This was a nationwide, multicenter, cross-sectional questionnaire survey conducted from February to April 2023 in 21 public hospitals across 7 geographic regions of China: North China, Northeast China, East China, Central China, Northwest China, Southwest China, and South China. The participating hospitals were located in 15 provinces, municipalities, and autonomous regions.

A pilot survey was conducted in December 2022 among 108 physicians and medical trainees to assess questionnaire clarity and the feasibility of electronically administering the survey. The pilot confirmed the feasibility of the survey procedures, informed refinement of the questionnaire, and provided data for sample-size planning. Pilot participants were not included in the final analysis.

The questionnaire was developed based on literature review, expert consultation, and pilot testing. Content validity was assessed through expert review using the content validity ratio (CVR). Internal consistency was evaluated using the Cronbach α, with values of 0.65 for the confidence domain and 0.68 for the complacency domain. The questionnaire can be found in Multimedia Appendix 1.

Eligible participants were physicians and medical trainees working or training at the participating hospitals during the study period. The survey focused on the physician workforce across career stages because physicians’ vaccination attitudes may influence patient counseling and vaccination-related clinical decisions, while medical trainees represent the earliest stage of physician training. Recruitment was organized at the hospital level using prespecified target numbers for each hospital and career-stage group, including medical trainees who had not begun standardized residency training, residents or physicians undergoing standardized residency training, attending physicians, associate chief physicians, and chief physicians.

At each hospital, 2 designated investigators coordinated recruitment and questionnaire distribution. To ensure broad representation across clinical departments, investigators sought to include participants from departments directly involved in COVID-19 care, including pulmonary and critical care medicine, emergency medicine, infectious diseases, and intensive care medicine, as well as other departments caring for patients at increased risk of severe COVID-19. Participants with and without experience treating patients with COVID-19 were actively recruited. Representation across clinical departments and experience treating patients with COVID-19 were recruitment targets rather than prespecified quotas.

The final questionnaire was self-administered electronically through Wenjuanxing (Changsha Ranxing Information Technology Co Ltd), and the survey link was distributed through direct electronic invitations and department-level WeChat (Tencent Holdings Limited) groups. Recruitment continued until the prespecified target numbers for each hospital and career-stage group were reached. Because the questionnaire link was distributed through both direct electronic invitations and department-level WeChat groups, the exact number of individuals who received or viewed the invitation could not be precisely determined; therefore, a response rate could not be calculated. The survey platform allowed only 1 submission per WeChat ID to reduce duplicate responses.

The sample size was estimated using PASS version 21.0 (NCSS, LLC) based on pilot survey data. The calculation incorporated a design effect of 1.2 to account for potential clustering within hospitals. Each hospital aimed to recruit approximately 150 eligible participants, yielding a target sample size of approximately 3150 participants across 21 hospitals. All submitted questionnaires with complete outcome and covariate information required for the main analyses were included in the final analytic dataset.

The primary outcome was hesitancy toward future annual COVID-19 vaccination, assessed using the following question: “Would you proactively receive the COVID-19 vaccine every year in the future?” Participants who answered “no” were classified as having hesitancy toward future annual COVID-19 vaccination and were asked to select reasons for hesitancy.

Secondary outcomes were hesitancy to support COVID-19 vaccination for older adults and for young children in participants’ families. These outcomes were assessed using the following questions: “If older adults in your family aged over 65 years have no clear contraindications, would you support their COVID-19 vaccination?” and “If there are children younger than 6 years in your family, would you support their COVID-19 vaccination?” Participants who answered “no” were classified as having hesitancy to support COVID-19 vaccination for the corresponding group.

The questionnaire collected information on age, education level, geographic region, hospital classification, clinical department, career stage, SARS-CoV-2 infection history, experience treating patients with COVID-19, prior COVID-19 vaccination, type of COVID-19 vaccine received, willingness to receive influenza vaccination, and selected perceptions related to COVID-19. Data on sex and income were not collected in this survey.

Regions were categorized as low, middle, or high according to the Human Development Index (HDI) of the province-level administrative regions in which the participating hospitals were located. Hospitals were classified as top grade A tertiary hospitals, other grade A tertiary hospitals, and nontertiary or other hospitals based on China’s hospital classification system and prespecified hospital ranking criteria. Clinical departments were categorized as COVID-19–related internal medicine departments, other internal medicine departments, surgical departments, or other departments. Educational attainment was categorized as doctoral degree, master’s degree, and college degree or lower.

The study protocol was reviewed and approved by the Ethics Review Committee for Biomedical Research Involving Human Subjects of the Chinese Academy of Medical Sciences and Peking Union Medical College (approval number CAMS&PUMC-IEC-2023‐030). This study was conducted in accordance with the Declaration of Helsinki. Before completing the questionnaire, all participants provided electronic informed consent (Multimedia Appendix 2).

The questionnaire did not collect participants’ names, national identification numbers, home addresses, or other identifying information for research analysis. Mobile phone numbers were collected solely for distributing participant compensation and were excluded from the analytic dataset. The WeChat ID restriction was used only to prevent duplicate submissions, and WeChat identifiers were not included in the analytic dataset. Survey data were stored in a password-protected database accessible only to authorized study investigators. Participants received CNY 50 (US $1=CNY 6.87 as of March 1, 2023) after completing the questionnaire.

Participant characteristics were summarized as frequencies and percentages and compared according to hesitancy toward future annual COVID-19 vaccination (Table 1). Group comparisons were conducted according to expected cell counts. The Pearson χ² test was used when all expected cell counts were 5 or greater, and the Fisher exact test was used when any expected cell count was less than 5. All tests were 2-sided, and a P value <.05 was considered statistically significant.

Because the primary and secondary outcomes were common, modified Poisson regression models with robust SEs were used to estimate unadjusted prevalence ratios (PRs), adjusted prevalence ratios (aPRs), and 95% CIs. All multivariable models were adjusted for age, education level, HDI level, hospital classification, clinical department, career stage, SARS-CoV-2 infection history, experience treating patients with COVID-19, and prior COVID-19 vaccination. The model for hesitancy toward future annual COVID-19 vaccination additionally included willingness to receive influenza vaccination, whereas the models for hesitancy to support vaccination for older adults and young children included participants’ own hesitancy toward future annual COVID-19 vaccination. Full univariable and multivariable modified Poisson regression results are reported in Table S4 in Multimedia Appendix 3. All analyses were performed using R software (version 4.3.2; R Foundation for Statistical Computing).

Among the 3169 participants included in the final analysis, the largest age group was 30 to 39 years (1337/3169, 42.2%), followed by 20 to 29 years (1125/3169, 35.5%). Overall, 1173 participants (37.01%) were residents, 1010 (31.87%) were attending physicians, 694 (21.9%) were associate chief or chief physicians, and 292 (9.2%) were medical trainees. Most participants reported a history of SARS-CoV-2 infection (2954/3169, 93.2%) and experience treating patients with COVID-19 (2568/3169, 81%).

Participant characteristics stratified by hesitancy toward future annual COVID-19 vaccination are shown in Table 1. Participants with and without hesitancy differed by age (P=.001), education level (P<.001), HDI level (P=.007), hospital classification (P<.001), career stage (P=.005), prior COVID-19 vaccination (P<.001), and willingness to receive influenza vaccination (P<.001). No significant differences were observed for clinical department, SARS-CoV-2 infection history, or experience treating patients with COVID-19.

Overall, 1899 of 3169 participants (59.9%) were willing to receive future annual COVID-19 vaccination, whereas 1270 (40.1%) were classified as having hesitancy toward future annual COVID-19 vaccination. Most participants had previously received at least 1 dose of a COVID-19 vaccine (3074/3169, 97%). Among previously vaccinated participants, 2862 of 3074 (93.1%) had received an inactivated vaccine, and 1864 (60.6%) were willing to receive future annual COVID-19 vaccination. Vaccine types previously received and preferred vaccine types among participants willing to receive future COVID-19 vaccination are shown in Figure S1 of Multimedia Appendix 3.

In addition, 1932 participants (61%) were willing to receive influenza vaccination. Among participants willing to receive influenza vaccination, 1587 of 1932 (82.1%) were also willing to receive future annual COVID-19 vaccination compared with 312 of 1237 (25.2%) among those unwilling to receive influenza vaccination.

In adjusted analyses, unwillingness to receive influenza vaccination was strongly associated with a higher prevalence of hesitancy toward future annual COVID-19 vaccination (aPR 4.11; 95% CI 3.71‐4.55). Not having previously received COVID-19 vaccination was also associated with a higher prevalence of hesitancy (aPR 1.27; 95% CI 1.11‐1.46). Compared with participants with college-level education or lower, those with a master’s degree (aPR 1.11; 95% CI 1.02‐1.21) or doctoral degree (aPR 1.20; 95% CI 1.07‐1.36) had a higher prevalence of hesitancy. Compared with participants from top grade A tertiary hospitals, those from other grade A tertiary hospitals (aPR 0.89; 95% CI 0.81‐0.97) and nontertiary or other hospitals (aPR 0.81; 95% CI 0.72‐0.90) had a lower prevalence of hesitancy (Figure 1). Full results for this outcome are shown in Table S4A of Multimedia Appendix 3.

Overall, 2857 of 3169 participants (90.2%) were willing to support COVID-19 vaccination for adults aged over 65 years without contraindications, and 2320 (73.2%) were willing to support vaccination for children younger than 6 years. For both target groups, the proportion willing to support vaccination was lower among participants with hesitancy toward future annual COVID-19 vaccination. Among hesitant participants, 1026 of 1270 (80.8%) supported vaccination for older adults and 705 (55.5%) supported vaccination for young children compared with 1831 of 1899 (96.4%) and 1615 (85%), respectively, among those willing to receive future annual COVID-19 vaccination.

In adjusted analyses, participants who were hesitant about receiving future annual COVID-19 vaccination themselves had a higher prevalence of hesitancy to support COVID-19 vaccination for older adults (aPR 5.13; 95% CI 3.95‐6.66). Hesitancy to support vaccination for older adults was also higher among participants from other internal medicine departments (aPR 1.36; 95% CI 1.05‐1.78) and surgical departments (aPR 1.69; 95% CI 1.24‐2.31) compared with those from COVID-19–related internal medicine departments. Attending physicians (aPR 0.57; 95% CI 0.35‐0.93), associate chief or chief physicians (aPR 0.48; 95% CI 0.26‐0.89), and participants with a history of SARS-CoV-2 infection (aPR 0.71; 95% CI 0.51‐0.98) had a lower prevalence of hesitancy to support vaccination for older adults (Figure 2A). Full results for this outcome are shown in Table S4B of Multimedia Appendix 3.

Similar associations were observed for hesitancy to support vaccination for young children. Participants who were hesitant about receiving future annual COVID-19 vaccination themselves had a higher prevalence of hesitancy to support COVID-19 vaccination for young children (aPR 2.88; 95% CI 2.54‐3.27). Not having previously received COVID-19 vaccination was also associated with a higher prevalence of hesitancy to support vaccination for young children (aPR 1.42; 95% CI 1.16‐1.74). Compared with participants aged 20 to 29 years, those aged 40 to 49 years (aPR 1.39; 95% CI 1.07‐1.79) and 50 years or older (aPR 1.52; 95% CI 1.13‐2.06) had a higher prevalence of hesitancy. Participants from middle-HDI regions (aPR 0.82; 95% CI 0.71‐0.94), from low-HDI regions (aPR 0.75; 95% CI 0.63‐0.89), and with a history of SARS-CoV-2 infection (aPR 0.82; 95% CI 0.67‐0.99) had a lower prevalence of hesitancy to support vaccination for young children (Figure 2B). Full results for this outcome are shown in Table S4C of Multimedia Appendix 3.

Among participants with hesitancy toward future annual COVID-19 vaccination, the most frequently reported reasons were concerns about vaccine effectiveness (674/1270, 53.1%), concerns about vaccine safety (424/1270, 33.4%), and insufficient disclosure of domestic clinical trial data (368/1270, 29%). Additional responses included perceiving vaccination as unnecessary because they were in good health (235/1270, 18.5%), other reasons (221/1270, 17.4%), fear of injection (115/1270, 9.1%), and concerns about underlying disease or contraindications to vaccination (73/1270, 5.7%) (Figure 3A).

Among participants with influenza vaccination hesitancy, the most frequently reported reasons were concerns about vaccine effectiveness (545/1237, 44.1%), perception of vaccination as unnecessary because they were in good health (485/1237, 39.2%), lack of time (256/1237, 20.7%), and concerns about vaccine safety (241/1237, 19.5%). Additional responses included out-of-pocket cost (168/1237, 13.6%), other reasons (149/1237, 12%), not knowing where to receive vaccination (62/1237, 5%), concerns about underlying disease or contraindications to vaccination (52/1237, 4.2%), and community vaccine shortage (35/1237, 2.8%) (Figure 3B).