Media factors in medical sociology have long been recognized as significant in influencing health management and medical outcomes. For example, media and media exposure have been shown to play a key role in heart disease prevention, vaccination, community care, and other health interventions [1,2]. The COVID-19 pandemic and long COVID in the digital age have caused unprecedented social disruptions due to the widespread use of social media and the circulation of mixed information. Therefore, it is essential to investigate how these new “media factors” impact the stability and development of society.

Furthermore, the prevalence of COVID-19 is accompanied by an infodemic [3]. The disaster stories about COVID-19 on social media serve not only as a significant force to unite the public in combating the pandemic but also as hidden triggers for the widespread resignation of health care workers. According to the World Health Organization (WHO), there will be a shortfall of 10 million health workers by 2030 [4]. Medical students are a vital reserve force for health care workers and play an essential role in public health events. For instance, during the COVID-19 pandemic, medical students’ involvement in patient care and health consultation complemented the work of medical professionals [5,6]. Their professional identity is critical to their career perception and choices [7], which may further impact the stability of the medical workforce.

Professional identity refers to the self-conceptualization of one’s work role [8] and is central to professional loyalty. In contrast to experienced medical professionals, medical students are young, vulnerable, and digital natives, with their professional identities still in formation and transformation [9]. Prior studies have typically applied descriptive analyses to explain the development of professional identity [10,11]. Although current studies are beginning to use cross-sectional methods to explore predictors of professional identity, they primarily discuss individual and educational environmental factors, such as clinical practice [12], experiences within the educational system [13], and working conditions [14]. Recently, there has been growing evidence suggesting that online media play an important role in influencing students’ knowledge and attitudes regarding the medical profession [15]; however, the theoretical mechanisms behind this influence remain unclear. Given that the COVID-19 pandemic has highlighted the impact of media on people’s daily lives [16], this research speculates that exposure to online media during the pandemic could also influence medical students’ professional identity. As online news media and social media are the most commonly used types of media, this study proposes the following research question: How does concurrent media exposure affect medical students’ professional identity?

In public health events such as the COVID-19 pandemic, the media play a crucial role in information distribution, pandemic warnings, gauging public emotions, facilitating social mobilization, and providing emotional support [17]. Specifically, online media not only deliver medical-related information [18] but also feature news reports that portray the professional and heroic images of health care workers [19]. Additionally, online media facilitate social interactions [20], allowing netizens to express their support for health care workers through understanding, encouragement, affirmation, sympathy, and care, among other forms [21]. As such, the social support gained through online media exposure may help medical students recognize the value of medical groups [22], develop their attitudes toward the medical profession, and understand public opinions about the field [23]. This study utilizes the Stimulus-Organism-Response (SOR) framework [24] to explain how media exposure relates to medical students’ professional identity, emphasizing the mediating role of social support.

To address the proposed research question, a survey method is used, yielding a total of 1087 valid responses. The results reveal that media exposure during the COVID-19 pandemic affects medical students’ professional identity, either directly or indirectly, through social support. Theoretically, this study provides a comprehensive understanding of the effects of media exposure on the development of medical students’ professional identity by extending the SOR framework. Practically, this study underscores the importance of media and information management in fostering professional identity, offering a viable approach to addressing the shortage of the medical workforce, promoting social unity, and reducing health disparities.

The objective of this study was to examine the effect of concurrent media exposure on medical students’ professional identity, with a specific focus on the mediating role of social support. In particular, this study investigated how concurrent media exposure (ie, online news media and social media) influenced social support (ie, informational support and emotional support), which in turn affected the development of medical students’ professional identity (ie, sense of belonging and professional commitment).

To empirically test the proposed research model, a cross-sectional survey was conducted to collect data. The survey design adhered to the CHERRIES (Checklist for Reporting Results of Internet E-Surveys) checklist [70] (see Multimedia Appendix 1 for the completed CHERRIES). A convenience sampling method was used, inviting medical students who were readily available to participate. Participants were recruited with the assistance of faculty members at a medical college. Specifically, the study team shared information about the survey in several medical courses and invited students to participate voluntarily. They were informed that their anonymity would be ensured to encourage honest responses and reduce social desirability bias [71]. The survey was conducted in May 2022.

Overall, the main body of the questionnaire consisted of 4 sections: (1) media exposure; (2) social support; (3) professional identity; and (4) demographic characteristics. Additionally, an introductory page was included at the beginning of the questionnaire to inform participants about the research purpose, target population, length of the survey, and policies regarding data security and storage. Furthermore, all participants were required to provide informed consent.

The questionnaire and methodology for this study were approved by the Ethics Committee of Tongji Medical College of Huazhong University of Science and Technology (approval number 2022S009). Informed consent was obtained from all participants. Measures were implemented to protect the privacy of all participants and ensure the confidentiality of the data. The questionnaire was anonymous, and the data file was stored on a secure drive accessible only to research team members who signed a confidentiality agreement. Each participant received a compensation of 10 RMB (approximately US $1.5) for their valid response.

This study used several strategies to develop the measurement instruments. First, to ensure the content validity of the instrument, measures for the variables were primarily adapted from well-validated scales and modified to fit the Chinese context and the characteristics of medical students. The back-translation method was used to adapt the original English scale to Chinese and ensure consistency between the 2 language versions. Specifically, the questions were first framed in English, then translated into Chinese, and finally back-translated into English to confirm consistency.

Second, we conducted several rounds of pretests to refine the instruments. Three professors and a group of graduate students from a medical school were invited to participate in these pretests. They completed the questionnaires and took part in follow-up interviews to evaluate whether the instruments effectively measured the constructs under investigation. Participants were also asked to provide feedback on the wording of the questions, clarity of expression, and question sequence. Based on their feedback, we enhanced the face validity and content validity of the instruments.

Third, we conducted a pilot test to assess the reliability and validity of the instruments. The results of the data analysis (shown in Multimedia Appendix 2; also see [72-75]) indicated that all measurements demonstrated ideal reliability and validity. Additionally, the measurement model analysis in the formal test further confirmed the reliability and validity of the instruments. The measures for the constructs are provided as follows.

Medical students from 4 majors (clinical medicine, basic medicine, nursing, and public health) at a medical college in China were recruited to complete a set of paper questionnaires. We selected these 4 majors due to their critical roles in the frontlines of the COVID-19 pandemic. Specifically, doctors, nurses, and epidemiologists from clinical medicine, nursing, and public health actively participated in the prevention and control of the pandemic. Students majoring in basic medicine were also included, as they play a vital role in combating the new coronavirus. By choosing a diverse range of majors, we aimed to enhance the generalizability of this study.

In the formal data collection process, a screening question was included to allow only medical students with online media usage experience to participate in the survey. Notably, participants from the pilot test were not permitted to complete the questionnaire. The survey included questions regarding demographic information, professional identity, and factors related to media exposure and social support. Additionally, we incorporated attention-trap questions to ensure the quality of the responses. A total of 1200 questionnaires were collected. After removing 113 invalid responses (eg, those that failed the attention test, contained missing values, or had excessively duplicated answers), we finalized a sample of 1087 valid responses, with students from the 4 majors approximately evenly distributed.

This study used structural equation modeling to test the research framework. We selected SmartPLS 4.0 (SmartPLS GmbH), based on partial least squares, as it allows for simultaneous testing of both the measurement model and the structural model [81]. In the current model, online news media exposure and social media exposure were incorporated as independent variables; informational support and emotional support served as mediators; and professional identity (ie, belonging and professional commitment) variables were included as outcomes. Following the approach suggested by Anderson and Gerbing [82], we first analyzed the measurement model to assess reliability and validity, and subsequently conducted structural model analysis to evaluate our research model. Furthermore, a mediation analysis was conducted to uncover the underlying mechanisms of media exposure’s influence. To account for potential confounding effects, gender, age, education, and internship experience were included as control variables in the structural equation model.

As mentioned above, a final sample of 1087 valid responses was obtained. The demographic information of this sample is summarized in Table 1. More than half of the students were female and pursuing an undergraduate degree. Additionally, most of the students were in their 20s or younger and reported having internship experiences during their studies.

This study assesses the convergent and discriminant validity to evaluate the measurement model. The descriptive statistics of the variables in our research model and the results are presented in Table 2. Notably, Cronbach α for all constructs exceeded 0.7. Additionally, the composite reliability values for all constructs were greater than 0.8, and all average variance extracted (AVE) values were above 0.5, meeting the recommended threshold values of 0.7 and 0.5, respectively [72,73]. In addition, all item loadings, except for the one related to professional commitment, exceeded 0.7. As this factor loading was close to 0.7, we retained it to consider content validity. Thus, the model demonstrates acceptable reliability and convergent validity. We further assessed discriminant validity by comparing the interconstruct correlation coefficients with the square roots of AVE [72]. The results in Table 3 indicate that the square roots of AVE exceed all interconstruct correlation coefficients, confirming that discriminant validity is not an issue.

This study also examined multicollinearity. As a rule, multicollinearity is considered high if the variance inflation factor of a variable exceeds 10 [72]. In this study, the variance inflation factor values for all variables were below 3, indicating the absence of significant multicollinearity. Additionally, because all data were collected from a cross-sectional survey, common method bias (CMB) was assessed. First, we conducted the Harman single-factor test using SPSS Statistics 26.0 (IBM Corp.) and found that the most important factor explained only 23.33% of the total variance, which is lower than the reference value of 50% [74]. Second, CMB may lead to high correlations between constructs. Table 3 shows that the highest correlation between constructs was 0.675, which is below the recommended threshold of 0.90 [75]. Therefore, CMB was not a significant threat in this study.

The results of the structural model test are shown in Figure 2. Consistent with our expectations, online news media exposure was positively associated with informational support (β=.163; P<.001) and emotional support (β=.084; P=.007). Additionally, social media exposure was positively associated with informational support (β=.122; P<.001) and emotional support (β=.235; P<.001). Therefore, H1a, H1b, H2a, and H2b are all supported.

Thereafter, this study analyzed the impact of social support on medical students’ professional identity. The results showed that informational support (β=.228; P<.001) and emotional support (β=.344; P<.001) were positively associated with belonging. Meanwhile, we found that informational support (β=.245; P<.001) and emotional support (β=.412; P<.001) were positively associated with medical students’ professional commitment. Overall, the model explained 28.1% of the variance in belonging and 35.6% of the variance in professional commitment. Therefore, H3a, H3b, H4a, and H4b are all supported.

This study also tested the impact of demographic variables on medical students’ professional identity. The results showed that the impacts of gender on belonging (β=.064; P=.22) and professional commitment (β=–.008; P=.87) were insignificant. Meanwhile, the impacts of age on belonging (β=–.074; P=.08) and professional commitment (β=–.053; P=.17) were also insignificant. Furthermore, the impacts of education on belonging (β=–.061; P=.15) and professional commitment (β=–.022; P=.54) were likewise insignificant. Notably, internship experience had positive effects on both belonging (β=.114; P<.001) and professional commitment (β=.091; P<.001).

The mediation effects of social support were further assessed following Hair et al [83] and Zhao et al [84]. Table 4 shows the results of our mediation analysis. The results indicated that specific indirect effects for both independent variables were significant, confirming the mediation effect of social support. Additionally, online news media exposure had significant direct effects on both belonging (β=.076; P=.01) and professional commitment (β=.059; P=.03). Therefore, informational support and emotional support partially mediated the relationship between online news media exposure and medical students’ professional identity [81]. Thus, H5a, H5b, H6a, and H6b were supported. In addition, as the direct effects of social media on belonging (β=.060; P=.06) and professional commitment (β=.045; P=.13) were both insignificant, social support fully mediated the relationship between social media exposure and medical students’ professional identity [81]. As such, H7a, H7b, H8a, and H8b were all supported.

In light of the far-reaching crisis brought by the pandemic, this study aims to uncover how media factors can help mitigate social risks within the field of medical sociology. Utilizing the SOR framework, we propose a research model to examine the impacts of media exposure on medical students’ professional identity. Our findings reveal several intriguing insights.

First and foremost, our findings indicate that both online news media exposure and social media exposure are positively related to perceived emotional and informational support. This suggests that medical students can derive both types of support from these media sources. Notably, the path coefficient between online news media and informational support was relatively higher than that for emotional support, whereas the opposite was true for social media exposure. The results can be explained by the inherent characteristics of the 2 types of media. Online news media are generally perceived as trustworthy due to their strict gatekeeping mechanisms [2], making them vital and reliable sources for informational support. By contrast, social media serves as an important platform for facilitating online interactions [46], thereby proving effective in providing emotional support.

Second, emotional support and informational support are positively associated with belonging and professional commitment, indicating that social support is a significant predictor of medical students’ professional identity. Notably, the path coefficients for emotional support on both belonging and professional commitment are higher than those for informational support. The results suggest that emotional support may have a greater impact on fostering medical students’ professional identity than informational support. This finding aligns with Wang et al [61], who emphasized the importance of emotional support in enhancing interpersonal relationships within online health support groups. Additionally, this can be explained by the nature of professional identity itself; both belonging and professional commitment relate to individuals’ feelings toward their professional group and the profession [37,38], making them more likely to be driven by emotional factors.

Third, the impact of media exposure on professional identity was mediated by social support. Specifically, social support partially mediated the effect of online news media exposure and fully mediated the effect of social media exposure. These results indicate that social media primarily relies on the provision of social support to influence professional identity, whereas online news media exert their effects on professional identity both directly and indirectly through social support. One possible explanation for this finding lies in the characteristics of different media. During the pandemic, individuals primarily relied on social media to transmit and obtain social support. By contrast, by reporting news about medical workers, online news media can directly assist medical students in developing their understanding of professional identity [19]. These findings further clarify that social media and online news media might play different roles in shaping medical students’ professional identity.

Fourth, the analysis of control variables shows that internship experience is positively associated with medical students’ belonging and professional commitment. For medical students, the medical profession places great emphasis on practicality, making internships an essential avenue for gaining knowledge about the field. On the one hand, internships help students develop their self-efficacy by providing hands-on experience and exposure to real-world medical environments [85]; on the other hand, practical experience can enhance medical students’ awareness of the profession [86]. Consequently, the training received during internships may better prepare students to become medical workers, which, in turn, inspires them to cultivate a sense of belonging and professional commitment to the field.

The internet has emerged as the predominant form of media globally, with social media serving as a crucial platform during the COVID-19 pandemic [3]. Given the media environment’s significant influence on shaping the cognition and behaviors of the younger generation [87], as well as the importance of medical students’ professional identity for the stability of the medical community, this study explores the impact of media exposure on medical students’ professional identity and seeks to uncover the mechanisms underlying this relationship. By comparing with prior work, this study offers several theoretical implications. First, it enriches the existing literature on professional identity in medical education. Previous studies typically focused on individual and educational environmental factors in the development of professional identity [12-14], often neglecting the role of the media environment. By integrating media exposure research with a social psychology perspective, this study empirically validates a new set of relationships and illustrates how media exposure influences medical students’ perceived social support and professional identity. Unlike prior work, this study examines 2 types of media exposure simultaneously, developing a nuanced understanding of the effects of online news media and social media on medical education. The findings emphasize the significance of media in fostering medical students’ professional identity, offering a new perspective for understanding the development of professional identity.

Second, this study extends the application of the SOR framework to explore the psychological processes underlying the development of professional identity, complementing the literature that examines the direct impacts of social factors on professional identity. It demonstrates that media exposure influences professional identity through both informational support and emotional support. Previous research has tended to focus on either cognitive or affective aspects of the organism component in the SOR framework [39]. This study further differentiates these 2 aspects by discussing the distinct impacts of informational support and emotional support on professional identity. Additionally, by illustrating how social support can be generated from different types of media exposure, this study contributes to the literature on the development of social support within various media environments.

This study also has practical implications for media platforms and medical educators. Above all, the findings confirm the importance of the media environment in fostering professional identity. These results are particularly relevant in the context of the Chinese media landscape. Since the outbreak of the COVID-19 pandemic, a substantial amount of information sharing praise and rewards for medical workers has circulated on Chinese media platforms. Exposure to such information may help students recognize the public’s appreciation for their profession, thereby enhancing their perception of social support as well as their professional identity. Consequently, the proper use and management of media during public health crises can facilitate the development of medical students’ professional identity. Considering the differences between online news media and social media, management departments and medical educators can leverage online news media to post reliable medical information, providing informational support, while encouraging favorable social interactions through social media to offer emotional support.

Additionally, our findings emphasize the importance of social support in enhancing professional identity. Sufficient social support should be provided to improve students’ well-being during their learning experiences [88]. By alleviating mental health issues such as depression, stress, and anxiety, students’ professional identity can be further promoted [89]. Furthermore, it is beneficial to present positive images of medical workers in online media so that medical students can develop a positive self-concept about the medical profession [77]. When reporting on medical workers, journalists should emphasize the value of medical professionalism rather than simply portraying them as angels or heroes [90,91]. In addition to focusing on providing social support through online media, medical educators and management departments can implement other measures to support students in coping with the challenges they encounter and improving their well-being, thereby reinforcing their professional identity.

Moreover, the significant effects of internships on belonging and professional commitment offer valuable implications for medical educators. Specifically, medical educators should provide opportunities for medical students to engage in practical learning. However, given that many medical students often have limited experience interacting with patients, leveraging media resources to provide professional knowledge can serve as an important complement to their internships. The practical experience gained from internships and the knowledge acquired through online media can be combined to motivate medical students to feel a sense of belonging and commitment to their profession.

There are several limitations that warrant future research. First, our study is designed based on the media environment in China. While the data illustrate the impacts of media exposure on the professional identity of Chinese medical students, future research could recruit respondents from other countries to enhance the generalizability of the findings. Second, this study focuses on online news media and social media without considering other media forms, such as short video platforms. Future research can consider the effects of other types of media. Third, this study collected self-reported data through a cross-sectional survey. While the relationships between variables can be tested, this approach limits the ability to validate causal relationships. Future studies could adopt longitudinal designs to collect data and examine the effects of media exposure. In addition, self-reported data may be less reliable in predicting actual behavior. Future research should consider using objective measures to assess professional identity. Fourth, while this study focuses on media exposure to uncover its impacts on the development of medical students’ professional identity, it is important to acknowledge that professional identity can be influenced by various factors. Future studies should explore additional factors in different contexts, particularly those that may hinder the development of professional identity.

Whether faced with a plague or war, health workers have always been on the front lines of combating social risks. Their professional identity serves as both a foundation for their beliefs and a source of strength. As members of the young generation in the age of social media, medical students’ cognitions and behaviors are easily shaped by the media environment [87]. It is thus urgent to conduct in-depth research on the impacts of media exposure on medical students’ professional identity, especially as the global wave of medical staff resignations intensifies. This study extends the SOR framework to examine the effects of media exposure related to long COVID on medical students’ professional identity, aiming to provide insights into the critical issue of medical workforce shortages. Based on a survey of 1087 medical students, this study reveals that exposure to online news media and social media can enhance medical students’ sense of belonging and professional commitment through emotional and informational support. It clarifies the role of media affordances in medical education, underscores the importance of media management and social support in shaping professional identity, and expands the understanding of the media’s role in public health events. In the new media environment, these factors may become crucial components and governance tools in medical sociology, influencing health outcomes and serving as a viable approach to achieving the One Health goal in the long COVID era.