goals for a caseload.	1	7	19	7	2
I feel that I am able to appropriately perform treatments for a caseload.	4	5	15	12	0
I feel that I am able to perform discharge planning for a caseload.	7	12	12	3	2
I feel that I am able to evaluate my treatments for a caseload.	2	7	16	10	1
I feel that I am able to progress interventions appropriately for a caseload.	1	4	14	15	2
I feel that I am able to deal with the range of patient conditions which may be seen with a caseload.	2	3	18	10	3

* The Likert values are 1—strongly disagree to 5—strongly agree.

,

Table A2. Frequency of the Likert answers to the posterior PSE.

Statement	1 *	2	3	4	5
I feel adequately prepared to undertake a caseload. I feel that I am able to verbally communicate effectively and appropriately for a caseload. I feel that I am able to communicate in writing effectively and appropriately for a caseload.	0	3	9	19	5
	0	1	8	18	9
	0	4	12	15	5
I feel that I am able to perform subjective assessments for a caseload. I feel that I am able to perform objective assessment for a caseload.	1	0	14	21	0
	0	2	6	22	6
I feel that I am able to interpret assessment findings appropriate for a caseload. I feel that I am able to identify and prioritize patient’s problems for a caseload. I feel that I am able to select appropriate short and long term goals for a caseload. I feel that I am able to appropriately perform treatments for a caseload.	0	1	12	19	4
	0	0	4	20	12
	1	3	5	20	7
	0	2	9	22	3
I feel that I am able to perform discharge planning for a caseload.	2	9	15	10	0
I feel that I am able to evaluate my treatments for a caseload.	0	6	12	15	3
I feel that I am able to progress interventions appropriately for a caseload.	0	1	12	19	4
I feel that I am able to deal with the range of patient conditions which may be seen with a caseload.	0	4	11	19	2

* The Likert values are 1—strongly disagree to 5—strongly agree.

and

Table A3. Response frequency of the PSE comparing previous and posterior responses.

Pair	Statement	Mean	Std. Deviation
1 PREVIOUS	My training has adequately prepared me	3.47	0.696
1 POSTERIOR	My training has adequately prepared me	3.72	0.815
2 PREVIOUS	My training has adequately prepared me for verbally communicating effectively and appropriately.	3.75	0.806
2 POSTERIOR	My training has adequately prepared me for verbally communicating effectively and appropriately.	3.97	0.774
3 PREVIOUS	My training has adequately prepared me for communicating in writing effectively and appropriately.	3.61	0.838
3 POSTERIOR	My training has adequately prepared me for communicating in writing effectively and appropriately.	3.58	0.874
4 PREVIOUS	My training has adequately prepared me for performing subjective assessments.	3.31	0.856
4 POSTERIOR	My training has adequately prepared me for performing subjective assessments.	3.53	0.654
5 PREVIOUS	My training has adequately prepared me for performing objective assessments.	3.50	0.737
5 POSTERIOR	My training has adequately prepared me for performing objective assessments.	3.89	0.747
6 PREVIOUS	My training has adequately prepared me for interpreting assessment findings.	3.39	0.688
6 POSTERIOR	My training has adequately prepared me for interpreting assessment findings.	3.72	0.701
7 PREVIOUS	My training has adequately prepared me for identifying and prioritizing patients’ problems.	3.61	0.964
7 POSTERIOR	My training has adequately prepared me for identifying and prioritizing patients’ problems.	4.22	0.637
8 PREVIOUS	My training has adequately prepared me for selecting appropriate short- and long-term goals.	3.06	0.860
8 POSTERIOR	My training has adequately prepared me for selecting appropriate short- and long-term goals.	3.81	0.951
9 PREVIOUS	My training has adequately prepared me for appropriately performing treatments.	2.97	0.971
9 POSTERIOR	My training has adequately prepared me for appropriately performing treatments.	3.72	0.701
10 PREVIOUS	My training has adequately prepared me for performing discharge planning.	2.47	1.082
10 POSTERIOR	My training has adequately prepared me for performing discharge planning.	2.92	0.874
11 PREVIOUS	My training has adequately prepared me for evaluating my treatments.	3.03	0.910
11 POSTERIOR	My training has adequately prepared me for evaluating my treatments.	3.42	0.874
12 PREVIOUS	My training has adequately prepared me for progressing interventions appropriately.	3.36	0.867

). The pre-intervention group (Group 2) comprised 63 participants, while the post-intervention group (Group 1) included 36 participants. We used the data of the students who answered both questionnaires, identifying them by their unique identification numbers. The mean score for the post-intervention group was significantly higher (M = 3.56, SD = 0.37) than the pre-intervention group (M = 3.13, SD = 0.37), indicating improved self-efficacy following the intervention.

A two-sample t-test was performed to compare the mean scores of the two groups, yielding a t-statistic of −3.2611 with a p-value of 0.001716. This result suggests a statistically significant difference in the physiotherapy self-efficacy scores between the pre- and post-intervention groups, with the post-intervention group showing a higher mean self-efficacy score. The effect size, as measured by the difference in means between the two groups, further supports the conclusion that the intervention positively affected the participants’ self-efficacy related to physiotherapy. The observed effect size (Cohen’s d) was 0.77, indicating a large magnitude of difference. This statistically robust outcome validates the intervention’s efficacy in enhancing participants’ self-efficacy.

3.2. Usability

Thirty-five students answered the SUS scale for the PCG, and thirty-three students answered the SUS scale for the ICs.

The SUS score for the PCG tool was calculated to be 65.93 (Figure 3, and

Table 3. Results of the SUS scores for the PCG and ICs.

Metric/Question	PCG	IC
SUS Study Score	65.93	77.5
Median	70	77.5
Standard Deviation	19.3	13.13
Adjective	OK	Good
Grade	C	B
Acceptability	Marginal	Acceptable
Quartile	2nd	3rd
Conclusiveness	100%	100
Answers to the questions	PCG	IC
Q1: I think that I would like to use this system frequently.	7.07	8.64
Q2: I found the system unnecessarily complex.	6.71	7.88
Q3: I thought the system was easy to use.	6.79	8.64
Q4: I think that I would need the support of a technical person to be able to use this system.	7.29	8.18
Q5: I found the various functions in this system were well integrated.	7.07	7.58
Q6: I thought there was too much inconsistency in this system.	6.86	7.12
Q7: I imagine most people would learn to use this system very quickly.	6.79	8.71
Q8: I found the system very cumbersome to use.	7.07	7.58
Q9: I felt very confident using the system.	6.14	7.65
Q10: I needed to learn a lot of things before I could get going with this system.	4.14	5.53

and

Table A4. Response frequency of the SUS for PCG.

Statement	1 *	2	3	4	5
I think that I would like to use this system frequently.	2	1	7	25	10
I found the system unnecessarily complex.	10	15	10	8	2
I thought the system was easy to use.	1	5	5	19	6
I think that I would need the support of a technical person to be able to use this system.	9	21	11	2	2
I found the various functions in this system were well integrated.	0	3	5	20	6
I thought there was too much inconsistency in this system.	11	15	9	8	2
I would imagine that most people would learn to use this system very quickly.	3	2	7	13	10
I found the system very cumbersome to use.	5	14	13	3	0
I felt very confident using the system.	3	5	8	14	5
I needed to learn a lot of things before I could get going with this system.	3	2	10	16	4

* The Likert values are 1—strongly disagree to 5—strongly agree.

). This score is positioned below the median SUS score of 70, indicating a moderate level of usability. The distribution of scores among the participants revealed a standard deviation of 19.3, highlighting variability in user experiences with the PCG tool. The overall usability of the PCG tool was deemed “OK” with a corresponding grade of “C”. Its acceptability was marginal, placing it in the second quartile of usability ratings.

The SUS study score for the IC tool was determined to be 77.5 (Figure 4, and Table 3 and

Table A5. Response frequency of the SUS for ICs.

Statement	1 *	2	3	4	5
I think that I would like to use this system frequently.	0	1	3	9	20
I found the system unnecessarily complex.	15	15	17	1	0
I thought the system was easy to use.	1	2	2	9	22
I think that I would need the support of a technical person to be able to use this system.	18	14	15	0	1
I found the various functions in this system were well integrated.	2	2	3	16	14
I thought there was too much inconsistency in this system.	15	14	16	2	1
I would imagine that most people would learn to use this system very quickly.	0	0	3	14	19
I found the system very cumbersome to use.	8	21	7	2	0
I felt very confident using the system.	0	1	6	16	10
I needed to learn a lot of things before I could get going with this system.	5	6	9	16	2

* The Likert values are 1—strongly disagree to 5—strongly agree.

). This score precisely matches the median, indicating a good level of usability that aligns with the median usability score across various systems. The standard deviation observed within the data was 13.13, suggesting a relatively consistent user experience among participants. The usability of the IC tool was categorized as “Good” and was awarded a grade of “B”. With an acceptability status of “Acceptable”, the tool positioned itself in the third quartile, signifying higher user satisfaction than many other systems.

3.3. Satisfaction

In the PCG, most students expressed positive attitudes toward integrating clinical cases into the curriculum (Figure 5 and

Table 4. Questions of the satisfaction questionnaire.

Number	Questions
Q1	To what extent would you like these types of clinical cases to be used in other subjects?
Q2	The content of the activity allows me to assess the patient’s functional status, considering the physical, psychological, and social aspects
Q3	The content of the activity allows me to establish effective communication with patients, family, social groups, and peers and to promote health education.
Q4	I believe I have a good command of the theoretical material of the Methods in Physiotherapy course
Q5	Carrying out the activity allows me a better understanding of the possibilities and limitations of interventions in physiotherapy
Q6	The content of the activity helps me integrate and relate the knowledge acquired from this and other subjects by applying them to a real clinical case
Q7	The knowledge acquired during the activity enables me to better identify and analyze the crucial elements to solve the problems of a real case
Q8	The content of the activity helps me consolidate the knowledge acquired in the Methods in Physiotherapy course
Q9	The knowledge acquired during the activity is relevant for the practice of physiotherapy

and

Table A6. Frequency of the Likert answers to the satisfaction questionnaire for PCG.

Statement	Totally Disagree	Disagree	Slightly Disagree	Slightly Agree	Agree	Totally Agree
To what extent would you like these types of clinical cases to be used in other subjects?	2	3	12	7	15	24
The content of the activity allows me to assess the patient’s functional status, considering the physical, psychological, and social aspects	5	6	8	10	18	16
The content of the activity allows me to establish effective communication with patients, family, social groups, and peers and to promote health education.	4	7	11	9	23	9
I believe I have a good command of the theoretical material of the General Physiotherapy Procedures II course	5	5	13	17	17	6
Carrying out the activity allows me a better understanding of the possibilities and limitations of interventions in physiotherapy	6	6	5	10	15	21
The content of the activity helps me integrate and relate the knowledge acquired from this and other subjects by applying them to a real clinical case	6	5	9	8	19	16
The knowledge acquired during the activity enables me to better identify and analyze the crucial elements to solve the problems of a real case	6	4	7	9	20	17
The content of the activity helps me consolidate the knowledge acquired in the General Physiotherapy Procedures II course	6	7	3	15	19	13
The knowledge acquired during the activity is relevant for the practice of physiotherapy	7	4	9	8	18	17

). The statement “To what extent would you like these types of clinical cases to be used in other subjects?” received notable endorsements, with 24 students totally agreeing and 15 agreeing. Similarly, for the statement regarding the content’s ability to assess the patient’s functional status across physical, psychological, and social dimensions, 16 students totally agreed, complemented by 18 who agreed.

Statements concerning the application of theoretical material and the relevance of the knowledge to physiotherapy practice also garnered strong agreement. However, a lower level of complete confidence in mastering the course material was evident, as the most robust responses shifted toward slight agreement, by 17 students, rather than total agreement, by 6.

Responses from the ICs demonstrated higher satisfaction across several vital aspects (Figure 6 and

Table A7. Frequency of the Likert answers to the satisfaction questionnaire for the ICs.

Statement	Totally Disagree	Disagree	Slightly Disagree	Slightly Agree	Agree	Totally Agree
To what extent would you like these types of clinical cases to be used in other subjects?	1	0	8	7	12	35
The content of the activity allows me to assess the patient’s functional status, considering the physical, psychological, and social aspects	1	2	9	6	23	22
The content of the activity allows me to establish effective communication with patients, family, social groups, and peers and to promote health education.	1	1	9	8	24	20
I believe I have a good command of the theoretical material of the General Physiotherapy Procedures II course	1	2	10	17	18	15
Carrying out the activity allows me a better understanding of the possibilities and limitations of interventions in physiotherapy	1	0	5	7	20	30
The content of the activity helps me integrate and relate the knowledge acquired from this and other subjects by applying them to a real clinical case	1	1	8	4	18	31
The knowledge acquired during the activity enables me to better identify and analyze the crucial elements to solve the problems of an actual case	2	1	5	6	21	28
The content of the activity helped me consolidate the knowledge acquired in the General Physiotherapy Procedures II course	1	2	5	6	18	31
The knowledge acquired during the activity is relevant to the practice of physiotherapy	1	0	6	6	21	29

). The use of clinical cases in other subjects was highly favored, with 35 students totally agreeing and 12 agreeing. Assessment of the patient’s functional status through the content was positively received, with 22 students totally agreeing and 23 agreeing.

The relevance of the content to practical physiotherapy and the effectiveness in communication with patients and related groups showed a substantial positive response, with 20 and 24 students, respectively, totally agreeing. The integration of theoretical and practical knowledge appeared robust, as reflected by the high agreeance on the statement related to identifying and analyzing crucial problem-solving elements in actual cases, with 28 totally agreeing and 21 agreeing.

4. Discussion

This study shows that PCG is a feasible method for physiotherapy education. Even though ICs have better satisfaction and usability, PCG’s good results make it a possible tool to be implemented in higher education. Nowadays, few studies have used PCG in education at the school level, in math [20] and reading skills [21]. Some automatic item generation (AIG) has been used to create multiple choice questions (MCQs) for exams and exercises in healthcare education. Studies have shown that AIG can produce high-quality MCQs comparable to traditionally developed items, as evaluated by medical experts [22]. AIG has also demonstrated success in generating case-based MCQs for clinical reasoning, with items reviewed and validated by specialists [23]. Additionally, AIG-generated items have been found to possess strong psychometric properties, effectively discriminating between low and high performers [24]. AIG has improved the quality of distractors in MCQs, ensuring they differentiate well between candidates [25]. Furthermore, AIG has shown promise in creating consistent and reliable items for mastery learning in continuous assessment environments [26].

This study found an increase in self-efficacy after using the tools. Increased self-efficacy in physiotherapy students has notable implications for their education and future clinical practice. Self-efficacy, the belief in one’s capabilities to achieve a goal or an outcome, is critical in health education, including physiotherapy. Enhanced self-efficacy is directly linked to better clinical performance, with students exhibiting improved patient interactions and outcomes [27]. Training interventions like simulation-based education significantly boost clinical skills and deep learning, improving overall performance [28]. Additionally, higher self-efficacy enhances students’ confidence in patient education, improving their communication and teaching abilities [29]. It also prepares students for lifelong learning and trains their adaptability to various clinical situations through traditional or self-directed learning methods [30]. Furthermore, students with higher self-efficacy report better stress management and preparedness for clinical exams and placements, correlating with improved performance in practical assessments [31].

However, self-reported self-efficacy outcomes are not perfect, due to subjectivity, so these results should be interpreted cautiously. Clinical reasoning should be assessed using validated tools that enhance educational outcomes and better prepare students for clinical practice. Practical tools include the Think-Aloud Standardized Patient Examination (TASPE) [32] and the Objective Structured Clinical Examination (OSCE) [33]. Both assess students’ diagnostic and treatment decisions in simulated scenarios. The Clinical Reasoning Reflection Questionnaire (CRRQ) and the Clinical Reasoning Grading Rubric [34] help track the development of reasoning skills over time.

Additionally, incorporating standardized patient and physician interactions in assessments and simulated patient encounters provides robust feedback and reflection opportunities. Lastly, faculty assessments using tools like the Clinical Reasoning Assessment Tool (CRAT) [35] offer structured feedback, which is crucial for students’ continuous improvement in clinical reasoning. Collectively, these approaches provide a comprehensive evaluation beyond self-reported measures.

The interactive case-based tool was more usable than the procedural content generation tool. Many factors can explain these differences in usability. First, the ICs likely provided a more intuitive and user-friendly interface. The students have to make a maximum of four choices, while PCG has a broader list of options that simulate realistic scenarios but may be more complex to navigate. Moreover, the ICs used videos to simulate a patient, as opposed to the PCG, which was text only. This makes ICs more engaging, and it is a factor that needs to be improved on the PCG.

Usability is a critical factor in the success of teaching innovation tools, ensuring they are effective, efficient, and satisfying for users, enhancing learning outcomes and user engagement. Usable tools improve the efficiency and effectiveness of learning by simplifying navigation and comprehension of content, leading to better learning outcomes and higher user satisfaction [36]. High usability also increases user satisfaction and engagement, encouraging consistent use and deeper interaction with the material, resulting in positive learning experiences [37]. Good usability minimizes learning barriers, such as confusion or frustration, enabling students to focus on content rather than interface issues. This is crucial for diverse student populations with varying technical proficiency [38]. Additionally, usable tools support different learning styles and needs, creating a more inclusive learning environment that caters to a wide range of students, including those with special needs [39]. Tools with good usability are more likely to be adopted and integrated into regular teaching practices, leading to more consistent and effective teaching and learning experiences across different educational settings [40]. Moreover, usability testing provides valuable feedback for continuous improvement, ensuring the tool remains relevant and practical, adapting to new educational needs and technological advancements [41].

The satisfaction of healthcare students with teaching innovations significantly influences various aspects of their education and future clinical practice. Understanding these consequences helps educators and institutions optimize teaching strategies to enhance educational outcomes. Students’ satisfaction with innovative teaching methods, such as problem-based learning (PBL) and high-fidelity simulation, is positively correlated with better academic performance and a deeper understanding of clinical skills [42,43]. Satisfied students show higher levels of motivation and engagement in their studies, particularly in active learning environments where they feel more involved and responsible for their learning [43]. Satisfaction with teaching methods that include direct patient care experiences, such as clinical clerkships, enhances students’ confidence and competence in patient interactions, often resulting in better patient satisfaction [44]. Furthermore, students’ satisfaction encourages faculty to adopt and refine these methods, leading to continuous improvement in teaching practices [45]. Teaching innovations that satisfy students often include components that mimic real-world clinical scenarios, such as high-fidelity simulations and interprofessional education, better preparing students for clinical practice [46]. High student satisfaction with teaching innovations ensures their sustainability and further development, securing ongoing support and resources for these practices and fostering a culture of continuous improvement in healthcare education [47].

4.1. Theoretical Contribution

The PCG tool developed for this study extends PCG’s application from its traditional realms in gaming and entertainment into educational settings, particularly in the health professions. By demonstrating how PCG can create diverse, realistic clinical cases, this study provides novel insight into how digital tools can enhance case-based learning. This addresses critical challenges in traditional CBL, such as the scalability of case diversity and the reduced workload on educators in developing and facilitating case discussions.

Secondly, the study contributes to the theoretical framework of constructivist learning theories, which advocate that learning is more effective when learners actively construct their understanding through real-world experiences. By integrating PCG and ICs into the curriculum, this study offers a practical examination of these theories, showing that digital tools can facilitate active learning and improve student engagement and self-efficacy in learning environments.

Furthermore, this study provides empirical evidence supporting the theory of self-efficacy in educational settings [48]. The findings suggest that PCG and ICs can significantly enhance physiotherapy students’ self-efficacy, aligning with Bandura’s concept that effective learning environments involve mastery experiences, social modeling, and social persuasion.

Lastly, using the SUS and satisfaction metrics to evaluate the usability and user satisfaction of educational technologies contributes to the broader discussion on how the design and functionality of educational tools influence learning outcomes. This aspect of this study highlights the importance of developing user-friendly educational technologies that meet the needs of learners and educators alike.

4.2. Practical Implications

This study on the use of PCG and interactive clinical cases (ICs) in physiotherapy education provides a compelling argument for integrating these technologies into educational settings. By automating the creation of complex clinical scenarios, PCG allows educators to focus more on guiding student learning than on content creation, enhancing the efficiency and diversity of case-based learning. This shift saves time and enriches the learning environment with a broader spectrum of realistic scenarios, increasing student engagement.

The improvement in student self-efficacy observed using these tools is particularly notable, as higher self-efficacy correlates with better academic performance and increased motivation. Institutions should consider this benefit as they integrate more interactive and technology-driven methods into their curricula to prepare students more effectively for clinical practice.

Moreover, the positive feedback on the usability and student satisfaction with these tools highlights their potential for broader adoption. This enhances the learning experience, suggesting that educational technologies, when well implemented, can significantly improve engagement and academic outcomes. Additionally, our findings could influence educational policy by encouraging the adoption of such technologies and advocating for ongoing professional development for educators to maximize the benefits of these innovations.

Overall, this study points to a transformative approach in physiotherapy education. This approach promotes a more dynamic and interactive learning environment that aligns with technological advancements and educational theories. This approach enhances learning outcomes and sets a precedent for future educational strategies and technological integration in health-related fields.

This research explored the integration of PCG and ICs in physiotherapy education, highlighting their potential to create a richer, more engaging, and effective learning experience. The findings suggest that PCG-driven can significantly enhance students’ self-efficacy and satisfaction with learning. This is crucial as it fosters a more active and self-directed approach to education, better preparing students for the complexities of clinical practice.

Further exploration of PCG applications in physiotherapy education, particularly its long-term impact on clinical reasoning and decision-making skills, is encouraged. Additionally, investigating the scalability and adaptability of this approach across diverse learning environments and educational institutions will be valuable for broader adoption and integration.

4.3. Limitations

This study is accompanied by some limitations that impact its broader applicability. One significant limitation is the absence of a control group, which constrains the ability to distinctly evaluate the effects of the tools due to the impossibility of dividing the classroom into different groups under the given constraints. Employing randomized controlled trials would provide a more robust framework for assessing causality. However, ethical concerns exist about using a teaching innovation only in one group. Future research could address this by implementing crossover interventions within split groups to assess learning gains more effectively.

Another notable limitation is the reliance on self-reported data, which can introduce bias as such data may not accurately reflect student outcomes or perceptions. The use of validated scales for the assessment of clinical reasoning is suggested. Some of these assessments have been previously explained in the discussion. An identification system for the student should accompany this to correlate the tools’ use and results.

Additionally, the study was conducted within a single educational institution, limiting the generalizability of the findings to other educational settings. To build upon this foundation, future studies should consider a broader participant base across multiple institutions to enhance the generalizability of the findings.

The technological aspect of the study predates the emergence of generative artificial intelligence, which presents an opportunity for future projects to integrate generative AI to enhance the procedural content generation process, potentially creating more dynamic and personalized student learning experiences.

Another limitation is the single assessment after the use of the tools. We cannot assure that the improvement in self-efficacy is permanent.

5. Conclusions

In conclusion, using procedural content item generators to acquire clinical skills for physiotherapy students showed promising results. Students demonstrated improved performance and engagement using the procedural content item generator and an interactive tool. Procedural content generation enhances the creation of personalized and adaptive learning experiences and provides opportunities for unsupervised training. Based on our research, PCG in educational settings, particularly in physiotherapy students’ acquisition of clinical skills, has shown positive outcomes.