Self-Regulation of Learning and Its Implications for Academic Performance and Well-Being of University Students in Health Sciences: A Systematic Review

Abstract

Self-regulated learning (SRL) is a fundamental competence for academic transition and success in higher education, especially in health sciences, where autonomy and learning management are essential. This systematic review analyzed the relationship between SRL, academic performance, and student well-being among undergraduate health sciences students. Following the PRISMA protocol, 39 articles published between 2015 and 2025 on Web of Science, Scopus, and PubMed databases were selected. The consolidated sample consisted of 24,835 participants. The methodological quality of the selected studies was assessed using the Newcastle–Ottawa scale (NOS). A predominantly positive association was found between high levels of SRL and academic performance (GPA) (with correlation coefficients ranging from r = 0.11 to r = 0.55 in the primary studies). Furthermore, evidence from standardized self-report questionnaires in the reviewed literature indicates that several studies report female students showed higher levels of organization and planning, but these findings were not consistently observed across all studies. SRL acts as a key protective factor against stress, anxiety, and academic burnout. However, a “stagnation paradox” was identified: SRL skills do not always evolve linearly, often showing regression or stagnation in advanced clinical years due to the high cognitive load and insufficient support structure in those environments. Regarding sociodemographic variables, female students reported higher levels of planning and responsibility. SRL does not develop spontaneously with academic progress. Therefore, higher-education institutions must implement systematic and intentional pedagogical strategies from the early years of training to foster student well-being and the development of resilient professionals.

1. Introduction

Entering higher education involves a challenging academic transition [1] that demands adaptation from students due to the autonomy required by the process and the necessary higher cognitive skills [2]. Due to this, students are expected to be able to manage their learning, which will directly influence their ability to adapt to university life [3].

A suitable educational environment can foster the development of these skills that not all students have before entering university [4], so the concept of self-regulated learning or SRL in higher education has become relevant as a measure to promote essential skills, facilitating integration into university life, and improving academic performance [5].

The SRL is fundamental to the university process and to lifelong learning, essential for a more sustainable and equitable society [6,7].

SRL is defined as the conscious ability to direct one’s own learning behavior [8], constituting a dynamic and self-directed process where control resides in the student for the development of academic skills [9,10].

SRL encompasses cognitive, motivational, and affective factors, involving self-generated thoughts, feelings, and behaviors oriented towards achieving a purpose [11]. Its objective is for students to actively participate in their learning process, setting goals and strategies based on external and internal resources [12].

The development of SRL generates better academic performance by being a specific process that must be adapted to each task, and not an intrinsic personal trait [11,12].

Within the SRL process, certain transversal characteristics can be described, which would be:

• SRL is dynamic, dependent on the academic environment, and can be controlled by the student. Therefore, learning skills can be acquired over time [13].
• The student has an active role in the SRL, managing their academic training instead of being a passive recipient [3].
• External factors (academic and social context) and internal (executive functions and motivation) influence the development of SRL and can be regulated by students. The university environment and interaction with peers/professors favor SRL [3,5,14].
• The development of the SRL is imperative for continuous learning and professional performance, since its skills transcend the academic [3,5].
• The motivational and strategic components of SRL can be measured quantitatively using standardized instruments [3].

Two SRL models stand out. The first is proposed by Zimmerman [12], which proposes three cyclical phases: (i) planning: analysis of the task, goals, resources, and self-motivation; (ii) performance: self-control, application of strategies and self-observation, recording actions and time spent; and (iii) self-reflection: self-evaluation of the process and emotional analysis.

On the other hand, Pintrich postulates a similar model with four cyclical phases [11] (i) planning: review of objectives, goal orientation, time planning, perceptions of difficulty and academic context; (ii) monitoring: metacognition of the process, motivation, effort, use of time and need for help; (iii) control: review of the effectiveness of cognitive and metacognitive strategies and volitional control used to adjust strategies; and (iv) reaction–reflection: self-evaluation of the effectiveness of the strategies used.

Although there are similarities, both authors agree that SRL is a cyclical and dynamic process, whose phases are not necessarily sequential and can occur in parallel [11,12].

SRL plays a fundamental role in higher education because it predicts better academic performance and grades [15,16]. It facilitates the transition and adaptation to university life, increasing satisfaction and acting as a protective factor against dropout [7]. Moreover, it fosters higher-order thinking skills and empowers students to autonomously manage their educational process, promoting motivation, self-efficacy, and proactive behavior [9]. This autonomy positively influences well-being, where time management is a protective factor against academic burnout [1]. Finally, SRL contributes to holistic development by fostering transversal skills such as critical thinking and strategic learning [2]. Therefore, SRL is fundamental to academic performance, adaptation to the university environment, and comprehensive professional development.

Also, following global trends in Curriculum harmonization, such as the Credit System in Latin America, the aim is to measure and harmonize student workload, declaring hours of direct instruction and independent work for the achievement of Expected Learning Outcomes and the Graduate Profile. This regulated context highlights the importance of students developing adequate SRL.

SRL is crucial for student well-being, acting as a protective factor against depression, stress, anxiety, and burnout [17]. SRL is negatively correlated with depression, promoting coping with stress [18] and contributing to its management [19]. Additionally, it reduces anxiety by facilitating clinical decision-making, where poor performance is associated with increased stress and SRL difficulties [20]. Similarly, SRL prevents burnout [1], while low academic well-being negatively impacts SRL, professional identity, and burnout in students [21]. In summary, SRL is essential for preventing these problems and for successful university adjustment.

Empirical evidence points to the critical importance of SRL in university students and the need for its systematic development from the beginning of their studies. Implementing early SRL pedagogical strategies and support programs would improve academic performance and curricular progress, while also acting as a preventative measure against demotivation, stress, and dropout. Therefore, promoting SRL is essential to foster persistence, successful completion of studies, and development of competent professionals.

Despite the evidence, SRL models in health sciences require special attention due to the theoretical-practical nature of their training. Understanding how SRL is integrated into these programs, considering curricular variability, is crucial to mitigate academic risks and ensure student well-being and adequate professional training. Therefore, this systematic review analyzed the relationship between SRL, academic performance, and student well-being among undergraduate health sciences students.

Understanding this phenomenon is highly relevant in the current academic panorama to design effective educational interventions. Therefore, to address this gap, this systematic review analyzed the relationship between SRL, academic performance, and student well-being. To achieve this goal, we implemented a rigorous systematic search strategy following PRISMA guidelines across Web of Science, Scopus, and PubMed, aiming to synthesize the most current evidence and provide a clear overview for future educational policies.

2. Methods

2.1. Design and Sources of Information

This systematic review, conducted according to the 2020 PRISMA guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) [22], examined the relationship between SRL, academic performance, and student well-being among undergraduate health sciences students. Systematic searches were performed in three broad-spectrum academic databases—Web of Science, Scopus, and PubMed—selected for their comprehensive coverage of the social science, education, psychology, and health. The final search was completed on 30 July 2025. In addition, the reference lists of included studies were manually reviewed to identify potentially relevant publications that were not captured by the automated search. This systematic review was not registered in PROSPERO. A formal protocol was not prepared; however, the review strictly adhered to PRISMA 2020 guidelines.

A ten-year time frame (2015–2025) was defined for the analysis to capture current and robust empirical evidence while avoiding biases arising from the COVID-19 lockdown period. This context significantly altered teaching and learning processes and patterns of SRL in students [23], which could affect the comparative validity of the results. This methodological decision aligns with the recommendations of the PRISMA framework, which allows for defining the search period whenever there is a clear theoretical and contextual justification [22].

2.2. Search Strategy

The literature search was carried out in accordance with the PRISMA protocol (see Figure 1) for systematic reviews and meta-analyses [22]. Searches were conducted across each database using the title, abstract, and keyword fields of every article. The following keywords were combined with Boolean operators (OR/AND): [(“academic self-regulation” OR “academic self regulation” OR “self-regulated learning”) AND (“Health students” OR “Healthcare students” OR “Health professions students” OR “medical students” OR “nursing students” OR “allied health students”)].

Figure 1. Search strategy for selected studies.

2.3. Eligibility Criteria

Before reviewing the full content of each article, the titles and abstracts were checked for relevance. The initial screening of titles and abstracts was conducted manually and independently by the authors, without the use of artificial intelligence assistants. Reference management software Mendeley (version 2.142.0) was used to detect and eliminate duplicates. The process of reading and filtering titles and abstracts took approximately six months.

Studies were selected based on the following criteria: (1) cross-sectional design; (2) study population comprising undergraduate health sciences students; (3) studies conducted with students attending face-to-face classes; and (4) studies written in English and Spanish.

The following exclusion criteria were established: (1) exclusively qualitative studies; (2) postgraduate or medical specialty students; (3) studies where the SRL variable was not directly related to the central focus of the study, and (4) studies with students attending online or hybrid classes. The extraction process was carried out independently by two reviewers, and a third reviewer acted as an external verifier in case of disagreement. Discrepancies were resolved through discussion and consensus.

Qualitative studies were excluded because this review aimed to synthesize quantitative evidence on the statistical link between self-regulated learning (SRL) and academic performance and well-being outcomes. Although valuable for understanding SRL experiences, qualitative studies do not offer the effect estimates needed for quantitative comparison across different studies.

2.4. Quality Assessment

The Newcastle–Ottawa Scale (NOS), modified for cross-sectional studies, was employed to assess the study quality [24]. NOS assesses quality by examining content, structure, and interpretation, demonstrating high reliability and validity, and performing better than other scales. Its main dimensions include outcomes, comparability, and selection. Seven categories evaluate aspects such as sample representativeness, justification for sample size, comparability between respondents and non-respondents, exposure estimates, comparability based on study design or analysis, the evaluation of results, and the adequacy of statistical analysis.

If the study satisfies these criteria, a maximum of ten stars can be awarded overall. Up to three stars are allocated for the selection dimension, four stars for comparability, and three stars for the outcome dimension [24].

However, it is important to acknowledge that NOS has certain limitations: it was initially designed for observational studies, which may lead to some subjectivity when applying it to different study designs. To address this, two independent reviewers evaluated each article, and in cases of disagreement, a third researcher reviewed the article.

3. Results

3.1. Search Results

Eight hundred (800) articles were found related to how SRL relates to the academic process of health sciences students. Of these, 443 were discarded as duplicates, and 221 were discarded after reviewing their titles and abstracts. Out of 136 full-text articles evaluated for eligibility, 97 were excluded. The main reasons for exclusion were non-cross-sectional study designs (n = 58), populations other than undergraduate health sciences students (n = 10), not applying a face-to-face modality (n = 21), and inadequate outcome data (n = 8).

3.2. Characterization of the Articles

The details of the articles and the results of this review are organized in Table 1. The sample sizes of the 39 articles range from 23 to 6546, totaling 24,835 participants across all the scientific articles analyzed. Additionally, the 39 articles were developed in different countries around the world: China (n = 7), Turkey (n = 6), the Netherlands (n = 5), Canada (n = 4), Iran (n = 3), Pakistan (n = 3), Saudi Arabia (n = 2), Australia (n = 2), USA (n = 3), Chile (n = 1), Indonesia (n = 1), Malaysia (n = 1) and Singapore (n = 1) (Figure 2).

Table 1. Summary of Selected Works.

Author(s) and Year	Country	Sample Size	Instrument	Main Findings
Jouhari et al., 2016 [25]	Iran	360	LASSI	The highest mean score on LASSI was for testing strategies (28.67 ± 4.44) and the lowest for self-assessment (21.91 ± 4.91). Significant sex differences appeared in main idea selection, attitude, and self-assessment, with higher scores among students with better academic performance.
Demirören et al., 2016 [26]	Turkey	561	SRLPS	Females outperformed males in SRLP ‘planning and goal setting’ (p = 0.017) and SPBL ‘accountability’ (p = 0.003). Second-year students scored higher than third-year students on SRLP ‘lack of self-direction’ (p = 0.001).
Lucieer et al., 2016 [27]	The Netherlands	949	SRLPS	Reflection improved significantly during medical school, differing between first and third years. A small positive link was found between monitoring, reflection, effort, and first-year GPA, but only effort related to third-year GPA. This suggests these skills have limited impact on overall academic performance due to their minor changes.
Salamonson et al., 2016 [28]	Australia	563	MSLQ	The study showed a positive link between greater sense of coherence and higher academic performance. It found that 40% of students had low, 42% medium, and 18% high sense of coherence, suggesting it may be key for successful adaptation to higher education.
Lucieer, van der Geest et al., 2015 [29]	The Netherlands	384	SRLPS	The study showed a decline in self-regulated learning skills in preclinical medicine, with some skills stagnating or declining over two years.
Denat et al., 2017 [30]	Turkey	524	SRLPS	The mean SRLSS score for nursing students was 149.34 ± 18.92. A significant difference existed between perceived academic success and SRLSS scores. First-year students scored higher on total SRLSS and subdimensions. No significant differences were found based on gender, age, school, or residence.
Cho et al., 2017 [31]	Australia	72	MSLQ	Extrinsic goal orientation increased significantly (p < 0.033), while metacognitive self-regulation decreased (p < 0.001). Prior clinical experience was negatively linked to metacognitive self-regulation (OR 5.0; p < 0.035).
Berkhout et al., 2017 [32]	The Netherlands	74	Own Method	Participants in 11 residency programs (e.g., family medicine, internal medicine, pediatrics) showed five self-regulated learning patterns, explaining 43% of the variance. Students cited a lack of clinical support, negatively affecting their active learning.
Cheema et al., 2019 [33]	Pakistan	339	MSLQ	Self-efficacy, time management/study environment, peer learning, and help-seeking declined from first to third year. Academic performance was weak but linked to intrinsic goal orientation (r = 0.123, p = 0.031), extrinsic goal orientation (r = −0.167, p = 0.003), elaboration (r = 0.182, p = 0.001), time management/study environment (r = 0.149, p = 0.009), and effort regulation (r = 0.148, p = 0.009). Females scored higher on task value (p = 0.009), males higher on self-efficacy (p = 0.002) and critical thinking (p = 0.012). Extrinsic goal orientation decreased significantly in the third year compared to preclinical years (p < 0.001).
Siddiqui & Malik, 2019 [34]	Pakistan	96	Own Method	The mean self-regulation score was 71.56 ± 11.19. Reflection scored highest (34.08 ± 1.30), planning lowest (27.01 ± 1.33). Students exhibited high levels of reflection but low levels of planning and monitoring.
Qaiser et al., 2020 [35]	Pakistan	253	SRLPS	There were no significant differences in self-regulated learning by gender (p < 0.2) or academic year (p < 0.4). Mean scores were 3rd year (145.83 ± 16.6), 4th year (138 ± 31.4), and 5th year (146.7 ± 21.59), with cutoffs of 102, 107, and 111. Eight students with lowest SRLPS scores were selected for focus groups. Participants identified four key factors negatively affecting self-regulated learning: curriculum, metacognitive strategies, infrastructure, and policies.
Demirören et al., 2020 [36]	Turkey	257	SRLPS, SPBL, TES	The mean scores were 3.68 for SRLP (self-regulation). SRLP positively correlated with SPBL (r = 0.55; p = 0.000) and TES (r = 0.26; p = 0.000), but not with achievement (r = 0.11; p = 0.073).
S. S. Lee et al., 2020 [37]	Singapore	457	MSLQ	NUS had the lowest extrinsic goal orientation, while UM scored lowest in self-efficacy and test anxiety. Help-seeking was lowest at UM (4.85) and NUS (4.89). UM also reported the lowest in Time Management/Study Environment (4.91). Both institutions had high intrinsic motivation.
Yang et al., 2020 [38]	China	2052	SRLS For College Students	The study found predictors of self-regulated learning (SRL) in medical students. Nomograms with AUCs of 0.733 and 0.749 predict motivation and learning strategies, highlighting the importance of self-regulation for better medical education outcomes.
Zheng & Zhang, 2020 [39]	USA	146	MSLQ	Help-seeking strategies are positively associated with learning outcomes among first- and second-year students, whereas cognitive strategy rehearsal is negatively associated. The study highlights the importance of self-regulated learning strategies for success in flipped classrooms.
Ghorbani Yekta et al., 2020 [40]	Iran	190	MSLQ	Intrinsic goal orientation boosts achievement motivation, leading to better academic results for self-regulated students.
J. Zheng et al., 2020 [41]	Canada	50	Analysis of SRL conduct	The SRL behaviors of ordering tests and verifying evidence positively predicted efficacy and confidence in an accurate diagnosis. However, linking evidence negatively predicted efficacy.
Gandomkar et al., 2020 [42]	Iran	76	MSLQ	The research revealed significant correlations: the metacognitive monitor microanalysis was linked to SRL, and the self-efficacy microanalysis showed a positive relationship with self-efficacy measured via the questionnaire.
B. Zheng et al., 2021 [43]	USA	146	MSLQ	The study showed self-efficacy had the highest score among students, followed by intrinsic motivation (4.59), resource management (4.48), metacognitive strategies (4.46), extrinsic motivation (4.24), and cognitive strategies (4.17).
Elfakki et al., 2021 [44]	Saudi Arabia	262	SRLQ	No correlation was identified between self-regulated learning (SRL) and academic achievement among medical students. A notable gender disparity was observed in SRL, with female students outperforming their male counterparts. Both genders, particularly males, demonstrated insufficient utilization of goal-setting and active learning strategies.
Li et al., 2021 [45]	Canada	23	Analysis of SRL conduct	Emotional frequency and variability affected clinical reasoning differently across SRL phases. Emotional variability negatively predicted performance in all phases. Emotional frequency during self-reflection was significantly associated with performance. Students had neutral affect in prediction and strong emotions in performance and self-reflection.
Kurt & Eskimez, 2022 [46]	Turkey	614	SRLS-CNP	Self-regulated learning in clinical practice was higher among women, first-year students (who showed greater motivation and learning strategies; p < 0.05), those interested in nursing, and those who viewed academics as role models. Fourth-year students felt 3.8 times more competent than first-years (p < 0.05). 84.2% reported needing help with clinical applications.
Zhang et al., 2022 [47]	China	193	SLSU	Personal factors like clear career plans and role models significantly predict self-regulated learning. Likewise, social support from mentors and peers also plays a key role.
Dogu et al., 2022 [48]	Turkey	417	SRLS-CNP	Nursing students exhibited high levels of SRL (60.28 ± 11.47) and moderate levels of self-efficacy (SE) (62.72 ± 11.04). A weak but significant positive correlation was observed between SRL and SE (r = 0.349, p < 0.001), indicating that as SRL increased, SE tended to increase as well.
Stoffels et al., 2023 [49]	The Netherlands	690	NCFQ Questionnaire Subsection-Perception/SRLWQ—SRL/BPN Scale	A positive learning atmosphere encourages self-regulated learning (SRL), which is fully mediated by the satisfaction of Basic Psychological Needs (BPN). Furthermore, SRL partially mediates the relationship between the learning atmosphere/BPN and perceived learning.
Subaş & Karaçay, 2023 [50]	Turkey	764	SRLS-CNP	Being female, having higher grades, and voluntarily choosing nursing are associated with higher self-regulated learning (SRL). Students comfortable in clinical practice and good at transferring knowledge showed higher SRL. Those who disliked the profession faced greater difficulties and had lower SRL. A high student-to-faculty ratio may limit interaction, lowering SRL.
L. Li et al., 2023 [51]	China	878	College Student SRL Questionnaire	The relationship between SRL and learning strategies was analyzed with multilinear regression. Factors influencing SRL in medical specialties remain unclear and need more study.
Romli et al., 2023 [52]	Malaysia	63	MSLQ	Metacognitive self-regulation, study time and environment, effort regulation, and help-seeking all scored below 5.00, indicating areas for improvement.
Wang et al., 2023 [53]	Canada	82	Fuzzer Miner Model	The complexity of the tasks influenced the timing of self-regulated learning (SRL) behaviors: students demonstrated more analytical reasoning and iterative processes in complex tasks than in simple ones.
Wang, Li, & Lajoie, 2023 [54]	Canada	27	Analysis of SRL conduct	The analysis showed cognitive load hampers diagnostic efficiency, with self-reflection (SRL) mediating. Higher load reduces SRL, lowering performance. The study confirms the link between load, SRL, and diagnostic efficiency.
Faisal E., 2023 [55]	Saudi Arabia	639	Analysis of SRL conduct	The study found positive correlations among self-regulation strategies: note-taking (NT), environmental structuring (ES), self-evaluation (SE), and self-consequence (SC). Self-evaluation had the highest mean (m = 4.14, SD = 0.59), whereas self-consequence had the lowest mean (m = 3.43, SD = 0.70).
Andrews et al., 2023 [56]	USA	44	Analysis of SRL conduct	Microprocesses of self-regulated learning (SRL) were studied in 44 internal medicine interns, focusing on strategic planning, metacognitive monitoring, causal attributions, and adaptive inferences during MCQ resolution. In planning, 45% prioritized relevant traits, but only 11% mentioned higher-order reasoning. During performance, 86% focused on integration, and 55% on diagnosis. In self-reflection, 93% identified management shortcomings, and 84% criticized their skills. Less than 10% planned to change their diagnostic approach.
H. Zhang et al., 2023 [57]	China	359	SRLAS	A significant positive relationship exists between self-efficacy and self-regulated learning (SRL) in nursing students (β = 0.417, p < 0.001). Positive academic emotions also directly impact SRL (β = 0.232, p < 0.001). Difficulty does not directly affect SRL but influences it indirectly through self-efficacy (77.778%) and positive emotions (14.184%).
Lepe-Martínez et al., 2024 [3]	Chile	101	Learning Management Questionnaire	The study found a significant association among executive functions, SRM, and academic achievement among first-year health sciences and medicine students. Moderate differences in SRM were observed based on performance (η2 = 0.071).
Huang et al., 2024 [58]	China	2246	College Student SRL Questionnaire	The study found that the educational system, parental background, parenting style, household income, parental occupation, GPA, and participation in SRL activities and training were significantly linked to higher SRL scores (p < 0.05). Specifically, the “5 + 3” educational system (p < 0.001; 95% CI, 9.842 to 15.686), GPA (p < 0.001; 95% CI, −10.009 to −5.712), engagement in SRL activities (p = 0.001; 95% CI, −11.903 to −3.162), and participation in SRL training (p < 0.001; 95% CI, −15.966 to −9.972) were positively associated with these skills.
Findyartini et al., 2024 [59]	Indonesia	1122	MSLQ	Males used social media less for learning than females (80 vs. 82, p < 0.007). Public university medical students showed more positive attitudes toward social media for learning than private university students (83 vs. 81, p < 0.007). Significant SRL differences were observed by educational stage and university type (scores: 426 vs. 418, p < 0.003; and 436 vs. 418, p < 0.001).
Zhouchen et al., 2024 [21]	China	322	College Student SRL Questionnaire	Self-regulated learning mediates the relationship between professional identity and student well-being (95% CI: 0.366–0.548, p < 0.001).
Broks et al., 2024 [60]	The Netherlands	1894	MSLQ	Three profiles of first-year medical students were identified: high TA/high SRL (36.6%), low TA/high SRL (34.9%), and moderate TA/low SRL (28.5%). Optimal performance was highest in the low TA/high SRL group (82.5%), then high TA/high SRL (71.9%), and moderate TA/low SRL (65.2%). Students in the low TA/high SRL group reported less stress (PSS = 23.9) than the others (≈28). Women are more likely to have the high TA/high SRL profile (Cramer’s V = 0.13, small).
Tian et al., 2025 [61]	China	6546	Own Scale	Teacher feedback predicted self-regulated learning (β = 0.136, p < 0.001). Teacher-student interaction mediated this relationship (effect size 0.116; 37.91%) and a sense of school belonging (effect size 0.045; 14.71%).

Figure 2. Distribution of the 39 articles obtained from the PRISMA protocol.

Based on the results of this review (see Table 1), it was possible to identify five central themes associated with SRL in the academic training of university students in the health field.

Due to variations in study design, measurement instruments for SRL, and how outcomes were reported, a meta-analysis was not conducted. Instead, results were summarized through a structured narrative, categorizing studies by outcome domains like academic performance, psychological well-being, and sociodemographic factors, and describing the direction and strength of the associations found.

1.

Relationship between SRL and Academic Performance

There is substantial evidence suggesting a predominantly positive association between SRL skills and the academic performance of health sciences students. Numerous studies report that students with higher Grade Point Average (GPA) scores perform significantly higher on SRL strategies and executive functions [3,25,58]. Moreover, students’ profiles characterized by high SRL and low-test anxiety have been reported to be associated with optimal academic performance [60].

While the evidence is extensive on this point, it is not uniform. Some studies did not find a significant direct association between SRL levels and GPA [36,44] or only found isolated components of SRL, such as effort regulation, that predicted academic performance in higher grades [27]. Furthermore, key mediating variables such as sense of coherence [28] and self-efficacy [43] were identified.

2.

Sociodemographic Variables and Curriculum Evolution

When analyzing demographic variables, a consistent gender trend is observed, where female students frequently obtain better SRL results globally or in sub-areas such as planning and accountability, compared to male students [26,46,50], although this difference was not statistically significant in all cultural contexts analyzed [30,35]; however, findings were not entirely consistent across all included studies.

A key finding relates to the evolutionary trajectory of SRL throughout the student training process. Contrary to expectations, multiple students report a stagnation or even a decline in SRL-related skills among upper-year students compared to first-year students [29,33]. This negative trend is observed primarily in the dimensions of metacognition and help-seeking [31], which could be explained by the fact that advanced courses or clinical immersion may slow down or even inhibit the continued development of autonomy in learning.

3.

Psychological Determinants: Self-efficacy, Motivation, and Emotion

Personal factors emerge as strong predictors of self-regulated behavior. Self-efficacy demonstrates a strong positive correlation with SRL capacity [48,57], frequently acting as a necessary precursor of the adoption of effective learning strategies. Intrinsic motivation, achievement orientation, and a defined professional identity also facilitate high levels of SRL [21,40,47].

However, the affective dimension plays an important regulatory role. Positive academic emotions promote the development of SRL [57], while unstable emotional variability has been negatively linked to performance in clinical reasoning [21]. Additionally, high levels of SRL act as a protective factor because they are associated with lower levels of stress and higher levels of student well-being [60].

4.

Influence of the Educational Environment and Cognitive Load

The educational context exerts a decisive influence on the quality of SRL. Effective teacher feedback and a pedagogical atmosphere that meets basic psychological needs are strong predictors of the development of SRL [49,61]. On the other hand, clinical settings are often perceived as lacking in support, which hinders participation in active learning [32].

In line with the above, the characteristics of the activities moderate the self-regulatory response. Experimental and microanalysis studies indicate that cognitive load and task complexity can alter SRL processes, promoting analytical reasoning but negatively affecting the self-reflection phase and diagnostic efficiency [53,54]. This underscores the importance of designing academic activities that balance cognitive challenge with appropriate instructional support.

5.

Profiles of Microanalytical Strategies and Processes

When the components of SRL are analyzed individually, a disparity is observed among health science students. On the one hand, reflection and task evaluation strategies generally show high scores [34], while frequent deficiencies are observed in critical skills such as planning, metacognitive monitoring, and seeking help [37,52,56].

Studies employing microanalysis methodologies show that, during clinical problem-solving, students frequently fail to adapt their strategies after making mistakes or do not employ higher-order reasoning processes in the planning phase [42,56]. However, specific methodologies such as the flipped classroom have demonstrated potential for improving collaborative strategies such as peer learning [39].

3.3. Analysis of Methodological Quality

Table 2 shows the results of the methodological quality analysis of the studies. Overall, the quality of the studies was satisfactory, with an average of 5.46 stars. The weakest point examined was the selection process, as most of the analyzed studies exhibited weaknesses stemming from their sample collection methods: most were non-probability, convenience, or snowball sampling methods. The highest score was seven stars, where eight articles obtained this score [21,31,38,42,47,50,60,62], and the lowest score was two stars with two works with this score [34,44].

Table 2. Analysis of the Quality of the Articles.

Authors	Year	Selection	Comparability	Outcomes
Jouhari et al. [25]	2016	★★★	★★☆☆	★★☆
Demirören et al. [26]	2016	★☆☆	★★★☆	★★☆
Lucieer, Jonker et al., 2016 [27]	2016	★☆☆	★★☆☆	★★★
Salamonson et al. [28]	2016	☆☆☆	★★★★	★★★
Lucieer, van der Geest et al., 2015 [29]	2016	☆☆☆	★★☆☆	★★☆
Denat et al. [30]	2017	★☆☆	★☆☆☆	★★☆
Cho et al. [31]	2017	★☆☆	★★★★	★★☆
Berkhout et al. [32]	2017	★☆☆	★☆☆☆	★★☆
Cheema et al. [33]	2019	★☆☆	★★★☆	★★☆
Siddiqui & Malik [34]	2019	☆☆☆	★☆☆☆	★☆☆
Qaiser et al. [35]	2020	★★☆	★★☆☆	★★☆
Demirören et al. [36]	2020	★★★	★★☆☆	★★☆
Lee et al. [37]	2020	★☆☆	★★☆☆	★☆☆
Yang et al. [38]	2020	★☆☆	★★★★	★★☆
B. Zheng & Zhang [39]	2020	☆☆☆	★★★★	★★★
Ghorbani Yekta et al. [40]	2020	☆☆☆	★★☆☆	★★☆
J. Zheng et al. [41]	2020	☆☆☆	★★☆☆	★★★
Gandomkar et al. [42]	2020	☆☆☆	★★★★	★★★
B. Zheng et al. [43]	2021	★☆☆	★★☆☆	★★★
Elfakki et al. [44]	2021	☆☆☆	★☆☆☆	★☆☆
S. Li et al. [45]	2021	☆☆☆	★★☆☆	★★★
Kurt & Eskimez [46]	2022	★★☆	★☆☆☆	★★☆
J. Y. Zhang et al. [47]	2022	★☆☆	★★★☆	★★★
Dogu et al. [48]	2022	★★☆	★★☆☆	★★☆
Stoffels et al. [49]	2023	★☆☆	★★☆☆	★★☆
Subaş & Karaçay [50]	2023	★★☆	★★★☆	★★☆
L. Li et al. [51]	2023	★☆☆	★★★★	★★☆
Romli et al. [52]	2023	☆☆☆	★★☆☆	★★☆
Wang, Li, Huang et al., 2023 [53]	2023	☆☆☆	★☆☆☆	★★☆
Wang, Li, & Lajoie, 2023 [54]	2023	☆☆☆	★★☆☆	★★★
Faisal E. [55]	2023	★★☆	★★☆☆	★★☆
Andrews et al. [56]	2023	☆☆☆	★★☆☆	★★☆
H. Zhang et al. [57]	2023	☆☆☆	★★☆☆	★★☆
Lepe-Martínez et al. [3]	2024	☆☆☆	★★☆☆	★★★
Huang et al. [58]	2024	★☆☆	★★★☆	★★☆
Findyartini et al. [59]	2024	★★☆	★★☆☆	★★☆
Zhouchen et al. [21]	2024	★☆☆	★★★★	★★☆
Broks et al. [60]	2024	☆☆☆	★★★★	★★★
Tian et al. [61]	2025	☆☆☆	★★★★	★★☆

Note: ★: The article does meet the criteria; ☆: The article does not meet the criteria.

4. Discussion

39 scientific studies that met the established inclusion criteria were analyzed, encompassing a cumulative sample of 24,835 undergraduate health sciences students. These cross-sectional studies were published between 2015 and 2025 and predominantly used standardized instruments such as the MSLQ, SRLPS, and SRLS-CNP to assess SRL. The evidence shows remarkable geographical diversity, with studies from 13 countries, notably China (n = 7), Turkey (n = 6), the Netherlands (n = 5), and Canada (n = 4). However, a limited representation of research from Latin American contexts was identified, with only one study conducted in Chile, highlighting a significant gap in regional scientific output. This finding opens a critical opportunity for future research in Latin America, where the study of SRL in health sciences students is still not sufficiently developed, despite its potential impact on improving academic performance and student well-being.

Regarding SRL, a consistent trend is observed with respect to gender, suggesting that female students exhibit higher level of SRL compared to males, which could be explained because women present a greater sense of responsibility and commitment to their training than men [26,44,50,63], as well as the fact that women show more goal-setting strategies, responsibility and planning, because it has been seen that they take better ownership of their training process. However, it is important to note that the gender gap is not universal across all the cultural environments analyzed, suggesting that contextual factors may mediate this relationship [30,35]. Beyond gender, other sociodemographic factors emerge as conclusive indicators of SRL behavior. Variables such as pre-university education, parents’ educational level, household income, and prior participation in occupational risk prevention training significantly influence students’ self-management skills [58]. Additionally, significant differences exist between students from different types of institutions, with public students reporting higher levels of SRL than private university students. This could be explained by two structural factors: (i) differences in admission requirements, given that public institutions tend to have more competitive selection processes, attracting students with strong academic profiles and a greater prior development of autonomy [59]. Therefore, it is suggested that SRL is not only an individual process but is directly influenced by cultural capital and the opportunities for advancement offered by the student’s socio-educational environment.

Undoubtedly, one of the most counterintuitive findings of this systematic review was the “stagnation paradox” in the development of SRL among health sciences students in their final years of study. Contrary to expectations that academic maturation would strengthen SRL competencies as students’ progress through their studies, the evidence shows that these skills do not develop linearly and may even regress during the preclinical and clinical stages [29]. Consistently, a significant decrease or even a high level of decline was observed in key dimensions such as time management, help-seeking, and metacognition in senior students compared to first-year [31,33].

This phenomenon suggests that the traditional organization of health training may be acting as an inhibitor of higher-order processes. To interpret this, the interaction between curriculum design and the demands of the clinical environment could be analyzed through the lens of cognitive load theory, which posits that working memory saturation prevents the activation of metacognitive processes [64]. Therefore, the progressive increase in the complexity of clinical tasks could not only affect diagnostic efficiency but also critically reduce the frequency of self-reflexive behaviors [54]. Moreover, when faced with highly demanding tasks, students prioritize immediate analytical reasoning to solve problems, neglecting the monitoring and evaluation processes of their own performance, which are fundamental components of self-reflective learning [53].

In addition, students often perceive clinical practice settings as spaces with little or no teaching support, which influences their ability to actively participate in their own learning [32]. This lack of external structure at preclinical and clinical levels could hinder the development of the autonomy required to complete the SRL cycle.

While more advanced students possess greater theoretical knowledge, they exhibit a notable deficiency in their ability to adapt strategies after making mistakes. Microanalytical evidence shows that less than 10% of students in clinical settings indicate that they would adjust their reasoning process in the face of future difficulties, revealing shortcomings in higher-order processes during planning [42,56].

Moreover, professional identity is positively correlated with student well-being, where SRL acts as a critical mediator in this relationship [21]. However, if SRL levels decrease, academic well-being is compromised, increasing the risk of stress and burnout, especially in the final stages of academic studies [21,60].

Therefore, the development of SRL in health science students does not evolve spontaneously, and the stagnation reported in this review highlights the need to include strategies such as the flipped classroom, which has shown satisfactory results with respect to academic performance and student satisfaction [65], and support programs that provide proper continuous development throughout professional training, to prevent the demands of the clinical environment from stifling the student’s reflective capacity.

The evidence found in this systematic review confirms that SRL promotes academic success and is a fundamental pillar for the mental health of healthcare students. A close relationship has been described where developing strategies that promote SRL act as a protective factor against the prevalence of symptoms associated with depression, stress, burnout, and anxiety [17]. In healthcare students, SRL shows a negative correlation with depression, explained by the strengthening of resilience mechanisms against stressors in the academic environment [18].

This mitigating effect occurs through various psychological and operational mechanisms identified in the analyzed studies. The first is a negative correlation between stress and SRL skills, where increased levels of perceived stress are associated with a decrease in students’ ability to develop SRL skills [19]. Furthermore, the development of SRL facilitates clinical decision-making, directly contributing to a reduction in student anxiety levels [20]. Additionally, adopting SRL strategies, such as time management, acts as a protective factor against burnout in university students [1] and even students characterized by high levels of SRL report very low levels of stress compared to those with low or even moderate metacognitive skills [60]. Similarly, academic well-being has been identified as directly influencing a student’s SRL-regulation capacity, impacting the construction of their professional identity [21]. Low levels of well-being can lead to cycles of low SRL, resulting in emotional exhaustion and difficulties in developing a professional identity [21]. Finally, self-efficacy emerges as a necessary precursor for the incorporation of effective academic strategies, demonstrating that as students’ confidence in their abilities is fostered, their SRL levels increase [48,57].

As a result, the importance of intentionally promoting the development of SRL skills from the early years of university becomes evident. This proactive approach, in addition to optimizing the educational process and curricular progress, acts as a strong protective factor against demotivation and academic dropout in demanding contexts [5,9]. Consequently, facilitating SRL is a vital strategy for developing resilient, competent healthcare professionals with enhanced adaptability skills required in higher education.

The results are conclusive regarding the predominantly positive relationship between SRL skills and the academic performance of healthcare students. A large corpus of evidence suggests that high-achieving students exhibit higher levels of SRL strategy use and greater proficiency in key executive functions such as monitoring and cognitive flexibility [3,25,58]. Additionally, profiles with high SRL capacity and low-test anxiety are reported to be associated with positive academic outcomes [60]. In this regard, a student’s ability to consciously guide their own study is fundamental, especially when control of the learning process rests with the student [10].

However, it is important to highlight that this relationship is not consistent across the analyzed literature. Some studies found no direct or significant relationship between SRL scores and academic success [36,44]. In other contexts, only specific components proved to be predictors of academic performance in later stages of the curriculum [27]. This inconsistency could be explained by the fact that the influence of SRL on grades is not linear, but rather conditioned by critical mediating variables identified in this study, such as self-sufficiency and sense of coherence [28,43]. In line with the above, self-efficacy has a preeminent precursor role in enabling students to implement effective learning strategies and achieve higher academic goals [40,48].

5. Strengths and Limitations

First, the PRISMA protocol was strictly followed to ensure transparency and replicability of the process [22]. The inclusion of 39 scientific studies allowed for the consolidation of a robust sample of 24,835 participants, providing robust statistical representativeness. Likewise, geographical diversity was achieved across multiple countries on different continents, enabling a global and transcultural perspective on the phenomenon of SRL in healthcare students (13 countries included). Finally, the Newcastle–Ottawa scale (NOS) was used for the standardized assessment of methodological quality, identifying critical gaps in the current literature [24].

The main strength of this study is that it constitutes one of the few, if not the first, systematic review that integrates and consolidates the evidence on SRL in students from different healthcare disciplines, going beyond previous reviews focused exclusively on medicine and nursing. This multidisciplinary approach allows the findings on the “stagnation paradox” to be generalized across different disciplines within the field. Additionally, the cumulative sample of 24,835 participants from 13 different countries provides adequate robustness with respect to the medical education literature, along with cross-cultural diversity of the data, strongly suggesting that the decline in SRL during clinical rotations is not an isolated finding, but rather a global systemic pattern.

Among the main limitations of this systematic review is the fact that all the included studies employed a cross-sectional design, which limits the possibility of establishing causal relationships between SRL and the variables analyzed. Also, this review did not include qualitative studies; despite the importance of these studies, they did not provide measurable and/or comparable data about the phenomenon. Longitudinal studies were not included to ensure methodological homogeneity and reduce the variability that could occur in such designs. Moreover, the overall methodological quality of the studies was rated as moderate, reinforcing the need for future research with more robust designs, such as longitudinal or quasi-experimental studies. Finally, it is worth noting that the studies relied on self-reported instruments, which may introduce biases associated with participants’ subjective perceptions and limit the external validity of the findings.

6. Conclusions

In conclusion, this systematic review synthesized evidence from 39 studies using rigorous selection criteria focused on undergraduate health sciences students across multiple global contexts. Going beyond a mere listing of findings, the integrated evidence demonstrates that SRL is a critical input that transcends academic performance, serving as an essential pillar for adapting to university life, student well-being, and the comprehensive training of healthcare professionals. It also confirms that high levels of SRL are associated with better academic performance and act as a protective factor against stress, anxiety, and burnout, especially in highly demanding environments.

However, the results revealed a “stagnation paradox”, because we suggest SRL skills may not develop spontaneously with academic progress, even declining in later years due to increased cognitive load and lack of institutional support in clinical settings. This highlights the need to develop SRL intentionally, systematically, and from the early years of the program.

In conclusion, universities in the healthcare field must implement teaching strategies that balance academic challenges with effective faculty support. Promoting lifelong learning is not only a measure to improve academic performance, but also a comprehensive way to ensure retention, well-being, and the development of competent and resilient professionals prepared for lifelong learning.