Social media has become a crucial bridge for hospitals to interact and communicate with the public in addressing rapid digital transformation issues. Disseminated health knowledge, shared medical information, and showcased professional services can also be provided [1-3]. Leveraging the extensive reach and immediacy of the internet, hospitals can use social media to establish close connections with patients, respond promptly to their needs and feedback, enhance their brand image, and build public trust and satisfaction [4-6]. Hence, social media influence is a means for expanding visibility, reputation, and critical support for achieving long-term sustainable development for hospitals [7,8].

Existing studies suggest that social media influence primarily encompasses 4 dimensions: audience reach, audience approval, audience interaction, and dissemination power [9-11]. Audience reach refers to the breadth and spread of information, reflecting the penetration and popularity of hospital content on social media platforms [12]. Audience approval indicates the audience’s acceptance of and trust in hospital-published information, highlighting the hospitals’ professional authority and brand image [13,14]. Audience interaction measures the quantity of user-hospital engagement on the platform, indicating the closeness between hospitals and the public [15]. Dissemination power focuses on content sharing and diffusion capability, assessing the impact and reach of secondary transmissions [16].

A hospital’s comprehensive capabilities are directly linked to its social media influence [17,18]. Large hospitals with stronger medical resources, professional talent, and broader patient bases enjoy inherent exposure and authority on social media [19]. These hospitals leverage their reputation and brand advantages to attract public attention and gain higher trust and recognition when sharing medical information [20].

However, a hospital’s comprehensive capabilities require long-term accumulation, which is challenging to enhance significantly in the short term. Promotional strategies are another key approach for enhancing hospitals’ social media influence [21,22]. Promotional strategies mainly focus on 3 aspects: activity, interactivity, and entertainment value. Activity emphasizes the frequency and timeliness of content updates, maintaining platform engagement through regular updates of health knowledge and medical information [23-25]. Interactivity emphasizes the frequency and quality of engagement between hospitals and other accounts, strengthening social networks through active and meaningful responses [26,27]. Entertainment value involves creating engaging and visually attractive content such as multimedia posts and short videos to enhance the appeal and dissemination of information [28,29]. With carefully curated content, effective interaction management, and collaborations with medical experts or influencers, hospitals can significantly boost their social media influence in the short term.

Although the aforementioned studies demonstrate that promotional strategies can effectively improve a hospital’s social media performance, the specific mechanisms impacting the 4 dimensions of social media influence remain underexplored. In addition, the effectiveness of promotional strategies may vary across different types of hospitals, and tailored strategies are crucial for optimizing resource allocation and enhancing promotional efficacy. Further research is needed to clarify how promotional strategies affect various dimensions of social media influence and identify heterogeneity across hospital types to provide actionable strategic recommendations.

In this study, we formulated hypotheses based on the uses and gratifications theory [30,31], dialogic communication theory [32,33], elaboration likelihood model [34,35], resource-based view [36,37], and institutional theory [38]. We empirically examined how the 3 dimensions of promotional strategies (activity, interactivity, and entertainment value) influence the 4 aspects of hospital social media influence (audience reach, audience approval, audience interaction, and dissemination power). In addition, hospital characteristics were incorporated as moderating variables to further explore their effects on these relationships.

Data were extracted from 721 verified hospital Weibo accounts and merged with various hospital characteristics sourced from the DXY National Hospital Database. Weibo, one of China’s largest social media platforms with a vast user base and robust information dissemination capabilities, has become a key venue for hospital social media promotion [39-41]. Within this context, this study collected data from hospital Weibo accounts officially verified as of January 2024.

The results of this study reveal how various promotional strategies impact the dimensions of social media influence and offer evidence-based guidance for formulating targeted strategies tailored to various hospital types. These findings not only help enhance hospital influence and public trust on social media platforms but also lay a solid foundation for future brand building and sustainable development.

This study investigated how to use promotional strategies to enhance the social media influence of hospitals. Hospital social media influence was categorized into 4 dimensions: audience reach, audience approval, audience interaction, and dissemination capacity. In detail, we measured audience reach on social media using the number of followers of the hospitals’ official accounts on Weibo (wbFans variable). Audience approval was evaluated using the maximum number of likes received on a single Weibo post (wbLike variable). Audience interaction was assessed through the maximum number of comments on a Weibo post (wbComment variable), whereas dissemination power was reflected in the maximum number of reposts of a single Weibo post (wbRetweet variable).

Next, this study examined the promotional strategies used by hospitals on social media focusing on 3 key dimensions: activity, interactivity, and entertainment value. Specifically, account activity was measured using the number of posts made by the hospitals’ official accounts on Weibo (wbNum variable). Interactivity was represented by the number of accounts followed by the hospitals accounts (wbFollow variable). Entertainment value was assessed through the percentage of posts containing images (wbPicture variable) and videos (wbVideo variable) relative to the total number of posts.

Finally, recognizing the diverse characteristics of hospitals, we considered several key attributes. These included the number of approved beds (Bed variable) as an indicator of hospital size; years of operation (History variable) to reflect institutional maturity; overall ranking (Rank variable) to represent hospital reputation; and binary indicators for whether the hospital was tertiary (Tertiary variable), private (Private variable), or general (General variable). In addition, we included the hospital’s provincial location. In terms of provincial characteristics, we further considered residents’ disposable income (Income variable), population density (Density variable), and age structure (the proportion of the population aged 14-65 years, denoted as the Age variable).

This paper presents descriptive statistics for the key variables, including the sample size, mean, SD, and minimum and maximum values. In addition, we examined the geographical distribution of hospitals to explore regional patterns in their presence.

To investigate how hospitals’ promotional strategies affected their social media influence, this study used a multiple linear regression model, as specified in equation 1. In equation 1, the dependent variable y represents various dimensions of hospitals’ social media influence, including the variables wbFans, wbLike, wbComment, and wbRetweet. To assess the percentage change in y, a logarithmic transformation was applied to these variables.

The key independent variables in equation 1 capture hospitals’ social media promotional strategies, including the variables wbNum, wbFollow, wbPicture, and wbVideo. The number of posts (wbNum) and accounts followed (wbFollow) were logarithmically transformed and represented as ln(wbNum) and ln(wbFollow), respectively. The proportions of posts with images (wbPicture) and videos (wbVideo) were multiplied by 100 to reflect percentages, denoted as wbPicture*100 and wbVideo*100.

In addition, equation 1 incorporates a set of control variables reflecting hospital-specific characteristics, including Bed, History, Rank, Tertiary, Private, and General. The variable Bed was logarithmically transformed and denoted as ln(Bed). Given the regional disparities across provinces in China, equation 1 also accounts for the effect of provincial location on hospitals’ social media influence. A set of province-level dummy variables D^k, which indicates whether a hospital was located in the kth province, was included. This approach provided a simplified yet effective means of controlling for unobserved characteristics associated with a hospital’s location. Given that provinces differ in terms of economic development, population structure, and other contextual factors that may systematically influence hospitals’ social media performance, the inclusion of D^k captured the composite effects of province-level attributes and mitigated potential bias from omitted variable issues.

The effectiveness of promotional strategies in enhancing social media influence may vary across different types of hospitals. This study further explored heterogeneity at the hospital level.

First, this study examined whether hospitals’ promotional strategies had already exhibited differentiation. To address this question, equation 2 was used. In equation 2, the dependent variable y represents various aspects of hospitals’ promotional strategies, including ln(wbNum), ln(wbFollow), wbPicture*100, and wbVideo*100. The independent variables in equation 2 capture hospital characteristics, including ln(Bed), History, Rank, Tertiary, Private, General, and D^k. In addition, D^k represents dummy variables for the hospitals’ provincial location. A multiple linear regression was conducted to analyze these relationships.

Second, this study grouped hospitals according to various characteristics to examine potential heterogeneity in the effectiveness of promotional strategies on enhancing social media influence. Hospitals were categorized based on the following dimensions: (1) large versus small size (based on the median number of approved beds), (2) long versus short history (also proxied by the median number of approved beds), (3) high versus low comprehensive ranking (using the median rank as the threshold), (4) high versus low hospital grade (ie, whether the hospital was a tertiary institution), (5) public versus private ownership, (6) general versus specialized hospitals, and (7) geographical region (with the northeastern and central regions combined due to limited northeastern observations, resulting in 3 regional groups: eastern, central and northeastern, and western).

For binary classifications (eg, ownership, ranking, size, history, grade, and whether the hospital was general or specialized), a dummy variable G was introduced. Building on equation 1, the core explanatory variables—ln(wbNum), ln(wbFollow), wbPicture100, and wbVideo100—each interacted with G—(ln(wbNum)*G, ln(wbFollow)*G, wbPicture*100*G, and wbVideo_i*100*G—as shown in equation 3. The coefficients of these interaction terms capture whether the effects of promotional strategies on social media influence differed by hospital type. Similarly, for the 3-category classification (ie, geographic region), 2 dummy variables, G₁ and G₂, were constructed. The interaction terms between the core explanatory variables and G₁ and G₂—ln(wbFollow)*G₁, wbPicture*100*G₁, wbVideo*100*G₁, ln(wbNum)*G_2I, ln(wbFollow)*G₂, wbPicture*100*G₂, and wbVideo*100*G₂—were included in the regression, as specified in equation 4.

This analysis aimed to identify which factors moderated the effectiveness of promotional strategies and provide tailored policy recommendations for hospitals with different characteristics.

To demonstrate the robustness of the main results, we conducted several additional analyses. First, the baseline regression shown in equation 1 can be interpreted as a fixed-effects model at the province level. Building on this idea, we extended the specification by estimating a mixed-effects model, which incorporated observable province-level variables and a random intercept for each province. As shown in equation 5, the set of province-level dummy variables D^k was replaced with observable province-level control variables (Income, Density, and Age) as well as a province-level random effect ε_k. Second, to test the generalizability of the model, we compared empirical results across different sets of control variables. In addition, to address the multiple comparison issue arising from the inclusion of numerous predictors, we applied the Benjamini-Hochberg false discovery rate (FDR) procedure to adjust the P values and reduce the risk of false positive errors.

This study did not require new approval from the institutional ethics committee, as it did not involve any clinical interventions and was based on publicly available data.

Descriptive statistics for the variables used in this study are presented in Table 1. Furthermore, the distribution of hospitals across provinces is illustrated in Figure 1. As shown in the figure, hospital resources are unevenly distributed at the provincial level, reflecting disparities closely tied to local economic development, population size, and other regional resource conditions.

This uneven distribution across provinces provides 2 key insights. First, local resource endowments may shape the market size and social media influence of hospitals. Therefore, in analyzing hospital social media performance, it is essential to consider not only hospital-specific characteristics and promotional strategies but also regional influences. Second, Figure 1 reveals that hospitals are most densely concentrated in the eastern coastal regions, relatively sparse in the western provinces, and moderately distributed in the central and northeastern regions.

To facilitate the analysis of regional heterogeneity, we divided China into 4 major regions: eastern (Beijing, Tianjin, Hebei, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong, and Hainan), central (Shanxi, Anhui, Jiangxi, Henan, Hubei, and Hunan), northeastern (Liaoning, Jilin, and Heilongjiang), and western (Inner Mongolia, Guangxi, Chongqing, Sichuan, Guizhou, Yunnan, Tibet, Shaanxi, Gansu, Qinghai, Ningxia, and Xinjiang). The distribution of hospital counts across these 4 regions is presented in Figure 2.

The regression results of equation 1 are presented in Table 2. First, the activity and interactivity of hospitals’ Weibo accounts significantly influenced all 4 dimensions of social media impact. Specifically, when the dependent variables were ln(wbFans), ln(wbLike), ln(wbComment), and ln(wbRetweet), the coefficients for ln(wbNum) were 0.482 (95% CI 0.399-0.565; P<.001), 0.435 (95% CI 0.338-0.532; P<.001), 0.309 (95% CI 0.224-0.395; P<.001), and 0.489 (95% CI 0.407-0.572; P<.001), respectively. This indicates that a 10% increase in the number of Weibo posts by a hospital corresponded to increases of 4.8% in followers (wbFans), 4.4% in maximum likes (wbLike), 3.1% in maximum comments (wbComment), and 4.9% in maximum shares (wbRetweet). Similarly, the coefficients for ln(wbFollow) were 0.447 (95% CI 0.320-0.574; P<.001), 0.241 (95% CI 0.092-0.391; P=.002), 0.235 (95% CI 0.104-0.367; P<.001), and 0.306 (95% CI 0.179-0.432; P<.001) when the dependent variables were ln(wbFans), ln(wbLike), ln(wbComment), and ln(wbRetweet), respectively. This suggests that a 10% increase in the number of accounts followed by a hospital corresponds to increases of 4.5% in followers, 2.4% in maximum likes, 2.4% in maximum comments, and 3.1% in maximum shares.

Second, the entertainment value of hospitals’ Weibo accounts had a significant positive effect only on certain dimensions of social media influence, specifically audience reach and audience approval. For instance, the coefficient for wbPicture*100 was significant only when the dependent variables were ln(wbFans) and ln(wbLike), with values of 0.006 (95% CI 0.001-0.012; P=.02) and 0.005 (95% CI −0.001 to 0.012; P=.10), respectively. This indicates that doubling the percentage of posts containing images increased followers by 0.6% and maximum likes by 0.5%. Similarly, the coefficient for wbVideo*100 was significant only for maximum likes—ln(wbLike)—with a value of 0.012 (95% CI 0.002-0.022; P=.02), indicating that doubling the percentage of posts containing videos increased maximum likes by 1.2%.

Moreover, among the hospital characteristics, hospital size and ranking had notable impacts on social media influence. Specifically, the coefficients for ln(Bed) were 0.466 (95% CI 0.251-0.680; P<.001), 0.690 (95% CI 0.438-0.942; P<.001), 0.381 (95% CI 0.160-0.602; P=.001), and 0.432 (95% CI 0.219-0.646; P<.001) when the dependent variables were followers, maximum likes, maximum comments, and maximum shares, respectively. This suggests that a 10% increase in the number of authorized beds was associated with increases of 4.7% in followers, 6.9% in maximum likes, 3.8% in maximum comments, and 4.3% in maximum shares. For the same dependent variables, the coefficients for Rank/100 were −0.02 (95% CI −0.030 to −0.009; P<.001), −0.014 (95% CI −0.027 to −0.001; P=.04), −0.008 (95% CI −0.019 to 0.003; P=.16), and −0.011 (95% CI −0.022 to −0.000; P=.049), respectively. This indicates that, except for maximum likes, a 100-rank improvement in hospital ranking increased followers by 2%, maximum comments by 1.4%, and maximum shares by 1.1%.

First, the regression results of equation 2 are presented in Table 3, indicating that promotional strategies did not exhibit significant differences across hospital types, with only certain characteristics such as ranking and ownership influencing some aspects of promotional strategies. Specifically, after controlling for other major hospital characteristics, higher-ranked hospitals tended to exhibit greater activity on Weibo. When the dependent variable was ln(wbNum), the coefficient for Rank/100 was −0.037 (95% CI −0.052 to −0.022; P<.001), indicating that, for every 100-place improvement in ranking, the volume of Weibo posts increased by 3.7%.

Specifically, public or private ownership and hospital size significantly affected the role of the entertainment value of social media content in enhancing social media influence. As shown in Table 4, when the dependent variables were ln(wbFans), ln(wbLike), ln(wbComment), and ln(wbRetweet), respectively, the coefficients for wbPicture*100*Private were 0.023 (95% CI 0.003-0.044; P=.03), 0.022 (95% CI −0.002 to 0.047; P=.07), 0.029 (95% CI 0.007-0.050; P=.009), and 0.024 (95% CI 0.004-0.045; P=.02), and the coefficients for wbVideo*100*Private were 0.036 (95% CI 0.004-0.068; P=.03), 0.027 (95% CI −0.011 to 0.065; P=.17), 0.036 (95% CI 0.003-0.070; P=.03), and 0.027 (95% CI −0.005 to 0.060; P=.09), indicating that private hospitals achieved better results across all 4 social media dimensions (audience reach, audience approval, audience interaction, and dissemination) by enhancing the entertainment value compared to public hospitals. In addition, in Table 5, G_scale is the dummy variable indicating whether the number of beds exceeded the sample median. Compared to small-sized hospitals, large hospitals’ entertainment content had a more significant effect in improving audience approval (the coefficient for wbPicture*100*G_scale was 0.010, 95% CI −0.001 to 0.022; P=.08) and engagement (the coefficient for wbPicture*100*G_scale was 0.009, 95% CI −0.001 to 0.019; P=.09).

Moreover, private hospitals demonstrated higher activity and interactivity on Weibo than public hospitals. When the dependent variables were ln(wbNum) and ln(wbFollow), the coefficients for Private were 1.175 (95% CI 0.335-2.015; P=.006) and 1.080 (95% CI 0.581-1.579; P<.001), respectively, suggesting that private hospitals had approximately twice as many Weibo posts and followed twice as many accounts as public hospitals. Other hospital characteristics did not show significant effects on promotional strategies.

Second, we examined the impact of hospital characteristics on the effectiveness of social media promotional strategies through interaction terms. The hospital characteristics included bed count (Bed); years of operation (History); overall ranking (Rank); and binary indicators such as whether the hospital was a tertiary institution (Tertiary), private hospital (Private), or general hospital (General), as well as the region where the hospital was located. The results indicate that a hospital’s ownership (public or private), size, ranking, and region were significant factors influencing the effectiveness of social media promotional strategies.

In addition, the heterogeneous analysis in terms of ranking is shown in Table 6, where G_rank is the dummy variable indicating whether the rank was higher than the sample median. A hospital’s ranking significantly influenced the role of interactivity in audience interaction and dissemination on social media. The interaction term analysis showed that higher-ranked hospitals significantly increased their social media influence by enhancing interactivity, especially in the dimensions of audience interaction (the coefficients for ln(wbFollow)*G_rank were 0.183, 95% CI −0.030 to 0.396 with P=.09 and 0.182, 95% CI −0.022 to 0.387 with P=.08 when the explained variables were ln(wbComment) and ln(wbRetweet)) and content dissemination (the coefficient for wbVideo*100*G_scale was 0.017, 95% CI 0.001-0.032; P=.04).

For the regional interaction analysis, the regression results are shown in Table 7, with G_region1 as the dummy variable indicating whether the hospital was located in the central and northeast regions and G_region2 as the dummy variable indicating whether the hospital was located in the eastern region. Improving activity in the eastern region significantly enhanced audience interaction and dissemination. In the western and central regions, increasing the entertainment value of social media content significantly boosted social media influence, particularly regarding audience approval (the coefficient for wbVideo*100*G_region2 was 0.021, 95% CI −0.002 to 0.044; P=.07) and audience interaction (the coefficient for wbVideo*100*G_region2 was 0.024, 95% CI 0.003-0.044; P=.02).

This study further conducted comparative analyses of the effects of hospital promotional strategies on social media influence under different model specifications, as presented in Multimedia Appendix 1. Due to space constraints, we report only the regression coefficients and the corresponding P values. Models 1 to 4 refer to fixed-effects models that included only promotional strategies. Models 5 to 8 were based on the baseline specification (equation 1) with hospital-level control variables, whereas models 9 to 12 were derived from the mixed-effects model described in equation 5.

Several key findings emerge from the regression results. First, a comparison between models 5 to 8 and models 9 to 12 shows that the main findings remained consistent regardless of whether a fixed-effects or mixed-effects model was used. Second, comparing models 5 to 8 with models 1 to 4 reveals that the estimated effects were not sensitive to the inclusion or exclusion of hospital-level or provincial-level control variables. Together, these results underscore the robustness of our findings regarding the effects of hospital promotional strategies on their social media influence. Third, the mixed-effects model allowed us to examine the additional effects of provincial-level characteristics. Among these, residents’ disposable income showed a significant positive association with hospitals’ social media influence, whereas population density and age structure appeared to have no statistically significant impact.

In addition, we adjusted the P values using the Benjamini-Hochberg FDR procedure. Due to space limitations, we report only the original and adjusted P values from the baseline specification under both the fixed-effects and mixed-effects models, as shown in Multimedia Appendix 1. The results indicate that, even after the FDR adjustment, the significance levels of the key explanatory variables—hospital promotional strategies—remained stable, providing further support for the robustness of our findings.

This study empirically examined the relationships among 3 dimensions of hospital promotional strategies (activity, interactivity, and entertainment value) and 4 aspects of hospital social media influence (audience reach, audience approval, audience interaction, and dissemination capability) while also analyzing the moderating effects of hospital characteristics. Overall, the findings largely support the hypotheses derived from the uses and gratifications theory, dialogic communication theory, elaboration likelihood model, resource-based view, and institutional theory.

First, hospital social media activity (posting frequency) showed significant positive correlations across all dimensions of social media influence, validating hypothesis 1. According to the uses and gratifications theory, users actively seek information, social interactions, and timely updates. Hospitals that frequently post content tend to effectively fulfill these user needs and, therefore, are more likely to achieve enhanced audience reach, approval, and interaction and content dissemination. In addition, frequent, high-quality content can boost user trust and loyalty, further enhancing the hospital’s online influence [43,44].

Second, hospital social media interactivity (number of accounts followed) exhibited significant positive correlations with all aspects of social media influence, supporting hypothesis 2 and dialogic communication theory. This theory underscores the importance of 2-way communication and interactive mechanisms for building trust and deepening audience engagement. Through actively interacting with other accounts, hospitals demonstrate transparency and openness, which tends to foster increased audience responsiveness and content dissemination [45-47].

In contrast, entertainment content (such as images and videos) primarily showed significant positive correlations with audience reach and approval but not with deeper interactions and dissemination effects, partially supporting hypothesis 3. According to the elaboration likelihood model, entertainment content acts as a peripheral cue, typically prompting superficial interactions such as likes or follows. However, when this content integrates substantial professional information, it might encourage deeper cognitive engagement (comments and shares). The current scarcity of high-information content posted by hospitals likely limits the deeper dissemination potential of entertainment-oriented posts [48]. Future research should further clarify and validate distinctions between central and peripheral processing pathways.

Interaction term analyses further explored how hospital characteristics moderated promotional strategy effectiveness, offering detailed insights for tailored strategies. The examined characteristics included approved bed numbers; years of operation; overall ranking; indicators for tertiary-level, public, or general hospitals; and regional classification (eastern, northeastern, central, and western China). The results revealed significant moderating effects of ownership (public vs private), size, ranking, and region, supporting hypothesis 4.

Hospital ownership and size critically moderated the relationship between entertainment content and social media influence. Private hospitals outperformed others in leveraging entertainment content to boost influence, likely reflecting their greater organizational flexibility and market orientation toward user-centric content strategies. In addition, private hospitals typically prioritize public communication and brand investment, making their entertainment content more engaging. Similarly, larger hospitals were more effective than smaller ones in enhancing audience approval and interaction through entertainment content due to stronger content creation capabilities and brand reputation. According to the resource-based view, larger and private hospitals are better positioned to use internal resources to create appealing content and amplify its dissemination.

Hospital ranking significantly moderated the impact of interactivity on social media effectiveness. Higher-ranked hospitals effectively leveraged interactive behaviors (eg, mutual following and comments) to enhance audience interaction and dissemination capabilities, consistent with dialogic communication theory, suggesting that trusted institutions more effectively stimulate persuasive interactions, enhancing user responsiveness.

Institutional theory suggests that regional economic and social development significantly impact the effectiveness of social media promotional strategies. In the economically developed eastern region, higher activity (frequent posting) significantly improved audience interaction and dissemination, likely due to user preferences for frequent updates and algorithm-driven content visibility. In contrast, in the less economically developed central, western, and northeastern regions, increasing entertainment content more effectively enhanced audience engagement and approval, indicating that users in these areas responded more strongly to visually and emotionally engaging content. Thus, hospitals should tailor content strategies based on regional cultural and informational preferences to enhance user engagement.

In summary, hospitals currently show minimal differentiation in promotional strategies; however, future communication plans should avoid a one-size-fits-all approach and, instead, consider internal resources and external environmental factors. Hospitals should maintain high levels of activity and interactivity while flexibly adjusting entertainment content according to ownership, size, ranking, and regional characteristics to achieve multidimensional influence enhancement. This optimization strategy aligns with the resource-based view, emphasizing internal capabilities as strategic success factors, and resonates with institutional theory’s emphasis on external norms, regulations, and expectations, jointly guiding hospitals toward building influential and trustworthy brands in the digital era.

The contributions of this study are as follows. First, this study integrated multidisciplinary perspectives from communication studies (uses and gratifications theory and dialogic communication theory), marketing (elaboration likelihood model), and organizational management (resource-based view and institutional theory) to systematically analyze the relationship between hospitals’ social media promotional strategies and their online influence. This integrated analytical framework transcends the limitations of previous research, which either focused solely on internal hospital characteristics (such as size, type, and geographic location) [1,49,50] or relied primarily on small-sample case studies for preliminary evaluation of promotional effectiveness [1,4,51]. Thus, this framework provides a novel theoretical perspective for comprehensively understanding the social media communication mechanisms of professional service institutions.

Second, this study innovatively categorized hospitals’ social media promotional strategies into 3 core dimensions—activity (posting frequency), interactivity (mutual following), and entertainment value (multimedia content)—and used empirical research methods to validate the differential effects of each dimension on the 4 elements of influence (reach, approval, interaction, and dissemination power). This theoretical innovation addresses the limitations of existing literature that analyzed single functions of social media (eg, health education or brand promotion) [5,52-55], providing a more comprehensive and systematic analytical framework for health communication research.

Third, this study innovatively incorporated hospital heterogeneity (size, level, and ownership) and regional differences into the analytical framework, revealing that organizational resource endowments and institutional environments significantly moderated the effects of promotional strategies. This finding not only provides an important theoretical basis for the contextual adaptation of social media promotional strategies but also expands the application boundaries of strategic communication theory in professional service fields.

Finally, based on the empirical findings, this study proposes a strategy principle emphasizing professional content first, supplemented by entertainment along with differentiated implementation plans, providing scientific evidence for medical institutions to optimize social media investments under resource constraints. In addition, this principle establishes an analytical perspective that integrates theoretical insights with practical applicability. It not only applies to the health care sector but also offers valuable implications for social media marketing practices in other professional domains such as education and public services, thus enhancing both the theoretical contribution and practical significance of the research findings.

This study has several limitations that warrant further exploration. First, the data primarily came from the Weibo platform. While Weibo is one of the largest social media platforms in China with a broad user base and robust information dissemination capabilities, integrating data from other platforms such as WeChat Channels or Douyin can provide a more comprehensive view of hospitals’ overall social media promotional strategies. Second, this study focused on official verified accounts, ensuring data reliability and credibility. However, the potential influence of unofficial accounts on public perceptions also merits further investigation. Third, this research was based on cross-sectional data, which limits the ability to analyze dynamic changes over time. Incorporating longitudinal data in future studies could help uncover trends in hospital social media performance and promotional strategies over time.

Moreover, future research could adopt longitudinal designs or experimental methods to rigorously validate the mapping between specific promotional content and theoretical constructs. For instance, content analysis could examine how different types of social media posts (eg, information-intensive vs entertainment-oriented content) differentially activate central or peripheral route processing. In addition, experimental designs could more strictly test causal mechanisms underlying the influence of interactivity and entertainment strategies on user behavior. Furthermore, the validity of theoretical constructs could be statistically assessed through scale development and validation methods such as exploratory factor analysis and structural equation modeling, further enhancing the robustness and theoretical validity of this study’s findings.

These extensions and deeper analyses will further enrich our understanding of the mechanisms behind hospitals’ social media operations.

This study provides an in-depth analysis of the relationships between 3 dimensions of hospitals’ social media promotional strategies (activity, interactivity, and entertainment value) and 4 dimensions of social media influence (audience reach, audience approval, audience interaction, and dissemination power). The findings reveal that activity and interactivity exhibit significant and broad positive associations across all dimensions, highlighting the critical importance of consistently publishing high-quality content and actively expanding interactive networks to solidify hospitals’ foundational social media influence. Although the overall contribution of entertainment value was relatively weaker, it can still play a supplementary role under specific contexts and for certain hospital types. The interaction term analysis revealed the alignment among hospital types, regional characteristics, and strategy elements, providing a clear road map for designing and implementing tailored strategies. The results indicate that hospital ownership (public vs private), size, ranking, and regional location are key factors influencing the effectiveness of social media promotional strategies. In summary, this study offers empirical evidence and strategic insights for health care institutions to construct more credible and patient-centered brand images amidst digital transformation, laying the groundwork for sustainable success in the digital age.