The study was conducted among the members of the Chen Xiaoping Academician Workstation of Hepatobiliary and Pancreatic Surgery. In China, there is a hospital rating system to evaluate the public hospital by 3 levels and 6 classes. Medical services and management, medical quality and safety, and medical technology level and efficiency are the 3 main assessment targets [20]. Tertiary class-A is the highest level of the comprehensive hospitals according to the “3 levels and 6 classes” assessment rules, meaning that it should have a high level of hospital information construction. Hence, this study included all tertiary class-A public hospitals among all members of the academician workstation. The questionnaires used in study were distributed by email in April 2020 and were collected in July 2020. A total of 24 questionnaires were distributed and 23 were collected. Among the 23 tertiary class-A public hospitals who provided feedback with the questionnaires, 3 were from East China, 5 were from Midland China, 3 were from South China, 6 were from West China, and 6 were from North China. The respondent at each hospital was the department director, vice dean, or dean.

This study was conducted using a self-designed questionnaire. The questionnaire was referred to in a previous study [21] and discussed, revised, and finalized by study groups to make it more suitable for this study. The questionnaire included questions around 4 latitudes: configuration of hospital information department, configuration of hospital information system, configuration and practical application of hospital internet service, and the satisfaction of hospital information construction. Furthermore, interviews were also conducted with the respondents in this study for requirements analysis.

This study was approved by the ethics committee of Tongji Hospital (TJ-IRB20220103). There were no active participants in the study data, and the ethics committee waived the requirement for informed consent. All methods in this study were carried out in accordance with relevant guidelines and regulations.

A database for the information obtained from the collected questionnaires was constructed using EpiData software (version 3.1; EpiData Association), and the data from each questionnaire were recorded twice by 2 staff members working together to ensure that the data were recorded without any typing errors. Data were presented by frequencies and percentages, and analyses were performed using SPSS Statistics software (version 21; IBM Corp).

Driven by the national policies, information construction has become one of the priorities of hospitals’ daily work. In our research, all hospitals had set up independent information department, which were almost all led by a senior leader of hospitals. However, only 70% hospitals in our study had built an IT team to construct and maintain the information projects at the hospital. Others had outsourced to other companies.

All hospitals were equipped with an EMR system in our study. This result is not surprising, since EMR is a very important evaluation index in China’s hospital ranking system. Besides, the utility rates of HIS and PACS were also very high (over 95%). The utility rates of LIS and RIS were a little lower than abovementioned systems, but they all exceeded 90%. Overall, the infrastructural construction of hospital information was relatively good.

In the aspect of hospital internet service, web-based appointment registration and EMR service were the most widely used. Other internet services such as external promotion and introduction, telework and portal application, web-based medical consultation, and business-to-business transaction were gradually developed and applied. In addition, we also found that hospitals in provincial capitals had a variety of internet services with higher utility rates.

Notwithstanding, the infrastructural construction of hospital information was satisfactory. However, the hospitals’ staff were not satisfied with the data utility. Half of them in our study thought that the huge scientific and economic value hidden in the massive data stored in hospital information systems had barely been mined.

In China, the distribution and allocation of medical resources are incredibly unbalanced: almost 80% of medical resources and patients are concentrated in large hospitals and 20% are in primary care institutions [23]. There is a similar trend in the construction of hospital information. Developed regions, first-tier cities, and top-tier hospitals are sitting on high-quality medical resources and have an early start and large investment in information technology construction, with enough funds and human resources [24]. This is a common problem in transitional countries. For over 20 years, the failures of medical informatization in transitional countries were mainly due to the infrastructural gaps [25]. According to the investigation results of the present study, the overall information system of large hospitals was relatively complete, and the hospital information systems, such as HIS, LIS, EMR, PACS, and RIS, were also well equipped. Based on the interviews, the next steps of information construction in large hospitals are in relation to the following aspects: higher-rated EMR system, widely interconnected information platform, more convenient and intelligent internet services, translating big medical data into valuable applications with cutting-edge technologies, and so on. In contrast, primary care institutions have limited resources in this area, and their level of investment and construction of information technology development is far lower than that of large hospitals. Noticeably, the gap between top-tier hospitals and primary care institutions will hinder the data interaction and limit the insights and extensive application of big medical data.

Medical information interconnection is a process of fusing multisource data into big data, which have great scientific and commercial value. However, the results of the present study showed that over 65% of the medical workers believed that the abundant medical data in their system was not utilized well. At present, most hospital information system still stay at computerizing the paperwork and data integration. In our interviews, most of the IT staff in the information department focused on the maintenance of the information system. The optimization of workflow is most focused on the quality control of EMR. Data mining is predominantly reserved for administrative management. However, with technology changing especially rapidly, the high application values of big data are being more and more realized. Medical scientists have foreseen the potential value of big medical data stored in hospital information systems [26]. Unfortunately, current hospital information systems cannot support this use for scientific research effectively, and medical scientists are always absent from hospital information construction. This is a common problem in the construction of medical informatization, which will seriously restrict further development of medical informatization [27].

In fact, the construction of medical information technology involves a wide range of aspects. The talent involved in the process of information technology construction needs to have a hospital-related medical background, be familiar with the hospital workflow, and have a theoretical knowledge on information technology [28]. Having this type of complex talent and clinical worker that coparticipate in the hospital information technology construction will certainly be able to broaden the insights into the information contained behind the massive data deposited in the hospital information system. This will further improve the use of medical data and build a synergistic system of “information plus” and clinical care, personnel training, scientific research, and the transformation of experimental findings, so that the hospital information construction meets the actual needs of clinical research. Moreover, this will meet the information data support for administrative decision-making and empower the improvement of medical service quality [29]. Several research had revealed the high willingness of medical students to receive medical informatics training in their study period (90% for American, 70% for German, 83% for South Korean, and 78% for British students) [30-32]. Therefore, we should actively train a group of composite talents with both medical and information backgrounds and encourage clinical workers to actively participate in the hospital information construction work through various forms, such as title evaluation, achievement recognition, and performance rewards.

The other very prominent problem in overall hospital information construction is the lack of Uniform Data Standards. There is an international series standard, which has been in development since the 1980s—Health Level 7 (HL7) Fast Healthcare Interoperability Resources (FHIR) standard—that was released in 2019 [33]. The HL7 FHIR standard includes a unified description and understanding of different data to facilitate rapid exchange and understanding of data from different sources and in different system. It can be said that HL7 FHIR is a facilitator of information exchange between different systems. However, FHIR is not widely used in China at present. Only a few hospitals as a pilot hospital have built relevant systems and platforms by FHIR [34]. In the early stage of information technology construction, almost no top-level planning and overall coordination were observed, and different manufacturers in the development and production of information technology systems used different standards in China. At present, most information platforms have been built according to the needs of each medical institution. The conceptual system and data type of medical institutions are different, and the same data may be recorded by multiple heterogeneous systems at the same time, which often leads to conflict and redundancy in the process of information interaction. The system data model used in medical units is not open, so data exchange and sharing cannot be realized, which affects the extraction and use of data and challenges the information exchange between systems in the medical field. The poor compatibility between the different information technology systems caused many difficulties in the completion of information interoperability with other hospitals or institutions.

The specialist medical information concept is being born at the right moment. In the process of specialist medical information construction, another level of data standard inconsistency is present, that is, the data recording standards are not uniform among the different medical institutions. Although the EMR can now enter part of the objective data in tabular form, most of the information related to disease development and treatment is still recorded using large texts, such as first course, daily note, surgical records, and so on, in the EMR. In retrospective studies, adequate human resources are required to extract valid data from medical records, and missing information in the extracted data is inevitable. In multicenter retrospective studies, different centers may have missing data due to different medical record writing habits. Moreover, the missing data will cause a lot of problems and greatly reduce the validity of the research results. The construction of information technology can save the process of manual extraction of information, that is, the poststructuring of textual information through natural language processing technology [35], but the lack of information cannot be compensated if the doctor does not write the information into the EMR. Thus, in the process of building medical information technology, according to the type of standard for data collection, what information should be recorded in the EMR? This is the most important point of the problem that must be solved, and these standards have not been established yet.

At present, the EMR is mostly a general EMR that was upgraded from the perspective of hospital-wide application. In recent years, the concept of specialized EMR has gradually emerged [36,37]. The specialized EMR is an integrated system of medicine, management, research, and education based on the general EMR and combined with specialized diagnosis and treatment process, so as to better serve the specialized diagnosis and treatment behavior and improve patient satisfaction. For the construction of specialized EMR, it is necessary to establish a standard data set that combines the characteristics of specialized medical treatment, and under the guidance of this standard data set, the general EMR can be renovated and upgraded in a targeted manner. The construction of the specialized medical data standards is not only for creating specialized EMR services but also for laying a solid data foundation for future data mining and specialized information application. With the rapid development and common application of big data mining technology in recent years, the application of medical big data, such as clinical auxiliary decision-making, clinical research and development, and medical intelligence, has already had a more mature technical environment, and what is lacking now is high-quality medical big data. Therefore, in the process of hospital information construction, the problems related to the construction of specialized medical data standards need to be addressed.

A limitation in our study is the small sample size, which may lead to being underpowered for inferential analyses. However, our findings indeed exposed existing problems in medical information construction. The interviews with the responsible persons involved in our study also provided valuable insight to improve the existing hospital medical information systems. Accordingly, we believe our findings are objective, and it calls for further study in larger samples.

Hospital medical information construction has always been a challenging issue in transitional countries. In China, although the infrastructural information construction of medical of tertiary class-A public hospitals has been completed, various problems are still present, hindering the development of hospital medical information construction. The resolution to improve the overall level of hospital information construction in transitional countries still requires long-term endeavors. Support for the construction of information technology in primary care institutions should be increased to balance the level of development of information construction in medical institutions at all levels; the training of complex talent with both IT technology and medical backgrounds should be emphasized; and specialized disease information standards should be developed to lay a solid data foundation for data use and improve the use of medical big data.