key: cord-1017556-9baof9z8
authors: Yan, Aihua; Zou, Yi; Mirchandani, Dinesh A
title: How hospitals in mainland China responded to the outbreak of COVID-19 using IT-enabled services: an analysis of hospital news webpages
date: 2020-04-20
journal: J Am Med Inform Assoc
DOI: 10.1093/jamia/ocaa064
sha: d33aefd82bc8a1027d1e1b834442416f5ed20280
doc_id: 1017556
cord_uid: 9baof9z8

OBJECTIVE: Many countries have implemented quarantine rules during the global outbreak of COVID-19. Understanding how hospitals can continue providing services in an effective manner under these circumstances is thus important. In this study, we investigate how information technology (IT) helped hospitals in mainland China better respond to the outbreak of the pandemic. MATERIALS AND METHODS: We conducted a content analysis of pages published on the websites of the Top 50 hospitals in mainland China between January 22, 2020 and February 21, 2020. In total, we analyzed 368 pages that the hospitals published during the initial days of the COVID-19 pandemic. The purpose was to identify common themes related to the utilization of IT by these hospitals in response to the pandemic’s outbreak. RESULTS: We identified five focal themes across the webpages published by the hospitals during our study period including (1) popular medical science education, (2) digitalized hospital processes, (3) knowledge management for medical professionals, (4) telemedicine, and (5) new IT initiatives for healthcare services. Our analysis revealed that Chinese hospitals spent greater effort in promoting popular medical science education in the initial stages of our study period and more on telemedicine in the latter stages. DISCUSSION: We propose a configurational approach for hospitals to design response strategies to pandemic outbreaks based on their available resources. CONCLUSION: Our study provides rich insights for hospitals to better utilize their IT resources and some recommendations for policy makers to better support hospitals in the future.

On December 30, 2019, the first case of COVID-19 was confirmed in Wuhan Jinyintan Hospital in Hubei, China. [1] On January 21, 2020, Nanshan Zhong, a well-known respiratory expert in China, announced in the mass media that there was a possibility of limited people-topeople transmission of COVID-19. [2] On January 23, 2020, Wuhan was placed in lockdown at 10am and residents were told to remain in the city unless they had special reasons for leaving. Figure 1 summarizes the key statistics of COVID-19 confirmed cases, suspected cases, and deaths between January 22, 2020 and February 21, 2020, using data from China's National Health Commission (NHC). [3] As illustrated, around 20% of the reported cases were critical and severe patients and around 95% of the mortalities occurred in Hubei province. In response to the pandemic's spread, China's central government launched a national effort aimed at implementing prevention and control mechanisms and cities started to implement quarantines.

Hospitals responded to the outbreak of COVID-19 by creating new rules and procedures for hospital management, redesigning the physical layout of outpatient services, and setting up specific areas for inpatient services for confirmed or suspected cases. They utilized information technology to facilitate these changes. In this paper, we seek to answer two research questions: To address these two questions, we integrated two divergent streams of research, i.e., ITenabled healthcare services and pandemic response actions. Extant literature on IT-enabled healthcare services has primarily focused on adoption and usage of electronic health records (EHRs), [4] [5] [6] [7] clinical decision support systems, [8, 9] mobile healthcare app, [10, 11] web-based platform such as patient portals, [12, 13] and remote healthcare service such as remote patient monitoring. [14] These empirical studies have investigated multi-dimensional factors that may influence the adoption and usage of IT-enabled healthcare services. Another stream of research has focused on the prevention or/and control measures of outbreak of epidemic or pandemic diseases, including the implementation of real-time outbreak and disease surveillance systems, [15] [16] [17] development of algorithms for computerized epidemic early detection systems, [18, 19] and response strategies for the public health system. [20] The response strategies include (1) standardized terminologies for patients to effectively communicate their symptoms and medical histories, (2) threat detection using healthcare data spanning time and location, and (3) response actions such as designing and disseminating guidelines, directing reminders to physicians for worrisome cases, and educating physicians in the application of these guidelines. [20] 

Content analysis was used to code and categorize news webpages published by hospitals in our study. As recommended by Weber [21] , content analysis can be used to identify and describe meaningful classifications, reveal patterns, and therefore gain new insights through a systematic analysis of a volume of written communication. [22] Our content analysis included three phases: preparation, organization, and reporting. [22, 23] In the preparation phase, we collected all the IT-utilization related news webpages published in the Top 50 hospitals in mainland China [24] between January 22, 2020 and February 21, 2020 (Appendix A). Five hospitals did not publish such news webpages during this one-month time window and were excluded. In total, we retrieved 368 news webpages that identified 360 events. The detailed coding procedure is described in Appendix B.

In the organization phase, we adopted an emergent coding approach and followed the steps outlined in Haney et al [25] . First, two authors independently reviewed all the hospital news webpages and came up with a set of codes to form a checklist. Second, they compared notes and reconciled any differences in the initial checklists. Third, a consolidated and mutuallyagreed upon checklist was used to independently code the pages. The inter-rater reliability was excellent using percentage agreement (92.5%) and Cohen's Kappa (0.90). [26] In the reporting phase, we used descriptive statistics, tables, and figures to interpret the common themes of how IT helped Chinese hospitals better respond to the pandemic.

Based on the 360 events identified in the published hospital pages, five themes emerged on how IT was used to respond to COVID-19, including Theme 1 (T1): dissemination of popular medical science education, Theme 2 (T2): digitalized hospital processes, Theme 3 (T3):

knowledge management for medical professionals, Theme 4 (T4): telemedicine, and Theme 5 (T5): new IT initiatives for healthcare services. The events in Theme 4 and Theme 5 were nonrecurring, but some of the events in Themes 1-3 were recurring such as publishing popular science articles. Figure 2 shows the distribution of these five themes from January 22, 2020 to February 21, 2020. From Figure 2 , we can observe the evolutionary patterns among these five themes. At the early stage of the study, hospitals focused more on publishing popular medical science articles on their websites (T1) and also digitalized hospital processes (T2), but in latter stages of the study they focused more on knowledge management for medical professionals (T3) and telemedicine (T4). Table 1 summarizes the coding underlying five themes. We also examined the changing patterns under each theme during this one-month study period and presented our finding in Figure 3 (for themes 1 and 2) and Figure 4 (for themes 3 to 5) respectively.

More than 37% of events were related to dissemination of popular medical science education. Many hospitals used websites to share popular medical education with the public.

We classified these events into four types, see Table 1 . Some hospitals reprinted articles related to COVID-19 from other sources such as WHO or China NHC (i.e. T1-PEAR). For example, the Peking University People's Hospital (ranked #13) reprinted a series of articles related to COVID-19 prevention such as how to wear a mask, how to wash hands, and how to self-isolate at home in February 2020. In contrast, some hospitals invited their medical experts from different departments to write original articles about COVID-19 prevention and control measures (T1-PEAO). A good example comes from Tongji Hospital in Wuhan, Hubei (ranked #8). It invited medical experts from infectious disease, respiratory medicine, neurology, and many other areas to write educational articles to the public. Another type of event is popular science article related to special needs people (T1-PESN), such as chronic disease patients, seniors, pregnant women, and children. Because of the quarantine and isolation policies, many patients with special needs were unable to see their doctors for follow-up visits. Therefore, some hospitals used their websites to provide medical guidelines for this group of patients to take care of themselves at home. For instance, Fuwai Hospital (ranked #23) published a series of articles for patients with heart diseases to deal with some symptoms during the quarantine period. Lastly, as people had to stay at home due to the quarantine and isolation policies, some began to exhibit symptoms of anxiety or stress. Hospitals thus rolled out professional advice on how to stay at home healthily and happily. For instance, on January 30, 2020, West China Hospital of Sichuan University (ranked #2) published an online book for reducing the level of anxiety and shared it freely with the public.

Popular medical education raised awareness in the population and reduced the infection risks. Furthermore, quarantine and isolation policy increased the level of stress and anxiety among people and people with special needs were unable to see doctors. The dissemination of medical information by hospitals helped address these issues.

As shown in Figure 3 , hospitals published more reprinted articles (T1-PEAR) at the early stage, but later added more originally written articles (T1-PEAO). The number of medical guidelines for special needs patients (T1-PESN) also increased significantly during the middle to latter stages. More articles related to psychological issues appeared at the latter stage (T1-PEPI). This is quite understandable because people were likely more anxious after being quarantined for a long term and thus psychological issues related articles were more suitable at that time point.

The second theme relates to the hospitals' digitalized processes to reduce the COVID-19 infection risk. Our findings indicate that 16 hospitals requested all their outpatients to use online methods (e.g. hospital's mobile app, hospital's service account on social media, and official website) to make appointment (T2-OA3). Three hospitals used robots or mobile devices to support their medical processes (T2-RPA). A good example is Guangdong Provincial People's hospital (ranked #27). On January 27, 2020, the hospital welcomed two new robotic employees to help deliver medicines from the pharmacy to the nurse workstation, as well as from the nurse workstation to the isolation ward. These robots are able to take the elevator, open and close doors, avoid obstacles, and recharge themselves. The utilization of these robots reduced close contact between healthcare workers, as well as between healthcare workers and confirmed COVID-19 patients. Later, on February 10, 2020, Guangdong Provincial People's Hospital added three new robots to their medical service team. These three new robots were mainly responsible for guiding patients to appropriate departments and answering questions related to COVID-19 prevention.

Hospitals also adopted IT-enabled services for administration (T2-VO). Online meeting tools were used for conferencing and social media apps such as WeChat were used for creating communication groups among healthcare workers and patients. A wide variety of digitized measures were also used for security control in the hospitals (T2-DSC). For instance, almost all the hospitals asked their visitors to report health status. Some hospitals used paper-based forms, but Ruijin Hospital (ranked #4) used QR code to capture this information. Patients could scan the QR code published on the website to fill in a form on WeChat. Then they could simply present the form to the security staff when they visited the hospital. Many hospitals also used infrared thermometers to detect visitor temperatures without touching the body. Furthermore, in order to save medical protective clothing and reduce the risk of human contact, only one doctor made the rounds of isolation wards and then used WeChat to communicate with other healthcare workers outside the wards. The final event in this theme is systems integration (T2-SI). Hospitals integrated their IT-infrastructure or information systems to better support diagnosis and clinical decision-making during the outbreak of COVID-19. One example comes from Shanghai Ninth People's Hospital (ranked #29). On January 31, 2020, the hospital integrated its laboratory information systems (LIS) to support its fever department. After the LIS integration, all the outpatients with fever symptoms could complete their doctor visits only in the fever department without visiting other departments in the hospital.

From Figure 3 , we can also observe that within the one-month study period, hospitals started with digitalized security controls (T2-DSC), then added more measures in coordinating administrative activities online (T2-VO), followed by emphasizing the usage of online appointments (T2-OA3). Interestingly, we also found that in the initial stages, the majority of hospitals suggested that outpatients use online appointments but not mandatorily (T2-OA1), then they implemented partial online appointments (T2-OA2), and by the end of the study period they had fully implemented online appointments for all the outpatient services (T2-OA3).

Our findings also reveal many events related to knowledge management, including knowledge discovery, knowledge capture, knowledge sharing, and knowledge application. [27] Medical knowledge is primarily tacit and procedural. [28] Thus, the dominant medical education and training method is learning by doing and medical training typically consists of lecture and practice. Medical training is extremely important to reduce the infection risk to healthcare workers. Training topics such as how to wear protective clothing appropriately, how to take care of confirmed or suspected COVID-19 patients, and how to provide professional counselling advices to anxious patients are thus critical. A pattern we observed in our data was that many hospitals recorded training videos and then shared them online internally and externally (T3-OT). By doing so, the duration of onsite training was shorter, reducing the infection risk. Moreover, the onsite training could focus more on the practice of medical skills.

Another major event in this theme was that many highly-ranked hospitals shared their treatment We can observe some evolutionary patterns in this theme in Figure 4 . At the early stage, hospitals paid more attention to launching online consultation services related to COVID-19 (T4-OMC), but then complemented the T4-OMC services with online prescription refills and follow-up consultation services (T4-OMF). Online psychological intervention services appeared in the early to middle stage but became more significant in the latter stage (T4-OMP).

During the outbreak of COVID-19, seven hospitals recognized the significance of IT in healthcare services. They began to make large investments in new IT platforms (T5-ITP) and promoted the concept of "Internet plus healthcare services" (T5-PHI). For instance, West China Hospital of Sichuan University (ranked #2) initiated two projects related to IT platforms on February 10, 2020. These two projects were a platform for precision diagnostics and a big data platform for disease epidemic research. Also, Xiangya Hospital of Central South University (ranked #18) collaborated with a private company to develop a COVID-19 prevention and popular science education system on February 16, 2020. The concept of " Internet plus healthcare services " has been promoted by China's central government since 2018. In our sample, we found that four hospitals promoted this concept on their websites. From Figure 4 , we can tell that the majority of theme 5 events happened in the latter stage of the one-month study window showing a learning curve. [29] 

Our findings revealed five themes related to IT-utilization in hospitals to better respond to the COVID-19 pandemic. Based on these five themes, we have developed an integrated framework for hospitals to better configure their pandemic response strategies using IT-enabled services. As depicted in Figure 5 , first, an effective control and prevention strategy highly relies on the IT-infrastructure. This infrastructure may be constructed and maintained at the hospital level, provincial level, or country level. A hierarchy may exist for these three levels of infrastructure. The more robust the IT-infrastructure is, the better will be the delivery of the ITenabled healthcare services. Second, the response strategy may contain one or more of these four categories of IT-enabled services, i.e., disseminating popular medical science education, knowledge management for healthcare workers, telemedicine, and digitalized hospital processes. Last, the success of the response strategy may be influenced by the hospital's vision and government policy. Government policies such as China's "internet plus healthcare services"

can have a significant impact on the resources that hospitals receive as well as their outlook.

Our study provides some recommendations for policy makers, medical professionals, and hospital managers. First, our study can help policy makers better prepare for similar medical crises. As suggested by our research, IT infrastructure is extremely important for implementing an "internet plus healthcare services" program in mainland China's hospitals. IT-infrastructure at the provincial or divisional level can provide greater benefits compared to isolated hospital IT-infrastructures. For instance, Sichuan province and Henan province have implemented provincial level online collaboration systems among hospitals. As such, online collaboration activities among hospitals in these two provinces was more compared to the others. In the future, provincial or divisional policy makers may consider allocating more resources for IT infrastructure development such as big data platforms or online collaboration systems.

Second, for medical professionals, our study suggests that medical training should place more effort on procedural training and incorporate more IT elements in the training. During the outbreak of COVID-19 disease, medical procedural training has become extremely important. [30] Inappropriate medical procedural taken by doctors or nurses can increase the infection rate. Our findings reveal that IT-enabled training such as flipped-classrooms were suitable during the outbreak of COVID-19. Because of the rushed schedule, many medical professionals who were supporting Wuhan had to attend the training session right after they arrived. In a flipped-classroom training, the hospitals posted online training videos to a platform like WeChat and then used the in-classroom format for practice training. Under a pandemic outbreak, medical professionals are highly motivated to receive appropriate training.

As such, watching training videos online might improve training efficiency.

Third, digitalization of hospital processes is a trend. Our findings confirm that AI technologies can be used for online self-assessment systems, robots can be used in guiding patients and delivering medicines within the hospital, and QR-code can be used for collecting patient and visitor information. All these digitalized processes can also be integrated into daily routines, and not just for preventing and controlling epidemic disease.

This study has some limitations. First, it only focuses on the news webpages on the hospital websites. However, because of the popularity of social media service accounts (e.g., hospital's WeChat service account), hospitals may update their news more frequently on the social media platforms than their official websites. Future research can extract the news from the social media platforms and examine whether there are any differences compared to the present study.

Another limitation is that we did not analyze the sequence of events in this study. Event sequence analysis is an approach to study the unfolding process of response events. [31] It can be used to identify emerging response strategies. Future research can thus investigate common patterns of event sequences in hospitals.

This study aimed to investigate the role of IT in helping hospitals respond to the outbreak of COVID-19. Based on a content analysis of 368 news webpages from 45 top hospitals in mainland China, we identified five common themes where IT was utilized by the hospitals between January 22, 2020 and February 21, 2020. These five themes are: (1) dissemination of popular medical science education, (2) digitalized hospital processes, (3) knowledge management for medical professional, (4) telemedicine, and (5) new IT initiatives. We also developed an integrated framework for hospitals to configure their response strategies to disease outbreaks using IT-enabled services. Our study offers some rich insights for policy makers, medical professionals, and hospital managers to better prepare for other medical crisis in the future.

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

The authors have no competing interests to declare.

AY conceived of and designed the project and was responsible for data acquisition, analysis, interpretation, and writing. YZ contributed to the data acquisition, analysis, and interpretation, and editing. DAM made substantial contribution to the data interpretation, writing and editing.

All the authors made substantial contributions to the revision and gave approval for the final version of manuscript to be published and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. 

Report of the who-china joint mission on coronavirus disease

China focus: China at "crucial stage" to control novel coronavirus

Paperless healthcare: progress and challenges of an IT-enabled healthcare system

The impact of transitioning from availability of outside records within electronic health records to integration of local and outside records within electronic health records

Electronic health records and burnout: Time spent on the electronic health record after hours and message volume associated with exhaustion but not with cynicism among primary care clinicians

Just tell me what ' s going on " : The views of parents of children with genetic conditions regarding the research use of their child ' s electronic health record

Learning curves of agents with diverse skills in information technology-enabled physician referral systems

Measuring implementation feasibility of clinical decision support alerts for clinical practice recommendations

Comparison of traditional versus mobile app self-monitoring of physical activity and dietary intake among overweight adults participating in an mHealth weight loss program

Quantifying usability: an evaluation of a diabetes mHealth system on effectiveness, efficiency, and satisfaction metrics with associated user characteristics

Patient reported barriers to enrolling in a patient portal

Design simplicity influences patient portal use: the role of aesthetic evaluations for technology acceptance

Emotion sharing in remote patient monitoring of patients with chronic kidney disease

Technical description of RODS: a real-time public health surveillance system

Healthmap: global infectious disease monitoring through automated classification and visualization of internet media reports

The EpiCanvas infectious disease weather map: an interactive visual exploration of temporal and spatial correlations

Modeling the effects of epidemics on routinely collected data

WTW-an algorithm for identifying "who transmits to whom " in outbreaks of interhuman transmitted infectious agents

The contributions of biomedical informatics to the fight against bioterrorism

Basic content analysis (No. 49)

Healthcare data scientist qualifications, skills, and job focus: a content analysis of job postings

The qualitative content analysis process

The most comprehensive recommendation! China hospitals ranking list by Fudan University

Drawing on education: using student drawings to promote middle school improvement

An application of hierarchical kappa-type statistics in the assessment of majority agreement among multiple observers

Knowledge management: Systems and processes. Routledge

The acquisition of tacit knowledge in medical education: learning by doing

Bounded rationality and organizational learning

Preparing and responding to 2019 novel coronavirus with simulation and technology-enhanced learning for healthcare professionals

An empirical taxonomy of implementation processes based on sequences of events in information system development