Hesitancy regarding providing vaccine recommendations among clinicians in China: Analysis based on the behavior and social driver model

To the Editor: To enhance the responsibility of clinicians in the prevention and control of infectious diseases, China has released a document on infectious disease prevention and control responsibilities of medical institutions.[1] Although the document released emphasizes the role of clinicians in providing vaccination advice, factors such as busy schedules and lack of incentives reduce the clinical implementation of vaccine-related activities.[2] Hence, exploring relevant measures to enhance clinicians’ engagement in vaccine-related initiatives is imperative to enable them to fulfill their public health roles and promote population well-being. In China, clinicians primarily focus on disease treatment, and few directly administer vaccines or offer vaccine-related services. Only a few tertiary and secondary medical institutions provide vaccination services, and clinicians are seldom involved in these activities. However, as the primary point of contact for patients and their families, clinicians’ attitudes toward vaccination not only impact personal health but also influence infectious disease prevention and control mechanisms within medical institutions. Their recommendation behaviors directly affect patient vaccination practices.[3] The World Health Organization introduced the behavioral and social drivers (BeSD) theoretical framework model in 2022,[4] emphasizing the crucial role of clinicians’ awareness and recommendations in shaping vaccination behavior. The BeSD framework describes the social and behavioral determinants of vaccine hesitancy, with clinicians contributing to social processes and practical issues. Although many studies have used the BeSD model to conduct vaccine hesitancy among clinicians, there are still few studies using this method to investigate the recommendation behavior and willingness of clinicians. Therefore, this study analyzed the factors influencing clinicians’ willingness to recommend vaccines to inform public health policy and strengthen their role in disease prevention. A cross-sectional survey was conducted in 29 provinces in China between December 25, 2023 and January 3, 2024. IQVIA (IQVIA Inc., Durham, North Carolina, United States) was used to distribute questionnaires on the WeChat platform in China. Questionnaires were distributed to clinicians who had registered on the platform. The questionnaire comprised questions on sociodemographic information, presence of a vaccine clinic in the hospital, clinicians’ willingness to provide vaccination advice or prescriptions to patients, vaccine-related education, vaccination behavior, attitudes toward providing advice or prescription, and attitudes toward the person who provided the vaccination service in the hospital. The Cronbach’s α coefficient of our questionnaire was 0.765, which had good reliability. The study was approved by the Medical Ethics Committee of the Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China (Approval No. CAMS&PUMC-IEC-2022-019). To investigate the factors associated with the propensity score for recommending influenza vaccination to patients with chronic diseases, two mediation models were hypothesized and explored. The first model reflected influenza vaccination behavior propensity in the thinking and feeling, practical issues, and social process categories. The second model focused on thinking and feeling, and explored negative opinions, including vaccine prevention not being effective, influenza infection not being serious, and vaccination not being required. RStudio (version 2023.03.0 + 386; RStudio Team, 2023) was used to perform data analysis. The correlation analysis and mediation regression effects were analyzed using psych and PROCESS R packages (Hayes, A. F., The Ohio State University, Columbus, OH, USA). Overall, a total of 2220 questionnaires from 198 cities in China were collected. Respondents who did not provide consent (n = 8) or were working in community health centers (n = 3) were excluded. The sample population exhibited extensive medical-related work experience, averaging 17.0 (interquartile range: 8.8–25.4) years, with women comprising 51.7% (1141/2209). 94.4% (2085/2209) clinicians were from general hospitals, and 53.6% (1183/2209) held positions as chief physicians or associate chief physicians. In addition, 47.5% (1049/2209) of the clinicians attained a master’s degree or higher through specialized training programs. Among all respondents, 55.9% (1234/2209) indicated that their hospital had vaccination clinics [Supplementary Table 1, https://links.lww.com/CM9/C642]. Among clinicians willing to provide vaccine-related advice or prescriptions (n = 2129), the highest proportion of participants who considered vaccine effectiveness paramount was in the “thinking and feeling” domain (55.8%, 1189/2129). Moreover, “practical issues” indicated that providing vaccine-related recommendations was deemed effective in promoting patient health (56.3%, 1198/2129). This surpassed the “social processes” associated with a history of clinician-recommended vaccination behavior (55.4%, 1180/2129) [Figure 1A]. Among clinicians who were unwilling to provide advice or prescriptions (n = 80), 45.0% (36/80) of clinicians who worried about adverse reactions caused by vaccination required risk responsibility. This surpassed concerns about the risk of influenza infection (21.3%, 17/80). In the “social process” domain, concerns regarding medical disputes stemming from recommendations (43.8%, 35/80) exceeded those regarding hospital requirements (41.2%, 33/80) [Figure 1B].Figure 1: Behavior and social drivers of clinicians’ attitudes toward providing vaccination recommendations to patients. (A) Reasons for clinicians’ willingness to provide vaccination advice or prescriptions to patients (n = 2129). (B) Reasons for clinicians’ unwillingness to provide vaccination advice or prescriptions to patients (n = 80).The correlation analysis showed that the propensity score for “recommending influenza vaccination to chronic patients” was negatively associated with “thinking and feeling” (r = −0.307), “practical issues” (r = 0.049), and “social processes” (r = −0.016). This indicated that “thinking and feeling” was associated with a lower tendency to recommend vaccination than social processes. Further exploration was conducted of specific factors related to “thinking and feeling”. The correlation analysis showed that the propensity score for “recommending influenza vaccination to chronic patients” was negatively associated with concerns about adverse reactions (r = −0.278), ineffective vaccine prevention (r = −0.312), and “influenza infection not being serious and not requiring vaccination” (r = −0.148). This indicated that “considering vaccine prevention ineffective” was associated with a lower tendency to recommend vaccines than “social processes”. The factors influencing clinicians’ recommendation intentions were analyzed based on the BeSD. The structural model was then estimated and tested. The Bootstrap method (bootstrap set: 10,000) was used to determine the mediating effect of clinicians’ individual vaccination behaviors on the intention to recommend vaccination to patients. Years of medical work experience, hospital level, and city tier were used as control variables. City tier was classified based on a comprehensive assessment of each city’s economic development level, population size, administrative importance, and cultural influence. The indirect effect indicated that an increase in vaccine behavior by 1 standard deviation (SD) was significantly associated with an increase in “recommendation intention” by 0.16 SD through “thinking and feeling”. Furthermore, the indirect effect indicated that an increase in “vaccine behavior” by 1 SD was significantly associated with an increase in “recommendation intention” by 0.04 SD through “worried about adverse reactions” and by 0.08 SD through “vaccine prevention is not effective” [Supplementary Table 2, https://links.lww.com/CM9/C642]. In addition, the mediating effect of the “thinking and feeling” accounted for 66.7% of the total effect size, indicated that the “thinking and feeling” accounted for the highest proportion of mediating effect between clinicians’ individual “vaccination behavior” and “recommendation intention”. Furthermore, the perception that “worried about vaccine prevention is ineffective” accounted for 34.5% of the total effect size, indicating that “worried about vaccine prevention is ineffective” accounted for the highest proportion of mediating effect between clinicians’ “individual vaccination behavior” and “recommendation intention” [Supplementary Table 2, https://links.lww.com/CM9/C642]. The results revealed a changing trend in the incentives for clinicians to participate in providing vaccine advice or prescriptions by investigating the expected incentives for clinicians. Professional recognition ranked highest, comprising 28.2% (623/2209), and policy or institutional requirements accounted for 20.7% (458/2209) as the second most frequently cited incentive. Our findings suggest that over 80% of clinicians are motivated to provide vaccine-related advice and prescriptions by four key incentives: policy requirements, personal recognition and prestigious titles, direct bonuses, and performance evaluations [Supplementary Figure 1, https://links.lww.com/CM9/C642]. This study examined clinicians providing recommendations for vaccinations to patients and revealed that clinicians’ personal vaccination behavior influenced their willingness to advise vaccinations to patients, with “thinking and feeling” about vaccinations playing a mediating role. Although most clinicians were willing to provide vaccine recommendations, concerns about vaccine safety and distrust have made clinicians hesitant to recommend vaccination to others. Enhancing vaccination and related knowledge education among clinicians is essential to bolster their ability to effectively promote immunization and address vaccine hesitancy. In the context of China’s efforts to promote the combination of treatment and prevention, vaccines are a key entry point for clinicians to fulfill their public health responsibilities. Our study investigated the critical role played by healthcare providers in vaccine prescription recommendation and application. Most of the doctors were willing to provide vaccine recommendations, indicating that the convenience and accessibility of vaccination services in various regions can be strengthened in the future, such as increasing the setting of vaccination clinics to promote the practical application of treatment and prevention. The primary limitation of this study lies in its data collection method, which relied on an online questionnaire and is susceptible to participants and recall biases. In addition, the cross-sectional design limits the ability to establish causal relationships. This study utilized the BeSD model to investigate clinicians’ vaccination recommendation intentions in China, offering compelling evidence for the adaptation of BeSD tools in China. The novel insights from this research provide recommendations for refining and implementing the BeSD model. In addition to clinicians’ individual vaccine hesitancy, factors impeding clinicians’ personal vaccination may indirectly influence their inclination to recommend vaccines to patients. The localization of pertinent models should be harmonized with the policy frameworks of diverse nations to guide model optimization aimed at addressing clinicians’ hesitancy to provide vaccine recommendations and promote vaccination. Acknowledgments Thanks are given to IQVIA platform for providing questionnaires. Funding This research was supported by grants from the Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences (No.2021-RC330-002), and Bill & Melinda Gates Foundation (Nos. INV-023808, INV-006373). Conflicts of interest None.