Comparison of Computer-based Information Support to Clinical Research in Chinese and Japanese Hospitals: A Postal Survey of Clinicians Views

Transcription

1 Schattauer GmbH Comparison of Computer-based Information Support to Clinical Research in Chinese and Japanese Hospitals: A Postal Survey of Clinicians Views G. Jiang, K. Ogasawara, A. Endoh, T. Sakurai Department of Medical Informatics, Hokkaido University Hospital, Japan Summary Objectives: The objectives of this research are to examine the current situation of computer-based information support of clinical research in hospitals and to determine the expectations of clinicians toward clinical research support functions of hospital information systems (HISs) in both China and Japan. Methods: 172 clinicians from 42 major hospitals in China (2 groups), and 568 clinicians from 79 university hospitals in Japan (2 groups), were surveyed by postal questionnaire during July and August, The Kruskal-Wallis test was performed to analyze the differences among the groups. Results: The total response rate was 66.9%. The result shows that 94.8% of the Japanese clinicians, 3.5 times more than those in China, use computers almost every day. High significance was shown for the frequency of non-his based information resources used by clinicians between China and Japan (p <0.001), whereas no significance for the frequency of HIS use by clinicians between the China I and Japan I groups (p = 0.725) was found. 33.3% clinicians in China thought they could obtain 30-50% of the necessary patient data for clinical research from the HIS, about 2 times more than in Japan (p = 0.009). Conclusions: Although the degree of computer involvement among clinicians in Japan is much higher than in China, the computer-based hospital information systems have not been developed well for supporting clinical research in both countries. The clinicians expect comprehensive computerized patient records (CPRs) and full use of patient related information in the existing HISs to support their clinical research. Keywords Hospital information systems, information storage and retrieval, clinical medicine, research, questionnaires Methods Inf Med 2002; 41: Introduction As computer systems become ubiquitous and easier to use in the clinic, hospital and at home, the potential to access information is enormous [1]. In terms of clinical research, computerized literature retrieval systems are available, at modest costs, on computer networks (including the Internet) and on CD-ROM [2]. It is also becoming common to access patient-related data through computer-based hospital information systems. Moreover, the telecommunication capabilities of clinicians in hospitals have increased remarkably [3]. Under these circumstances, it is still not clear to what extent computer-based information meets clinical research information needs in both China and Japan, in which the information infrastructure and the development of hospital information systems are on different levels of advancement. In this paper, we describe our investigation of the current situation of computerbased information support to clinical research in hospitals and the clinicians expectations of clinical research support functions of HISs in both China and Japan. To do so, we have used a self-made postal questionnaire. By comparing the two countries, we provide a base for discussing methods to improve the support functions of Chinese and Japanese HISs. 2. Methods The questionnaire was designed according to the following two aims: 1) To what extent have computers been involved in clinical research in hospitals with -installed HISs? 2) What expectations do clinicians have of HISs in regard to clinical research? The questions involved the frequency of computer use, computer-based literature retrieval, Internet-based information collection, communication among researchers, and HIS-based patient data collection (Table 1). The postal survey was conducted during July and August Four groups of clinicians from both China and Japan were selected for the survey. 120 clinicians were from Lanzhou Jincheng hospital in China (China I), 52 clinicians from 40 major hospitals in China (China II). 100 clinicians were from Hokkaido University Hospital, Japan (Japan I), and 468 clinicians from 78 university hospitals in Japan (Japan II). In the China II group, the clinicians were selected from a clinical research group supported by a Sasakawa medical scholarship, 63% of them from university hospitals. The clinicians of the other three groups were selected randomly. All clinicians selected for the survey have more than 5 years of work experience in a clinical setting. The Kruskal-Wallis test was used to analyze the statistical significance among the groups using the Excel Statcel statistical software. For each information resource, 5 kinds of the tests were performed. First, the Kruskal-Wallis test was performed among the 4 groups of clinicians. If the results were significant, tests comparing the subgroups (China I vs. Japan I, China II vs. Japan II, China I vs. China II, Japan I vs. Japan II) were performed. Received November 30, 2000 Accepted July 17, 2001

2 142 Jiang et al. Table 1 The list of questions the clinicians were asked in the questionnaire. used to communicate their clinical research. This percentage is about 11% in Japan.About 82% of the clinicians in Japan used computer-based literatures databases to retrieve medical literatures at least 1 to 2 times per week. This percentage is about 19% in China. Figure 1 shows the frequency of computer, Internet, and computer-based literature database use by clinicians at the level of at least 1-2 times per week in China and Japan. Table 2 shows the p values of the Kruskal-Wallis tests ( = 0.05) for comparing the use of different information resources among different groups. High significance was shown for the frequency of non-his based information resources used by clinicians between China and Japan (p <0.001). 3. Results Valid responses were obtained from 105 clinicians of the China I group (87.5%), 46 for the China II group (88.5%), 61 from the Japan I group (61%) and 283 from the Japan II group (60.5%). The total response rate was 66.9%. 3.1 Non-HIS-based Information Collection In order to determine the degree of computer literacy of clinicians in both countries, the clinicians were questioned about the frequency with which they use computers. The results show that 94.8% of the Japanese clinicians, about 3.5 times more than in China, use computers almost every day. In addition, 24.2% of the clinicians in China have never used a computer. In Japan this percentage is less than 1%. More than 80% of the Japanese clinicians use the Internet at least1-2 times per month for clinical research, this percentage drops to 26.4% in China. Moreover, 47.8% of the Chinese clinicians have never used the Internet for clinical research. This percentage is less than 10% in Japan. About 75% of the clinicians in Japan use to communicate at least 1-2 times per month. This percentage is about 11% in China. In addition, about 74% of the clinicians in China have never Fig. 1 Comparison of the frequency of computer, Internet, and computer-based literature database use by clinicians at least 1-2 times per week between China and Japan (p <0.001, Kruskal-Wallis test). 3.2 HIS based Data Collection Frequency and Usefulness of HIS Use The survey of the frequency HIS usage showed that only 5.9% of the Chinese clinicians and 15.5% of the Japanese clinicians used HIS frequently to conduct clinical research. 54.1% of the clinicians in both countries rarely or have never used HIS to retrieve needed patient data for their clinical research (Fig. 2). Table 2 shows that although the significance was shown by performing the Kruskal-Wallis test on all 4 groups (p = 0.033), no significance was found for the frequency of HIS use by clinicians between China I and Japan I groups (p = 0.725). In addition, the question what is the maximum percentage of related patient data available for your clinical research by using computer-based hospital information system in your hospital? was answered by clinicians. The survey showed that only less than 10% of the clinicians in both countries thought that they could obtain more than 50% of the relevant data for their clinical research (Fig. 3) in this way. 33.3% of the clinicians in China thought that they could get 30-50% of the relevant patient data for clinical research from HIS. This is about 2 times higher than in Japan (about 14.5%).

3 143 Computer-based Information Support in China and Japan expectation to be able to access the current patient data stored in HIS (34 respondents). The third is the expectation to be able to exchange patient-related information among hospitals (12 respondents). The fourth is the expectation to give clinical research a relatively high priority, based on the principal of protecting the privacy of patients (10 respondents). Fig. 2 Comparison of the frequency of HIS-based data collection by clinicians between China and Japan (p = <0.05, Kruskal-Wallis test). Table 2 shows significance was identified for the usefulness of HIS by using the Kruskal-Wallis test on all 4 groups (P = 0.015). Also, the significance was determined for the usefulness of HIS as perceived by the clinicians between the China I and Japan I groups (p = 0.009), although no significance was found between the China II and Japan II groups (p = 0.197) Opinions of HIS The next question inquired into the major problems arising during extraction of patient data from HIS for clinical research. The survey showed that about 50% of the clinicians in both countries thought that the lack of necessary patient data in HIS was the main problem for their clinical research. Approximately 25% of the clinicians thought the lack of a support function for accessing patient data stored in HIS was the main problem (Fig. 4). In addition, clinicians opinions were obtained from a free-text question: What is your opinion about the kind of help you really expect to get from computer or hospital information systems for your clinical research? 128 of 344 (37.2%) respondents in Japan and 36 of 151 (23.8%) in China answered the question. According to the answers, the expectations of clinicians in both countries can be summarized into the following categories. The first category is the expectation to computerized patient records (CPRs), which include comprehensive and long-term patient-related data (39 respondents). The second category is the Table 2 The p values of the Kruskal-Wallis tests ( = 0.05) used to compare the use of different information resources among different sample groups. 4. Discussion 4.1 Clinical Research and Computer-based Support Functions By comparing clinicians views between the two countries, some basic problems are revealed regarding the methods required to meet the clinicians information needs for clinical research by using current clinical information systems. The survey showed that there was a high frequency of computer use by clinicians in Japan. It indicated that the computer has penetrated into the daily work of clinicians in Japan, probably related with the widespread introduction of order-entry types of HISs in Japanese university hospitals. In contrast, only 26% of the clinicians use computers daily in China. It can be inferred that the interaction between computer systems and clinicians in China was not frequent and close. Hospital information systems (HISs) are computerized information banks that deal with patient-related data, typically including demographics, medical information (such as history, diagnosis, laboratory findings) and financial information [4]. While many HISs can manage large amounts of clinical patient data, few have built-in capabilities for supporting clinical research [5]. There are some complaints from clinicians that the current HISs cannot meet their information needs for clinical research. For instance, some HISs lack research support functions, while some systems limit researchers to accessing databases directly to extract the data they need [6, 7]. Lack of common standards to describe clinical data is also a problem in collecting clinical research data [8].

4 144 Jiang et al. Fig. 3 Fig. 4 Comparison of the usefulness of HIS by clinicians between China and Japan (p = <0.05, Kruskal-Wallis test). Comparison of the opinions about the problems of HIS from clinicians between China and Japan. With the clinicians increasing information needs for their clinical practice and research, the functions of HISs have expanded constantly. Connecting with non-patient related information resources have become an important support function in integrated HISs, especially with the rapid involvement of Internet/WWW technologies in hospitals. The support function has great potential for improving the capability to extract related information for clinical research. Electronic information retrieval (IR) systems have been accepted well and are being used frequently by clinicians for clinical research in Japan. The survey showed the significant difference in frequency of Internet use, literature databases use and communication for clinical research by clinicians between China and Japan. The main reason is probably due to a large difference in the necessary hardware and software available for supporting retrieval systems in hospitals between the two countries. We noticed that 62% of the clinicians from the China I group have never used the Internet for clinical research, and 26% of the clinicians in this group have never used computer-based literature retrieval databases. In fact, the Internet connection environment was not established in the hospital, but the literature retrieval services were provided by CD-ROM retrieval systems, such as MEDLINE and Chinese Biomedical Literature Databases. This is probably the reason of the lowest rate of use for clinical research by clinicians in the China I group. Therefore, the way the electronic IR systems were used is different between the two countries. Japan is more on-line oriented in comparison to China. It is conceivable that the situation may change with the improvement of the Internet connection environment in hospitals in China. Although the degree of computer involvement of clinicians in Japan is much higher than in China, no significant difference was found on HIS-based patient data collection for their clinical research between China I and Japan I group. It may be due either to the lack of required data in HISs, or the lack of functionality to support acquiring related data from HISs. The following questions in the survey proved the former explanation. Only less than 10% clinicians in both countries thought that they could get more than 50% of the relevant data from HISs for their clinical research. Moreover, about 50% clinicians in both countries thought that the lack of related patient data in HISs was the major problem for their clinical research. The results indicated the current HISs in both countries could not meet the information needs of clinicians for their clinical research. It seems the problem originated from current HISs in both countries. In Japan, the successful introduction of order entry HISs in the early 1990s marked the beginning of clinical information management in hospitals. In 1998, approximately 90% of university hospitals had installed order entry systems [9]. In China, with the implementation of the Golden Health Network Project by the Ministry of Health in 1995, many hospitals have begun introducing network-based HISs with client-server distributed structures [10]. In Japan, advantages brought on by the introduction of order entry systems have been proven to be effective in shortening patient waiting time and improving management [11]. Meanwhile, disadvantages have also been found by some studies. Kaihara [9] pointed out that important items of medical records, history of present illness, physical examination, progress notes, surgical operation notes, outcome, etc, are still missing from this database in order entry systems. Hirose [12] thought that many of Japan s current patient-centered order-entry systems lack certain desirable database characteristics or functions and fail to integrate patient data/information across subsystems. In China, network based HISs are just being introduced into hospitals, replac-

5 145 Computer-based Information Support in China and Japan ing the monolithic version of HISs in recent years. However, it is pointed out that the current HISs in China still concentrate on administrative and financial management [13]. As a result, it can be concluded that current HISs in both countries were not developed for supporting clinical research. However, based on the current situation of HISs in China, it is difficult to explain why the clinicians in China I group felt the HISs more useful for their clinical research than those in the Japan I group. A possible explanation can be given according to the experiences of the author from China: Some services for supporting patient record retrieval were available in hospitals in China. For example, the hospital corresponding to China I group provided ICD-9 based retrieval service and clinicians could retrieve the index numbers of paper-based patient records that interested clinicians. Thus, a seemingly insignificant service made a relevant good response. It implies that it is very helpful to establish some support functions in HISs that make accessing patient-data easier for clinicians. Clinicians expectations mentioned in the survey provided some clues for seeking solutions for improving current HISs to meet the information needs of clinicians and to support their clinical research. A long-term solution is probably to establish a computerized patient record (CPR) system, which includes comprehensive patient record data and a support module for clinical research with a user-friendly interface. A practical plan has been drafted in Japan to change the order-entry systems to the CPR within the next 6 years, in spite of many difficulties [12]. In China, the research on CPR and its related standardization have also been attached much importance. In addition, the survey showed that clinicians eagerly expected to be able to retrieve the patient data stored in the existing HISs for their clinical research. Therefore,endeavor should be taken in improving and enhancing clinical research support functions in existing HISs. A medical information and retrieval system was designed to support clinical research by establishing an independent database mirroring virtually all the information held in the HIS and providing flexible retrieval and analysis of data contained in the database [5]. A Webbased disease tracing system was established by scanning HIS for the integration of clinical research within clinical practice and for the coordination of a multi-center trial [14]. 4.2 Methodological Issues A questionnaire survey is a blunt instrument and a good method for identifying clinicians information needs [15]. The method is suitable particularly for identifying clinicians information needs on clinical research, because it is difficult to use direct observation methods. Much of the early work was conducted by measuring the physicians information needs for their clinical practice [16, 17] rather than for clinical research. A questionnaire must be reliable: the random error of response must be minimized so that consistency of measurement is achieved. The questionnaire must also be valid: it must be a true measure of what it intends to measure and must not be subject to bias [18, 19]. In this survey, we focused on clinical research support functions provided by computer systems in Chinese and Japanese hospital settings. Considering the reliability and validity of the questionnaire, we reviewed previous studies of physicians information needs for their clinical practice and research and focused only on the usual computer, Internet, , computer-based literature database, and HIS use to develop the questionnaire. The response rate of the survey was much higher than a previous questionnaire survey on HIS in Japan [11]. It indicated that most of clinicians were interested in the questions included in the self-made questionnaire survey. All clinicians selected were from major hospitals, most of them from university hospitals. Therefore, we believe that the views of clinicians between China and Japan were comparable. We mainly focused on the current situation of computer-based support of clinical research by clinicians in both countries. The questions were designed to be rather general. All of the clinicians in Japan selected for the survey were from university hospitals, but only 65% were from university hospitals in China. Therefore, it seemed that the clinicians in Japan selected in the survey were more research-oriented. In addition, the number of clinicians in the China II group was relatively small in comparison to the Japan II group. It is possible that the Chinese clinicians were not represented comprehensively. 5. Summary Non-patient related data collection and patient related data collection are the two most important parts of the information needs of clinicians for their clinical research.the computerized literature retrieval systems (online or CD-ROM) have been widely used and well accepted by clinicians for their clinical research, whereas the computer-based hospital information system has not been well developed to support clinical research in both China and Japan. Acknowledgments The survey was partly supported by the Japan- China Sasakawa Medical Fellowship.The authors are very grateful for the cooperation of all clinicians involved in the survey in both China and Japan. References 1. Hubbs PR, Rindfleisch TC, Godin P, et al. Medical information on the Internet. JAMA 1998; 280: 1363 (editorials). 2. Hersh WR, Hickam DH. How well do physicians use electronic information retrieval systems? A framework for investigation and systematic review. JAMA 1998; 280: Lindberg DAB, Humphreys BL. The promise: a time of change for medical informatics in USA. Yearbook of Medical Informatics IMIA. Stuttgart-New York; Schattauer 1999: Satya-Murti S. Hospital information systems: The physician-computer connection: a practical guide to physician involvement in hospital information system. JAMA 1993: 270 (7): Dorda W, Wrba Th, Duftschmid G, et al. ArchiMed: A medical information and retrieval system. Methods Inf Med 1999; 38: Sata H, Hori M. Medical information system in near future: requests from physicians. The proceedings of the 18th joint conference on medical informatics in Japan, 1998: 15. (In Japanese). 7. Sakon M, Monden M. Problems in hospital information system and its future: the viewpoints from surgeons. The proceedings of the 18th

6 146 Jiang et al. joint conference on medical informatics in Japan, 1998: 18. (In Japanese). 8. Kaihara S. The promise of medical informatics in Asia. Yearbook of Medical Informatics IMIA. Stuttgart-New York: Schattauer 1999: Kaihara S. Realisation of the computerised patient record: relevance and unsolved problems. Med Inf 1998; 49: Yang MJ, Cao L, Lin F, et al. Rational thoughts on the development of medical informatics in China for new century. Proceedings of the first China-Japan-Korea joint symposium on medical informatics, Academy press, 1999: Beijing. 11. Haruki Y, Ogushi Y. Okada Y, et al. Status and perspective of hospital information system in Japan. Methods Inf Med 1999; 38: Ball MJ, Peterson H, Douglas JV. The computerized patient record:a global view. MD Computing 1999: Wang JZ. Recent trends and challenges of HIS in China: The growth potential using M and Cache. Proceedings of the first China-Japan- Korea joint symposium on medical informatics, Academy press 1999: Beijing. 14. de Groen PC, Barry JA, Schaller WJ. Applying World Wide Web technology to the study of patients with rare diseases. Ann Intern Med 1998; 129: Smith R. What clinical information do doctors need? BMJ 1996: 313 (7074): Covell DG, Uman GC, Manning PR. Information needs in office practice: are they being met? Ann Intern Med 1985; 103: Williamson JW, German PS, Weiss R, et al. Health science information management and continuing education of physicians. A survey of US primary care practitioners and their opinion leaders. Ann Intern Med 1989; 110: Mckinley RK, Manku-Scott T, Hastings AM, et al. Reliability and validity of a new measure of patient satisfaction with out of hours primary medical care in the united kingdom: development of a patient questionnaire. BMJ 1997; 314: Friedman CP, Wyatt JC. Evaluation methods in medical informatics. Springer-Verlag Press, Correspondence to: Guoqian Jiang Department of Medical Informatics Hokkaido University Hospital West 5, North 14, North District Sapporo, , Japan guoqian@med.hokudai.ac.jp