This nationwide cross-sectional survey in China was conducted by West China Hospital, a national medical center located in Chengdu, Sichuan Province [29], from September 5 to December 31, 2023. The study adhered to the CHERRIES (Checklist for Reporting Results of Internet E-Surveys) reporting guidelines [30].

Inclusion criteria were (1) participants must be anesthesia providers (certified registered anesthesiologists and nurse anesthetists), providing direct clinical care in both public and private health care facilities; (2) participants should be able to independently complete the questionnaire; (3) informed consent must be obtained; and (4) participants must have access to the internet and suitable devices for the web-based survey.

The questionnaire was developed based on the knowledge, attitude, and practice (KAP) theory, which highlights the relationships among health behavior changes [31]. In addition to an extensive literature review pertaining to PAEs [23,32-37], we conducted a focus group interview with 10 stakeholders, including surgeons, anesthesia providers, epidemiologists, medical informaticists, health care policymakers, and information technicians from different health service systems, ensuring the relevance and comprehensiveness of the questionnaire items.

Internal consistency was assessed using Cronbach alpha, which yielded a value above 0.7, indicating satisfactory reliability [36]. Exploratory factor analysis showed a Kaiser-Meyer-Olkin value of 0.87, and the Bartlett test of sphericity was significant (P<.001) [38]. Factor analysis demonstrated that the cumulative contribution rate of the 6 factors with eigenvalues greater than 1 was determined to be 54.01, confirming good structural validity. Content validity was assessed by 5 experts from various fields (anesthesiology, nursing, statistics, health care policy, and administration) using a 4-point Likert scale (1=not relevant, 2=somewhat relevant, 3=quite relevant, and 4=highly relevant) [39]. The item-level content validity index was determined using the formula (the item-level content validity index=number of experts rating 3 or 4/total number of experts) [40]. Items with a content validity index greater than 0.8 were retained. After expert revisions, a pretest was conducted with 20 anesthesia professionals to further adjust and refine the questionnaire. The final version included 51 items divided into 4 sections (Multimedia Appendices 1 and 2).

The first section of the survey gathered general information about participants, including their age, gender, professional role (anesthesiologist or nurse anesthetist), educational background, professional title, length of service in health care, and the city and hospital tier at which they were used. In China, public hospitals are categorized into 3 tiers based on their medical capabilities: tier 1, tier 2, and tier 3 [41]. The higher the tier, the higher the level of medical care offered. Tier 3 hospitals are usually large academic centers in major cities [42]. Similarly, anesthesia professionals are categorized by a hierarchical system, including junior (eg, resident and nurse practitioner), intermediate (eg, attending physician and supervisor nurse), deputy senior (eg, associate chief physician and associate chief nurse), and senior (eg, chief physician and chief nurse) [43]. The promotion process for these professional titles is based on a comprehensive evaluation that considers clinical proficiency, years of experience, academic contributions, and formal education [43].

This section assessed the respondents’ knowledge of PAEs, with 12 items consisting of 2 single-choice and 10 true or false questions, based on established medical guidelines, clinical expertise, published articles, and procedural announcements related to PAEs. These sources formed the basis for our scoring criteria.

The questions encompassed various aspects, including the definition, preventability, risks to patients, and the protocols for reporting and managing PAEs. Each question had a definitive correct answer, with binary scoring: 1 point for each correct response, 0 points for incorrect responses, and allowing a maximum score of 12. Higher scores indicate a more comprehensive understanding of PAEs, while lower scores highlight areas where further knowledge enhancement is necessary.

Participants’ attitudes toward PAEs and the utilization of a mobile app–based CDSS for managing PAEs were explored in this section, which included 12 questions using a 5-point Likert scale (1=strongly disagree, 2=disagree, 3=neutral, 4=agree, 5=strongly agree). Since this section is subjective, there are no correct or incorrect answers, and the objective is to quantify attitudes. Total scores ranged from 12 to 60, with higher scores indicating a more positive outlook toward the management of PAEs.

The final segment evaluated participants’ experiences with reporting incidents, along with their perceptions of facilitators and barriers to adopting the mobile app–based CDSS for PAE management. This section included dichotomous and Likert scale questions to provide a comprehensive view of practice behaviors. No formal scoring system was applied, but detailed statistical analyses were performed on response distributions to identify trends and variations in reporting practices. The questionnaire concluded with an optional open-ended question, allowing participants to provide additional insights on PAE management and the proposed system.

The questionnaire was created and published using Wen Juan Xing, a popular web-based survey platform in China [44]. The survey homepage displayed the objectives, scope, and instructions for participants. A dedicated QR code was generated for easy access and distribution. The director of the Department of Anesthesiology at West China Hospital coordinated with the Chinese Association of Anesthesiologists, which represents anesthesia professionals from most medical institutions nationwide, to endorse and share the survey. With this association’s support, the QR code was distributed through WeChat groups affiliated with anesthesiology departments in various health care organizations, which is a common communication channel within China.

Upon submission, Wen Juan Xing’s backend automatically recorded responses. Given WeChat’s strict real-name verification policy for account creation and Wen Juan Xing’s enforcement of a single submission per unique IP address, it was ensured that each participant used a distinct WeChat account and made only one submission. The platform concealed each participant’s privacy, and no personally identifiable information was collected. Thereafter, data were directly downloaded from Wen Juan Xing to a Microsoft Excel spreadsheet. After excluding duplicate and invalid responses, the organized dataset was prepared for the final analysis.

Sample size estimation was based on a previous study surveying a conceptual mobile app–based system for reporting AEs following influenza vaccination, which indicated an 86% acceptance rate among physicians [45]. Using PASS 15 software (NCSS, LLC), we calculated the required sample size as 866, adjusting for a 10% attrition rate, an SD 5% margin of error, and a 95% CI. To reflect the distribution of anesthesiologists and nurse anesthetists in China, which is 76.7% and 23.3% respectively, the final sample consisted of 664 anesthesiologists and 202 nurse anesthetists.

Data were extracted from the Wen Juan Xing platform and analyzed using SPSS (version 26.0; IBM Corp) and R (version 4.3.1; R Core Team). Continuous variables conforming to a normal distribution were described as means with SDs, while nonnormally distributed variables were presented as medians with IQRs. Categorical variables were expressed as frequencies and percentages. Descriptive statistics, including chi-square tests, Kruskal-Wallis tests, and ANOVA, were applied according to the data characteristics.

The analysis began by examining the correlations between participants’ sociodemographic characteristics, KAP, and the economic development of their regions. A framework analyzing the determinants of KAP was developed through 3 progressively complex models. These models incorporated a broad spectrum of variables, including age, gender, professional role, educational background, professional title, years of service in health care, hospital tier, gross domestic product (GDP) per capita, geographic region, experience with PAEs, use of informatic tools, participation in departmental discussions on PAEs, and satisfaction with current systems.

The analysis proceeded with univariate assessments using linear regression to evaluate individual impacts, followed by multivariate analyses across the 3 adjusted models. Variables were incrementally incorporated from model 1 through model 3, thereby deepening the investigative scope. Model 1 explored the influence of various independent variables on knowledge levels, designated as the dependent variable. Model 2 extended this examination by evaluating the effects of independent variables on attitudes. Ultimately, model 3 used a comprehensive approach, centering on the willingness to use the CDSS as the dependent variable and integrating all previously considered variables. This structured progression facilitated a multifaceted exploration of how knowledge, attitudes, sociodemographic, and economic factors collectively influence the willingness to practice using CDSSs. The significance level was set at P<.05.

To mitigate potential responder bias and ensure accurate representation [46], we used a weighted average approach based on the population size of anesthesiologists in each region [47], aiming to assess the willingness of anesthesia providers across various provinces to adopt this mobile app–based CDSS, which served as the main outcome indicator. This method addressed the disproportionate impact of regions with varying anesthesiologist populations on the national average. The approach involved calculating a weighting factor for each province’s reported willingness to use this system, based on its contribution to the total national anesthesiologist population. The formula for the weighting factor is as follows:

The weighted value was obtained by multiplying the average reported willingness by the calculated weighting factor, ensuring equitable comparison across provinces. This methodology strengthens the reliability and validity of our findings, mitigating biases from unequal data distribution and providing a more accurate national assessment of adoption intentions.

Based on per capita GDP metrics from the National Bureau of Statistics of China for the year 2022, provinces have been classified into 3 economic tiers: high, medium, and low (Multimedia Appendix 3). In addition, this study included 31 mainland provinces, excluding the Hong Kong Special Administrative Region, the Macao Special Administrative Region, and Taiwan Province due to inconsistencies in statistical standards and data collection methodologies. These provinces were grouped into 6 administrative regions based on economic indicators and geographical locations, with further details available from the National Bureau of Statistics website [48].

Responses were collected in a free-text format, and the data were systematically organized and categorized into key themes based on the frequency of similar comments by 2 independent researchers (XF and PL). Although no formal qualitative analysis was performed, any discrepancies that emerged during the analysis were addressed through discussions with a third researcher (XH).

The study was approved by the Research Ethics Committee of West China Hospital of Sichuan University (2024-132). Participants were invited to complete the anonymous questionnaire accessed via a QR code, following informed consent. A brief introduction outlining the study’s objectives and privacy assurances was provided to respondents. The survey platform ensured that no personally identifiable information was collected, and all data were deidentified to protect confidentiality. Participants were informed that their responses would remain confidential and used only for research purposes. Participation was entirely voluntary, and participants had the right to withdraw at any time. No compensation was offered for participation.

The survey engaged anesthesia providers throughout China, yielding 2440 valid responses (99.3% of 2456 participants), including 2226 anesthesiologists (91.2%) and 214 nurse anesthetists (8.7%). The median age of respondents was 42 years, with a slight male majority (1307/2440, 51.6%) and fewer females (1133/2440, 46.4%). The distribution of gender and age was consistent across regions, regardless of the GDP per capita. Table 1 provides a descriptive overview of the significant variations observed in the demographics and professional characteristics based on the regional GDP per capita. Notably, the attitudes toward PAEs and willingness to use this mobile app–based CDSS among participants varied in accordance with the GDP per capita of their regions (P=.003), while their knowledge levels about this system did not display a similar correlation.

To enhance the robustness of our findings, a sensitivity analysis was conducted that specifically accounted for the perspectives of anesthesia nurses (Multimedia Appendix 5). Regarding the level of knowledge, those holding deputy senior professional titles exhibited greater awareness (coefficient=1.86, P=.04). However, nurse anesthetists used in tier 3 hospitals (coefficient=–1.18, P=.04) and residing in the northwestern region of the country (coefficient=–1.09, P=.04) demonstrated lower knowledge levels. Moreover, those with deputy senior professional titles also displayed more negative attitudes (coefficient=–4.79, P=.02). In contrast, anesthesia nurses who were neutral (coefficient=5.13, P<.001), satisfied (coefficient=2.94, P=.007), and very satisfied (coefficient=1.46, P=.03) with the current reporting system held more positive attitudes toward incidents management, in comparison to those who were dissatisfied. Furthermore, the characteristics associated with a high willingness to adopt among nurse anesthetists included residing in southwest (coefficient=0.43, P=.002) and northwest China (coefficient=0.47, P=.002), satisfaction with the current system (coefficient=0.37, P=.006), and a positive attitude toward PAEs and their management (coefficient=0.05, P<.001). Complete details of the survey responses are available in Multimedia Appendix 6.

This study is the first comprehensive investigation based on KAP theory regarding PAEs among anesthesia providers in China, which identified key factors influencing the adoption of a mobile app–based CDSS in perioperative care. The high response rate underscores the importance of PAE management and a collective willingness to embrace technological innovations for enhancing patient safety. Many respondents showed a solid understanding and positive attitudes toward PAE management, and a strong willingness to use this informatics solution. These findings are consistent with previous studies in various medical fields, highlighting the increasing acceptance and benefits of integrating informatics tools in health care [49,50].

Grounded in the KAP theory, which elucidates the interrelationships among knowledge, attitudes, and practices in health behavior change [31,51], our study offers valuable insights into the factors influencing the intention to use the CDSS for PAE management among anesthesia providers. This theory posits that behavior change occurs in 3 stages: knowledge accumulation, attitude formation, and behavior promotion, and this framework suggests a positive correlation where knowledge enhances attitudes, which in turn promotes practice [31]. However, our findings revealed a notable divergence: while participants exhibited adequate knowledge regarding PAEs, this knowledge did not significantly influence their attitudes; attitude itself emerged as a strong predictor of the intention to use the CDSS. This suggests, that despite sufficient knowledge, it is participants’ attitudes (shaped by perceived usefulness, ease of use, and trust in the technology), that primarily drive their intention to adopt the CDSS. Therefore, addressing attitudinal factors through targeted interventions is paramount. Comprehensive training and institutional support can cultivate positive attitudes toward the CDSS, ultimately enhancing its integration into clinical practice. In addition, the usability and interface design of this app are critical for its adoption; an unintuitive design might hinder user engagement. Other potential barriers include concerns regarding data privacy and security, resistance to change, and insufficient feedback mechanisms for continuous improvement. An organizational culture that does not encourage innovation, along with the financial implications of implementing such tools, can further restrict adoption. Consequently, understanding these multifaceted factors is crucial for developing effective strategies to promote the adoption of a mobile CDSS.

Building on these findings, it’s essential to consider the broader economic context influencing attitudes toward advanced medical technologies like the mobile app–based CDSS. Despite being the world’s second-largest economy, China’s perioperative medical services lag behind those of developed countries, with notable regional disparities in resource allocation [52,53]. Our study highlighted that staff from tier 3 hospitals, often located in wealthier regions with higher per capita GDP, typically possess higher academic credentials and professional ranks. Interestingly, knowledge about PAEs was consistent across diverse economic zones. However, anesthesia providers in affluent areas demonstrated a stronger inclination to adopt the CDSS, reflecting the influence of regional economic differences on technology perception and accessibility. This openness is likely driven by better financial support and prioritization of perioperative informatics infrastructure in these areas. Wealthier regions, with more comprehensive medical resources, tend to place greater emphasis on these aspects [54]. Furthermore, anesthesiologists in southwest China and nurse anesthetists in western China displayed greater openness to adopting CDSS technology. Despite rapid technological growth, these regions still lag in information technology infrastructure and health care resources compared to more developed eastern provinces [55]. These disparities underscore the need for targeted strategies to bridge these gaps. It is essential to increase financial investments in the perioperative health care sector, offer policy incentives, and enhance medical publicity campaigns [56,57]. Such measures aim to reduce the disparities within China and align its perioperative medical services more closely with those in developed countries.

The analysis of gender disparities suggests that female professionals possess a deeper understanding of PAEs, whereas male professionals exhibit a more positive attitude toward these events and a greater inclination toward adopting this mobile app–based CDSS. This variance might be influenced by gender-specific traits, educational backgrounds, and career paths [49]. Literature suggests women often engage in meticulous learning and adopt a cautious clinical approach, while men display confidence and decisiveness, fostering more optimistic attitudes toward the management of PAEs [58]. These insights warrant further exploration into how gender impacts professional development and service provision for PAEs. Emphasizing gender equity and appreciating diverse perspectives in informatics technology development can lead to more inclusive workplace environments and equitable implementation of technological solutions in the PAE sector [19,49]. Regarding work experience and educational background, the study indicates that higher education levels and professional designations correlate with a better understanding of PAEs. Interestingly, the duration of clinical practice does not significantly enhance the knowledge of PAEs, likely due to China’s health care system, which emphasizes both substantial patient care and scientific research capabilities for career advancement [43]. As physicians gain experience, they often become autonomous and confident [59], potentially reducing their focus on PAEs. Departmental management committees should recognize this trend and promote a culture that values continuous education and prioritizes patient safety PAE management.

In clinical practice, approximately half of the respondents experienced PAEs in the past year, yet nearly 50% of these incidents went unreported. This underreporting is exacerbated by the reliance on voluntary reporting and the absence of effective identification tools [60]. Less than 40% of anesthesia professionals are satisfied with the current incident reporting system, citing its complexity, burdensome workload, and time constraints as major issues. Concerns about new information technology solutions include potential workload increases and data security, reflecting apprehensions noted in previous studies [61]. To address these challenges, an innovative informatics solution could leverage the Institute for Healthcare Improvement’s global trigger tool, proven effective in managing ADEs [62]. As outlined in our early protocol [21], this proposed application should feature real-time monitoring, risk prediction, alerts, and enhanced support for reporting and medical decision-making. Barriers to the current reporting framework include legal and financial concerns, fear of negative personal evaluations, and a lack of reporting incentives. Effective implementation of an intelligent decision-support system for PAEs requires fostering an environment of openness, transparency, collaboration, consistent training, and clear legal protections [61,63]. Such a climate is crucial for maximizing the system’s potential in perioperative care [64,65].

In addition, this analysis highlighted differences in perceptions between anesthesiologists and nurse anesthetists regarding PAEs. Anesthesiologists generally possessed a deeper understanding and a more positive attitude toward PAE management and were more inclined to adopt the CDSS for this purpose. In contrast, nurse anesthetists demonstrated higher satisfaction with the current incident reporting system. These variations likely stem from the distinct roles and working conditions of the two professions [25]. Anesthesiologists focus on intraoperative patient monitoring, while nurse anesthetists emphasize patient interaction, vigilance, and proactive care, requiring sustained bedside presence and attentiveness beyond merely monitoring and executing medical orders [66]. This makes nurse anesthetists critical stakeholders in perioperative care and PAE monitoring. However, their heavy workload and focus on surgical patients might lead to less attention on PAE management [27], which could explain the study’s limited capture of their perspectives. To improve this situation, respondents emphasized the need for a user-friendly, efficient, and streamlined informatics tool for managing PAEs. Enhancing the role of nurse anesthetists in PAE management and providing targeted training and enhanced education for both anesthesiologists and nurse anesthetists are recommended. Such initiatives could be facilitated by health care departments to improve PAE management and ensure comprehensive stakeholder involvement.

Our study has several limitations that should be considered in future research. First, the uneven sampling distribution across different provinces led to potential bias, which we attempted to mitigate through regional analysis and result weighting. Future research could address this by using a more balanced geographic sampling strategy. Second, the open web-based distribution of the survey meant we could not calculate an exact response rate. Future studies might benefit from a closed survey system with targeted recruitment to improve sample representativeness. Third, the low participation of nurse anesthetists, possibly due to the slow growth of the anesthetic nursing sector [25], and a national shortage of anesthesia nurses in China [42], limits the generalizability of our findings to this crucial group. Recognizing their potential contribution to PAE management could affirm their role in perioperative care and encourage further engagement. This recognition could initiate a virtuous cycle, where valuing nurse anesthetists leads to increased engagement and professional development, enhanced patient outcomes, and further acknowledgment of their critical importance. Fourth, our respondent pool was primarily from tertiary hospitals, offering limited insights into PAE management in primary care settings. Fifth, despite our efforts to minimize response bias through clear question design, pilot testing, and anonymity, the risk of response and reporting bias remains. In addition, even though the self-developed questionnaire underwent validity and reliability checks, it still carries the potential for reporting bias. Last, as a cross-sectional study, while we identified factors influencing PAE management, we did not provide a detailed management framework. Future research should focus on developing strategies for implementing CDSSs in managing PAEs.

In light of these limitations, future studies should focus on achieving a more balanced sampling distribution, improving the representation of underrepresented groups like anesthesia nurses, and including primary health care institutions. Incorporating a combination of qualitative and quantitative methods can enhance response validation and provide deeper insights into participants’ motivations. Implementing advanced statistical techniques to identify and adjust for biases can enhance the reliability of the findings. In addition, conducting expert focus groups or Delphi surveys could help develop a comprehensive framework for PAE management using CDSSs. Despite these constraints, the survey offers valuable insights for future management of PAEs and the development of related information technologies.

This national study highlights a strong readiness among Chinese anesthesia professionals to adopt mobile CDSSs for PAE management. However, critical barriers, including role-specific knowledge disparities and ineffective organizational communication, must be addressed to ensure successful implementation. Collaborative efforts among local authorities, health care facilities, anesthesia departments, and technology developers are essential to design and implement tailored strategies. Key recommendations include interdisciplinary training programs to enhance nurse anesthetists’ competencies, institution-level incentives to promote incident reporting, and user-centered CDSS designs that prioritize seamless integration into clinical workflows. These measures are vital for improving perioperative incident reporting systems and ultimately advancing the safety and outcomes of surgical patients.