Trajectory changes of clinical empathy ability of Chinese nursing interns: A longitudinal study using latent transition analysis.

Trajectory changes of clinical empathy ability of Chinese nursing interns: A longitudinal study using latent transition analysis - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Medicine (Baltimore) . 2026 May 29;105(22):e49054. doi: 10.1097/MD.0000000000049054 Trajectory changes of clinical empathy ability of Chinese nursing interns: A longitudinal study using latent transition analysis Mengmeng Hu a, b , Xin Liu b , Li Zhou b , Hongjuan Chang b , Chun Zhang c, * Author information Article notes Copyright and License information a China Resources & WISCO General Hospital, Wuhan, Hubei, China b Wuhan University of Science and Technology Medical School, Wuhan, Hubei, China c The Third People’s Hospital of Hubei Province, Wuhan, Hubei, China. * Correspondence: Chun Zhang, The Third People’s Hospital of Hubei Province, Wuhan 430033, Hubei, China (e-mail: 55132838@qq.com ). Received 2026 Jan 17; Revised 2026 May 7; Accepted 2026 May 11; Collection date 2026 May 29. Copyright © 2026 the Author(s). Published by Wolters Kluwer Health, Inc. This is an open access article distributed under the Creative Commons Attribution License 4.0 (CCBY) , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. PMC Copyright notice PMCID: PMC13225516  PMID: 42216316 Abstract This study utilized a longitudinal design and latent transition analysis to examine the distinct profiles, transitional trajectories, and influencing factors of clinical empathy among nursing interns. An 8-month follow-up survey, initiated in June 2024, was conducted among 762 nursing interns from 2 Grade A tertiary teaching hospitals in Hubei Province, China. Results identified 4 discrete latent profiles: low empathy-stable, low empathy-increasing, high empathy-stress, and high empathy-distress. The low empathy-stable profile demonstrated the lowest stability (7% maintenance probability) and a significant tendency to transition to the high empathy-stress profile (43% probability). In contrast, the maintenance probabilities for the remaining profiles ranged from 18% to 44%. Multivariate logistic regression revealed that gender, physical exercise, peer relationships, and reading empathy-related literature significantly predicted transition pathways ( P < .05). Additionally, emotional intelligence served as a significant positive moderator in empathy development ( P < .05). These findings highlight the dynamic nature of empathic capacity during clinical practicums. To promote long-term career development and prevent emotional exhaustion, hospitals and nursing schools should implement tailored emotional intelligence interventions, particularly for interns within the stress and distress-dominant empathy profiles. Keywords: emotional intelligence, empathy, latent profile analysis, latent transition analysis, nursing interns, psychological growth 1. Introduction Within the framework of the biopsychosocial model, health transcends the mere absence of disease; rather, it is conceptualized as a holistic state of optimal physical, psychological, and social well-being. [ 1 ] Modern nursing practice has imposed deeper requirements on nursing professionals. In addition to providing basic care, nurses are expected to offer psychological support to patients. [ 2 ] Empathy refers to the capacity to understand and resonate with the emotions, thoughts, and needs of others, thereby generating analogous emotional experiences. It is conceptualized as a psychological process wherein an individual assumes the perspective of another, comprehends their viewpoints, and vicariously experiences their emotions. [ 3 ] Empathy is a cornerstone for fostering harmonious nurse–patient relationships and is one of the essential qualities that nursing personnel should possess. [ 4 ] The empathic abilities of nurses can assist patients in better managing the psychological stress induced by illness, enhance their courage and confidence in confronting challenges, and serve as a vital psychological resource during the recovery process. [ 5 ] For nursing interns, their empathic abilities significantly influence their future professional attitudes and the development of their role personalities. [ 6 ] According to the emotional intelligence (EI) model proposed by Rivers and Brackett et al, EI is defined as an individual’s capacity to adaptively perceive, understand, and regulate the emotions of oneself and others, as well as to utilize emotions for effective problem-solving. High levels of EI enable individuals to mitigate psychological distress and buffer against the adverse effects of negative emotions. [ 7 ] In the context of nursing education, EI is recognized as a vital psychological resource that underpins students’ clinical competence. [ 8 ] Furthermore, a study by Aghajani et al [ 9 ] involving nursing interns demonstrated that those with higher EI exhibit superior emotional awareness and management capabilities. This allows them to flexibly implement emotion regulation strategies and maintain appropriate emotional engagement. Conversely, nursing students with lower EI are more vulnerable to emotional exhaustion due to a deficiency in effective emotion regulation resources. In accordance with the Regulations on Nurses in China, undergraduate nursing students are mandated to complete a minimum of 8 months of clinical practicum prior to undertaking the nursing professional qualification examination. [ 10 ] This period extends beyond mere technical skills training; it constitutes a critical transitional phase from academic study to professional clinical practice. During this practicum, empathy serves as a fundamental prerequisite for cultivating therapeutic nurse–patient relationships and delivering personalized nursing interventions. Crucially, empathy is not a static disposition but rather a dynamic construct shaped by situational contexts. As nursing students systematically engage with clinical patients for the first time, their empathic capacity may exhibit nonlinear developmental trajectories. This variability is often driven by the complexities of real-world healthcare settings, repeated exposure to patient suffering, and discrepancies in available clinical support resources. However, most prior research relies on cross-sectional analyses that treat empathy as a continuous construct, thereby obscuring unobserved population heterogeneity and distinct developmental pathways. Latent transition analysis (LTA) addresses this limitation by identifying categorical latent statuses of nursing students based on their empathic capacity, and by quantifying the transition probabilities between these statuses across the clinical practicum. [ 11 ] This approach offers a nuanced perspective on the developmental dynamics of empathy. Accordingly, the present study utilizes LTA to delineate the latent profiles and transitional trajectories of empathy among Chinese nursing interns. By incorporating EI as a predictor, we further analyze its influence on these stage-sequential transitions. Ultimately, this research aims to yield robust, evidence-based insights to inform the design of targeted empathy cultivation strategies. 2. Method 2.1. Study design and participants A convenience sample of 762 nursing interns was recruited from 2 tertiary hospitals in Wuhan, Hubei Province, as the study participants. The 2 testing periods were the first week of the clinical internship and the final week after the internship, respectively, with an 8-month longitudinal follow-up investigation conducted between them. The first testing period (T1) occurred in June 2024, yielding 766 valid participants. The second testing period (T2) took place in March 2025, resulting in 762 valid participants. Seven hundred sixty-two nursing students who completed both T1 and T2 were included in this study, achieving a follow-up rate of 99.4%. This study was reviewed and approved by the Ethics Committee of the Medical College of Wuhan University of Science and Technology (Approval No: 2023177). All participants gave their informed consent to participate in this study. 2.2. Research tools 2.2.1. General information survey Based on a comprehensive literature review, the researchers developed a questionnaire themselves, which encompasses basic information such as gender, age, peer relationships, menstrual cycle regularity, and whether books related to empathy have been read. 2.2.2. Interpersonal Reactivity Index (IRI) Empathy was assessed using the IRI, originally developed by Davis (1980), [ 12 ] which has been extensively utilized to evaluate empathic capacity in medical and nursing populations. The present study employed a validated Chinese version of the scale, comprising 22 items distributed across 4 dimensions: perspective-taking, personal distress, fantasy, and empathic concern. Items are rated on a 5-point Likert scale. Higher aggregate scores indicate elevated levels of empathy among the participants. [ 13 ] In the current study, the Cronbach α coefficients for the scale were 0.79 at baseline (T1) and 0.84 at follow-up (T2). 2.2.3. Wong and Law Emotional Intelligence Scale EI was assessed using the scale developed by Wong and Law. [ 14 ] This instrument encompasses 4 dimensions: self-emotion appraisal, others’ emotion appraisal, use of emotion, and regulation of emotion, with each dimension comprising 4 items. Responses are rated on a 7-point Likert scale. Higher aggregate scores indicate elevated levels of EI among the participants. In the present study, the Cronbach α coefficients for this scale were 0.93 at baseline (T1) and 0.95 at follow-up (T2). 2.3. Testing procedure After obtaining informed consent from all participants, the research team briefed the students on the study’s objectives and provided clear guidelines for survey completion. Data collection was conducted longitudinally across the respective clinical practicum groups. At baseline (T1), sociodemographic characteristics alongside measures of clinical empathy and EI were collected. At follow-up (T2), only the empathy and EI scales were readministered. During each data collection session, 2 nursing postgraduate research assistants provided standardized verbal instructions, ensuring that the interns completed the questionnaires independently. All surveys were retrieved on-site immediately upon completion to ensure data integrity. 2.4. Data measurement Data management and preliminary descriptive analyses were performed using SPSS version 26.0. Subsequently, LTA was conducted via Mplus version 8.3 to identify the latent profiles and state transitions of the IRI. Model fit was evaluated using several established information criteria: the akaike information criterion, the Bayesian information criterion (BIC), and the sample-size-adjusted BIC, with lower values indicating superior model fit. Furthermore, entropy was used to assess classification precision; it ranges from 0 to 1, with values closer to 1 reflecting higher accuracy. Finally, the Lo–Mendell–Rubin adjusted likelihood ratio test and the bootstrapped likelihood ratio test were employed. Significant P -values ( P < .05) for these tests signify that a k -class model provides a significantly better fit than a ( k -1)-class model. [ 15 ] 3. Results 3.1. General demographic The average age of the nursing interns included in the sample was 21.79 ± 2.042 years. There were 113 males (14.8%) and 649 females (85.2%). 301 were specialized degree holders (39.5%) and 461 had a bachelor degree (60.5%). 358 exercised regularly (47%), 404 did not exercise regularly (53%), 233 had regular menstrual cycles (35.9%), and 416 had irregular menstrual cycles (64.1%). 102 had poor relationships with classmates (13.4%), 660 had good relationships with classmates (86.6%), 120 read books about EI (15.7%), and 642 did not read books about EI (84.3%). 3.2. Common method deviation test The Harman single-factor test was employed to assess the common method deviation. Unrotated principal component factor analysis was conducted on the measured item data collected at 2 time points. The results indicate that the variance explained by the first extracted common factor in both analyses is below the threshold of 40% (T1: 20.12%; T2: 30.4%). This suggests that this study has no significant common method bias issue. 3.3. Repeated measures analysis of variance A repeated measures analysis of variance was conducted to examine the differences in the main variables at 2 time points. [ 16 ] The results indicated that, in the IRI scale, significant differences were observed for opinion selection ( F = 7.47, P = .001), personal distress ( F = 3.671, P = .026), Fantasy ( F = 18.429, P < .01), empathic concern ( F = 24.647, P < .01), and the total score ( F = 70.23, P < .01). In the EI Scale, significant differences were also found for self-emotion appraisal ( F = 3.355, P = .047), regulation of emotion ( F = 20.87, P < .01), use of emotion ( F = 24.273, P < .01), others’ emotion appraisal ( F = 23.75, P < .01), and the total score ( F = 58.23, P < .01). See Table 1 . Table 1. Analysis of variance for repeated measurements at time points T1 and T2. item T1 T2 F P M ± SD M ± SD Perspective-taking 17.26 ± 6.015 15.67 ± 4.242 7.47 .001 Personal distress 8.87 ± 4.435 12.33 ± 4.685 3.671 .026 Fantasy 11.95 ± 3.707 11.39 ± 3.687 18.429 <.01 Empathic concern 12.91 ± 3.721 15.78 ± 4.109 24.647 <.01 Self-emotion appraisal 19.93 ± 4.779 20.34 ± 4.341 3.355 .047 Regulation of emotion 27.08 ± 7.271 18.42 ± 4.422 20.87 <.01 Use of emotion 9.97 ± 2.750 18.84 ± 4.502 24.273 <.01 Others’ emotion appraisal 19.36 ± 2.750 19.44 ± 4.508 23.75 <.01 IRI 46.74 ± 10.346 58.53 ± 14.494 70.23 <.01 WLEIS 76.34 ± 16.454 77.05 ± 14.733 58.23 <.01 Open in a new tab IRI = Interpersonal Reactivity Index, WLEIS = Wong and Law Emotional Intelligence Scale. 3.4. Latent transition analysis Using the response situations of the 4 dimensions from the T1 and T2 IRI Scales for Chinese nursing interns as initial models, one category was incrementally added each time. The fitting indicators of the resulting models were comprehensively analyzed to determine the optimal model. Five-category models were constructed at 2 time points (T1 and T2), and the fitting indices of each model were systematically compared. At the T1 time point, Model 4 demonstrated the best goodness-of-fit among all models, with an Entropy value of 0.886 (higher than Models 2, 3, and 5, with Entropy values of 0.851, 0.857, and 0.877, respectively). Additionally, the P values for the Lo–Mendell–Rubin adjusted likelihood ratio test and bootstraped likelihood ratio test were <.01, indicating statistical significance. Furthermore, the classification accuracy of Model 4 was more consistent with the actual situation, with an average probability of 93.0% across profile. At the T2 time point, the akaike information criterion, BIC, and sample-size adjusted BIC decreased progressively as the number of profile increased. The Entropy values of Models 4 and 5 were closest to 1 (0.896 and 0.925, respectively). However, when transitioning to a 5-category model, the Lo–Mendell–Rubin test results were no longer significant ( P > .05). These findings suggest that Model 4 provides a better fit than Model 5, with an average probability of 94.0% across profile. See Table 2 . Table 2. The LTA model at time points T1 and T2. Model AIC BIC aBIC Entropy LMR BLRT Category probability 1 17,665.984 17,703.072 17,677.668 – – – – 2 16,446.955 16,507.222 16,465.941 0.851 <0.01 <0.01 0.654/0.346 3 15,838.807 15,922.254 15,865.097 0.857 0.008 <0.01 0.502/0.304/0.194 4 15,458.163 15,564.790 15,491.755 0.886 <0.01 <0.01 0.188/0.088/0.271/0.453 5 15,234.459 15,364.265 15,275.353 0.877 0.08 <0.01 0.135/0.322/0.306/0.071/0.165 1 17,724.892 17,761.980 17,736.576 – – – – 2 16,644.598 16,704.866 16,663.585 0.854 <0.01 <0.01 0.571/0.429 3 16,169.203 16,252.650 16,195.493 0.863 0.019 <0.01 0.396/0.422/0.182 4 15,738.014 15,844.641 15,771.606 0.896 0.012 <0.01 0.189/0.321/0.347/0.142 5 15,464.605 15,594.412 15,505.500 0.925 0.766 <0.01 0.180/0.297/0.336/0.158/0.029 Open in a new tab Bold indicates the importance of the fourth model. AIC = akaike information criterion, aBIC = adjusted Bayesian information criterion, BIC = Bayesian information criterion, BLRT = bootstrapped likelihood ratio test, LMR = Lo–Mendell–Rubin. According to the characteristics of the conditional mean distribution map across the 4 dimensions of the IRI scale, the 4 profile were systematically named. As illustrated in Figures 1 and 2 , the mean values for each dimension of Profile 1 and 2 are relatively low and designated as “Low Empathy – Stable Profile” and “Low Empathy – Rising Profile,” respectively. For Profile 3 and 4, the mean values in opinion selection and personal distress gradually increase higher than those observed in other dimensions. Consequently, these profile are labeled “High Empathy – Stress Profile” and “High Empathy – Distress Profile,” respectively. Figure 1. Open in a new tab Categorization diagram of clinical empathy ability for Chinese nursing interns at Time Point T1. Figure 2. Open in a new tab Categorization diagram of clinical empathy ability for Chinese nursing interns at Time Point T2. 3.5. Latent transition analysis LTA were performed on the 4 latent profile at time points T1 and T2, with the results presented in Table 3 . The diagonal elements of the transformation matrix represent the probabilities that subjects remain in their original latent profile at different time points. Specifically, for the low empathy-stable category, the likelihood of remaining in this category is 7%. In contrast, the probabilities of transitioning to the low empathy-rising, high-empathy-stress, and high-empathy-distress profile are 14%, 43%, and 36%, respectively. The likelihood of maintaining this category for the low empathy-rising category is 18%, with transition probabilities to the low empathy-stable, high empathy-stress, and high-empathy-distress profile being 31%, 32%, and 19%, respectively. For the high-empathy-stress category, the likelihood of remaining in this category is 24%, with transition probabilities to the low empathy-stable, low empathy-rising, and high empathy-distress profile being 35%, 18%, and 24%, respectively. Finally, for the high-empathy-distress category, the probability of remaining in this category is 14%, with transition probabilities to the low empathy-stable, low-empathy-rising, and high empathy-stress profile being 27%, 19%, and 44%, respectively. Table 3. Analysis of latent transition in clinical empathy ability among nursing students at time points T1 and T2. Time T2 Low empathy-stable Low empathy-rising High empathy-stress High empathy -distress T1 Low empathy-stable 0.07 0.14 0.43 0.36 Low empathy-rising 0.31 0.18 0.32 0.19 High empathy-stress 0.35 0.18 0.24 0.24 High empathy-distress 0.27 0.15 0.14 0.44 Open in a new tab Bold indicates the importance of the fourth model. 3.6. Influencing factors on the occurrence ratio and transformation probability To investigate the effects of factors such as gender, exercise, regularity of menstrual cycles, peer relationships, reading empathetic literature, and EI on the latent profile of clinical empathy among nursing students during different internship stages, a multiple logistic regression analysis was conducted with the low empathy-stable profile serving as the reference category. The odds ratios (OR) were calculated to determine the ratios of probability changes for subjects transitioning into the low empathy-rising, high empathy-stress, and high empathy-distress profile relative to the low empathy-stable profile under the influence of covariates. The results are presented in Table 4 . Male interns were less likely to belong to the low empathy-rising and high empathy-stress profile than female interns. Neither the frequency of exercise nor the regularity of menstrual cycles significantly affected the classification outcomes. Peer relationships, however, showed notable differences: compared to poor peer relationships, good peer relationships were associated with a higher probability of belonging to the high empathy-stress and high empathy-distress profiles. Additionally, interns who had not read empathetic literature were likelier to belong to the low empathy-rising and high empathy-stress profiles than those who had. Finally, interns with high EI were less likely to belong to the high empathy-stress and high empathy-distress profiles compared to those with low EI. Table 4. The occurrence ratio of T1 potential probability under the influence of predictor variables. Factors Low empathy-rising type High empathy-stress type High empathy-pain type Gender (0 = female, 1 = male) 0.699 * 0.905 * 1.236 Exercise (0 = No, 1 = Yes) 0.856 1.338 1.094 Regular menstrual cycle (0 = No, 1 = Yes) 0.769 1.259 1.972 Peer relationship (0 = poor, 1 = yes) 0.423 1.106 * 1.807 * Empathetic literature reading (0 = No, 1 = Yes) 1.762 ** 2.113 * 1.127 Emotional Intelligence 0.958 0.754 ** 0.213 * Open in a new tab The regularity of menstruation is only investigated in females. * P < .05. ** P < .01. To investigate the influencing factors of latent category transformations in clinical empathy among nursing students during their internship period, variables such as gender, frequency of exercise, regularity of menstrual cycles, peer relationships, reading of empathy-related literature, and EI were incorporated as predictive variables in the latent class model. Subjects remaining in their original latent category were the reference group for multinomial logistic regression analysis. The results are presented in Table 4 below. An OR >1 indicates that the likelihood of a subject undergoing the specified transformation increases under the influence of the predictor variable; conversely, an OR <1 suggests a decrease in probability. According to Table 5 , male nursing students exhibit a reduced likelihood of transitioning from the low empathy-stable profile to the low empathy-rising profile (OR = 0.685) and the high-empathy-stress profile (OR = 0.685) compared to female nursing students. Nursing students who exercise regularly demonstrate a decreased likelihood of transitioning from the low empathy-stable profile to the high empathy-stress profile (OR = 0.794) and the high empathy-distress profile (OR = 0.312) relative to those who do not exercise frequently. Students with positive peer relationships are more likely to transition from low empathy-ascending to high empathy-distressful (OR = 2.010) than those with poor peer relationships. Furthermore, nursing students who have read books on EI exhibit a higher probability of transitioning from the high empathy-stress profile to the low empathy-upward profile (OR = 1.025) than those who have not read such literature. EI also influences the likelihood of transitions: the probability of the low empathy-ascending profile transforming into the low empathy-stable profile increases (OR = 1.023), as does the possibility of the high empathy-stress profile transforming into the low empathy-stable profile (OR = 1.008). Table 5. shows the occurrence ratio of the probability of changes in clinical empathy ability of intern nursing students under the influence of predictive variables. Factors Latent state Low empathy-stable Low empathy-rising High empathy-stress High empathy-distress Gender (0 = female, 1 = male) Low empathy-stable REF 0.684 * 0.685 * 1.157 Low empathy-rising 1.462 REF 1.264 1.686 High empathy-stress 1.156 0.791 REF 1.334 High empathy-distress 0.867 0.593 0.750 REF Exercise (0 = No, 1 = Yes) Low empathy-stable REF 0.293 0.794 * 0.312 * Low empathy-rising 0.774 REF 1.387 1.015 High empathy-stress 0.558 0.721 REF 0.732 High empathy-distress 0.762 0.985 1.367 REF Regular menstrual cycle (0 = No, 1 = Yes) Low empathy-stable REF 0.817 0.900 0.806 Low empathy-rising 1.224 REF 1.101 0.987 High empathy-stress 1.112 0.908 REF 0.896 High empathy-distress 1.240 1.013 1.116 REF peer relationship (0 = poor, 1 = yes) Low empathy-stable REF 1.001 1.158 2.012 Low empathy-rising 0.999 REF 0.987 2.010 * High empathy-stress 0.863 0.864 REF 1.736 High empathy-distress 0.497 0.497 0.576 REF Empathetic literature reading (0 = No, 1 = Yes) Low empathy-stable REF 0.960 0.976 1.109 Low empathy-rising 1.041 REF 1.016 1.154 High empathy-stress 1.025 * 0.984 REF 1.136 High empathy-distress 0.902 0.866 0.880 REF WLEIS Low empathy-stable REF 1.016 0.991 1.007 Low empathy-rising 1.023 ** REF 1.017 1.009 High empathy-stress 1.008 * 0.983 REF 0.993 High empathy-distress 1.016 0.991 1.007 REF Open in a new tab REF is the reference group. The reference category of the dependent variable is the subjects who retain the original group. Listed as a potential state of T1. WLEIS = Wong and Law Emotional Intelligence Scale. * P < .05. ** P < .01. 4. Discussion This study utilized LTA to examine the latent states of clinical empathy among nursing students during their internship and the transformation patterns across time. Additionally, a systematic investigation was conducted into the effects of various factors, including gender, exercise habits, regularity of menstrual cycles, peer relationships, reading EI literature, and EI, on the transition model of empathy ability. The findings indicate that the clinical empathy of trainee nurses can be categorized into 4 distinct latent classes: low empathy-stable profile, low empathy-rising profile, high empathy-stress profile, and high empathy-distress profile. These results differ from the 3-class model of empathy proposed by Xie et al [ 17 ] for undergraduate nursing students (high opinion selection-low personal distress group, low empathy ability group, and high empathy ability group). This discrepancy may arise from differences in research design and the unique stress characteristics inherent to clinical settings. Notably, LTA analysis uncovered the dynamic evolution of empathy during the clinical internship process. Under the high empathy dimension, influenced by internship-related stress, 2 subprofiles emerged: the high empathy-stress profile (characterized by short-term improvements in empathy accompanied by emotional burden) and the high empathy-distress profile (psychological exhaustion resulting from prolonged maintenance of high empathy levels). Importantly, nursing students in all 4 latent profile exhibited the highest scores in the opinion selection dimension across both measurement points, suggesting exceptional emotional perception abilities in clinical practice. As a critical component of empathy, opinion selection reflects an individual’s inclination to comprehend others’ psychological states actively. [ 18 ] Clinical practice is a pivotal stage for nursing students transitioning into professional roles, where professional identity development drives them to deepen their understanding of patient needs. This role transformation may serve as a key driver for the manifestation of enhanced opinion selection ability. [ 19 ] The analysis of latent transformation results indicates that the low empathetic-stable profile has the lowest probability of maintaining its original potential category (7%). In contrast, the high empathetic-distressful profile has the highest probability of maintaining its original potential category (44%). The low empathetic-stable profile is more likely to transform into the high empathetic-stress profile (43%) and the high empathetic-distressful profile (36%). These findings suggest clinical empathy among nursing students during their internship exhibits significant temporal dynamics. Unlike classroom simulations or theoretical learning, clinical practice demands that nursing students confront patients’ distress, anxiety, and even death directly over extended periods. This role transition and establishing professional identity contribute to fluctuations in nursing students’ empathy abilities. [ 20 ] Dynamic systems theory also underscores that empathy is not a fixed trait but fluctuates based on environmental interactions. [ 21 ] This study reveals an “upward trend” in the clinical empathy of trainee nurses during their internship period. High empathy significantly enhances patient satisfaction, improves treatment adherence, and reduces medical disputes. [ 22 , 23 ] However, Marshman et al argue that excessive empathy among healthcare professionals can lead to self-depletion and emotional suffering, [ 24 ] consistent with this study’s classification results. Individuals with excessive empathy are prone to somatic reactions such as anxiety, sleep disturbances, and physical distress. [ 25 ] An animal experiment demonstrated [ 26 ] that when monkeys observe similar distress in others, their mirror neurons activate distress-related brain regions (e.g., the insula and anterior cingulate cortex), simulating the emotional experience of others and causing involuntary frowning or tearing. Wang et al found that nursing students’ empathy abilities and job burnout are influenced by factors such as professional identity and clinical environmental stress. Long-term emotional involvement may diminish work enthusiasm and potentially lead to a desire to distance oneself from the medical field. [ 27 ] Therefore, this study posits that empathy is a double-edged sword in the medical profession. Moderate empathy enhances the quality of care, while prolonged excessive empathy may result in psychological exhaustion, occupational risks, and physical symptoms. Schools and hospitals should guide nursing students to manage emotional boundaries scientifically, strengthen support systems, and ensure sustainable career development while preserving humanistic care. Multivariate logistic regression analysis revealed that female nursing students who exercised infrequently exhibited high levels of peer support and had not read literature on empathy. This demonstrated a significant tendency toward subprofiles by high empathy traits. This finding aligns with most psychologists’ research, indicating that women score significantly higher than men in both emotional empathy (feeling others’ emotions) and cognitive empathy (understanding others’ perspectives). [ 2 , 19 ] Gupta et al further highlighted that moderate exercise can regulate emotions and alleviate stress, indirectly influencing empathy. Conversely, female nursing students who do not exercise regularly may lack the emotional regulation and stress relief benefits of physical activity. [ 28 ] Additionally, the study found that nursing students with high peer support were more susceptible to negative effects associated with heightened empathy abilities. highly empathetic nursing students might be more prone to perceiving and caring for patients’ negative emotions, leading to potential emotional exhaustion. [ 29 ] To mitigate emotional depletion among highly empathetic-stress profile and highly empathetic-distress profile nursing students, interventions such as emotional teaching courses, [ 30 ] mindfulness training, [ 31 ] and self-care training [ 32 ] could be encouraged. EI is a protective factor in the low empathetic-stable and the low empathetic-rising profiles. Some highly empathetic Chinese nursing interns may experience excessive emotional empathy, insufficient cognitive empathy, and become overly immersed in others’ narratives. For such students, participation in emotional management training programs can be encouraged to develop their capacity for emotional regulation and enhance their EI. Ma Z et al [ 33 ] demonstrated that simulating nurse–patient interaction scenarios through virtual reality and training selective empathy enables individuals to distinguish between their emotions and those of others, thereby avoiding over-involvement or ineffective assistance. Additionally, required nursing students to document “the emotions of others I perceived” and “my own needs understood” daily, fostering self-awareness through comparative reflection and preventing excessive self-sacrifice. A daily 10-minute mindfulness practice can also assist nursing students in efficiently disengaging from emotional residues following empathetic engagement. [ 34 ] Furthermore, maintaining a healthy lifestyle, encompassing regular routines, a balanced diet, and moderate exercise, significantly contributes to nursing students’ emotional and intellectual development. [ 35 ] Consequently, enhancing EI facilitates improved communication with patients and healthy interpersonal relationships, promotes a more precise self-understanding, strengthens self-awareness and self-regulation abilities, and enables nursing students to better adapt to future professional environments. 5. Conclusion and limitations The primary innovation of this study lies in its comprehensive and dynamic exploration of the latent profile and transformation characteristics of clinical empathy among nursing interns. Specifically, it investigates how demographic factors and EI influence the transformation of these profile. Nevertheless, certain limitations and deficiencies in this research warrant further investigation in future studies. First, this study only examined a limited set of variables affecting clinical empathy, while many other significant influencing factors were not included. Second, although a longitudinal design was employed to explore the temporal development of clinical empathy subprofiles among nursing interns, future studies could benefit from incorporating samples with diverse professional and cultural backgrounds to examine the similarities and differences in the development and changes of empathy across various contexts. Acknowledgments We thank the nursing students who participated the study and completed questionnaire forms. Author contributions Data curation: Xin Liu, Li Zhou. Writing – original draft: Mengmeng Hu. Writing – review & editing: Hongjuan Chang, Chun Zhang. Abbreviations: BIC Bayesian information criterion EI emotional intelligence IRI Interpersonal Reactivity Index LTA latent transition analysis OR odds ratio 2023 Annual Open Fund Project of the Key Laboratory for Occupational Hazard Identification and Control in Hubei Province (Number: OHIC2023K02) and 2023 Annual Project Funded by the Education Department of Hubei Province for Scientific Research (Number: B2023002). The authors have no conflicts of interest to disclose. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request. How to cite this article: Hu M, Liu X, Zhou L, Chang H, Zhang C. Trajectory changes of clinical empathy ability of Chinese nursing interns: A longitudinal study using latent transition analysis. Medicine 2026;105:22(e49054). Contributor Information Mengmeng Hu, Email: 2927788144@qq.com. Xin Liu, Email: 512736027@139.com. Li Zhou, Email: humm2026@126.com. Hongjuan Chang, Email: changhj0812@126.com. References [1]. Nakao M, Komaki G, Yoshiuchi K, Deter H-C, Fukudo S. Biopsychosocial medicine research trends: connecting clinical medicine, psychology, and public health. BioPsychoSocial Med. 2020;14:30. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [2]. Lee SE, Seo J-K, Squires A. Voice, silence, perceived impact, psychological safety, and burnout among nurses: a structural equation modeling analysis. Int J Nurs Stud. 2024;151:1046669. [ DOI ] [ PubMed ] [ Google Scholar ] [3]. Wang J, Xu X, Sun J, et al. A study of latent profile analysis of empathic competence and factors influencing it in nursing interns: a multicenter cross-sectional study. Front Public Health. 2024;12:1434089. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [4]. Coffey A, Saab MM, Landers M, et al. The impact of compassionate care education on nurses: a mixed-method systematic review. J Adv Nurs. 2019;75:2340–51. [ DOI ] [ PubMed ] [ Google Scholar ] [5]. Cheng L, Yang J, Li M, Wang W. Mediating effect of coping style between empathy and burnout among Chinese nurses working in medical and surgical wards. Nursing Open. 2020;7:1936–44. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [6]. Cunico L, Sartori R, Marognolli O, Meneghini AM. Developing empathy in nursing students: a cohort longitudinal study. J Clin Nurs. 2012;21:2016–25. [ DOI ] [ PubMed ] [ Google Scholar ] [7]. Brackett MA, Rivers SE, Shiffman S, Lerner N, Salovey P. Relating emotional abilities to social functioning: a comparison of self-report and performance measures of emotional intelligence. J Pers Soc Psychol. 2006;91:780–95. [ DOI ] [ PubMed ] [ Google Scholar ] [8]. Ksiksou J, Maskour L, Alaoui S. The relationship between perceived stress and emotional intelligence in Moroccan nursing students. J Educ Health Promotion. 2023;12:238. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [9]. Kikanloo AAI, Jalali K, Asadi Z, et al. Emotional intelligence skills: is nurses’ stress and professional competence related to their emotional intelligence training? A quasi experimental study. J Adv Med Educ Professionalism. 2019;7:138–43. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [10]. Chen S, Liu W, Li Y, et al. Transition shock among nursing interns: roles of achievement motivation and clinical practice environment-a cross-sectional study. Nurs Open. 2025;12:e70297. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [11]. Li H, Fong TCT, Hsu YC, et al. Change in psychological distress and associated factors among Hong Kong young adults in post-COVID-19 era: a latent transition analysis. Soc Psychiatry Psychiatr Epidemiol. 2025;60:2823–33. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [12]. Davis MH. Empathy: a social psychological approach. Social psychology series;1994. [ Google Scholar ] [13]. Zheng F, Dong Y, Wang K, et al. Reliability and validity of the Chinese version of Interpersonal Reactivity Index (IRI-C). Chin J Clin Psychol. 2010;18:155–7. [ Google Scholar ] [14]. Wong C-S, Law K. The effects of leader and follower emotional intelligence on performance and attitude: an exploratory study. Leadersh Q. 2002;13:243–74. [ Google Scholar ] [15]. Nguefack HLN, Pagé MG, Katz J, et al. Trajectory modelling techniques useful to epidemiological research: a comparative narrative review of approaches. Clin Epidemiol. 2020;12:1205–22. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [16]. Langenberg B, Helm JL, Mayer A. Repeated Measures ANOVA with Latent variables to analyze interindividual differences in contrasts. Multivariate Behav Res. 2022;57:2–19. [ DOI ] [ PubMed ] [ Google Scholar ] [17]. Xie J, Zhang JH, Liu R, et al. Influencing factors analysis and intervention strategies of empathy ability of nursing undergraduates based on latent class analysis. Nurs Res. 2022;36:3804–9. [ Google Scholar ] [18]. Deng X, Chen S, Li X, et al. Gender differences in empathy, emotional intelligence and problem-solving ability among nursing students: a cross-sectional study. Nurse Educ Today. 2023;120:105649. [ DOI ] [ PubMed ] [ Google Scholar ] [19]. Heggestad AKT, Konow-Lund A-S, Christiansen B, Nortvedt P. A vulnerable journey towards professional empathy and moral courage. Nurs Ethics. 2022;29:927–37. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [20]. Jia-Ru J, Yan-Xue Z, Wen-Nv H. Empathy ability of nursing students: a systematic review and meta-analysis. Medicine (Baltim). 2022;101:e30017. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [21]. Klocek A, Premus J, Řiháček T. Applying dynamic systems theory and complexity theory methods in psychotherapy research: a systematic literature review. Psychother Res. 2024;34:828–44. [ DOI ] [ PubMed ] [ Google Scholar ] [22]. Lind M, Koch MK, Bluck S. Nursing students’ empathic communication: role in recognizing and treating chronic pain patients. Patient Educ Couns. 2024;123:108236. [ DOI ] [ PubMed ] [ Google Scholar ] [23]. Wang Q, Wang L, Shi M, et al. Empathy, burnout, life satisfaction, correlations and associated socio-demographic factors among Chinese undergraduate medical students: an exploratory cross-sectional study. BMC Med Educ. 2019;19:3804– 9. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [24]. Marshman C, Hansen A, Munro I. Compassion fatigue in mental health nurses: a systematic review. J Psychiatr Ment Health Nurs. 2022;29:529–43. [ DOI ] [ PubMed ] [ Google Scholar ] [25]. Xie W, Chen L, Feng F, et al. The prevalence of compassion satisfaction and compassion fatigue among nurses: a systematic review and meta-analysis. Int J Nurs Stud. 2021;120:103973. [ DOI ] [ PubMed ] [ Google Scholar ] [26]. Pascolo PB, Bucci A. Mirror neurons in the macaque monkey: a critical review. Crit Rev Biomed Eng. 2019;47:507–13. [ DOI ] [ PubMed ] [ Google Scholar ] [27]. Xia HJ, Li HF, et al. Current status and influencing factors of empathy fatigue among intern nursing undergraduates. Chin High Med Educ. 2023;19:19–21. [ Google Scholar ] [28]. Gupta S, Bharatha A, Cohall D, Rahman S, Haque M, Azim Majumder MA. Aerobic exercise and endocannabinoids: a narrative review of stress regulation and brain reward systems. Cureus. 2024;16:e55468. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [29]. Berduzco-Torres N, Medina P, San-Martín M, Delgado Bolton RC, Vivanco L. Non-academic factors influencing the development of empathy in undergraduate nursing students: a cross-sectional study. BMC Nurs. 2021;20:245. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] [30]. Napolitano F, Calzolari M, Pietro SD, et al. Pedagogical strategies to improve emotional competencies in nursing students: a systematic review. Nurse Educ Today. 2024;142:106337. [ DOI ] [ PubMed ] [ Google Scholar ] [31]. Yang C, Zhu Y-L, Xia B-Y, Li Y-W, Zhang J. The effect of structured empathy education on empathy competency of undergraduate nursing interns: a quasi-experimental study. Nurse Educ Today. 2020;85:104296. [ DOI ] [ PubMed ] [ Google Scholar ] [32]. Soner G, Avci IA. The effect of a disability-friendly nursing education program based on story theory on nursing students’ empathic tendencies and perceptions of competence in providing care: a mixed methods study. Nurse Educ Today. 2025;151:107085. [ DOI ] [ PubMed ] [ Google Scholar ] [33]. Ma Z, Huang K-T, Yao L. Feasibility of a computer role-playing game to promote empathy in nursing students: the role of immersiveness and perspective. Cyberpsychol Behav Soc Netw. 2021;24:750–5. [ DOI ] [ PubMed ] [ Google Scholar ] [34]. Aloufi MA, Jarden RC, Gerdtz MF, et al. Reducing stress, anxiety and depression among undergraduate nursing students: a systematic review. Nurse Educ Today. 2021;102:104905. [ DOI ] [ PubMed ] [ Google Scholar ] [35]. Arrogante O, Ortuño-Soriano I, Fernandes-Ribeiro AS, Raurell-Torredà M, Jiménez-Rodríguez D, Zaragoza-García I. The effects of high-fidelity simulation training on emotional intelligence in undergraduate nursing students: a pre-post intervention study. Teaching Learning Nurs. 2025;20:e829–35. [ Google Scholar ] Articles from Medicine are provided here courtesy of Wolters Kluwer Health ACTIONS View on publisher site PDF (806.5 KB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• A scoping review of factors associated with pain empathy among nursing professionals and students.#1053943
• Adult Attachment Predicts Parental Empathy: Evidence Across Four Samples of Parents of Children from Preschool Through Middle Childhood#453241
• Empathy, Timeliness, and Virtuous Hearing#108898
• Continuity Empathy - A Structural Framework for Reconstructing the Continuity Beneath Emotional, Relational, and Cognitive States#694674
• Use and effectiveness of the arts for enhancing healthcare students' empathy skills: A mixed methods systematic review#393453