Utilizing health information technology in the treatment and management of patients during the COVID-19 pandemic: lessons from international case study sites

Abstract Objective To develop an in-depth understanding of how hospitals with a long history of health information technology (HIT) use have responded to the COVID-19 pandemic from a HIT perspective. Materials and methods We undertook interviews with 44 healthcare professionals with a background in informatics from six hospitals internationally. Interviews were informed by a topic guide and were conducted via videoconferencing software. Thematic analysis was employed to develop a coding framework and identify emerging themes. Results Three themes and six sub-themes were identified. HITs were employed to manage time and resources during a surge in patient numbers through fast-tracked governance procedures, and the creation of real-time bed capacity tracking within electronic health records. Improving the integration of different hospital systems was identified as important across sites. The use of hard-stop alerts and order sets were perceived as being effective at helping to respond to potential medication shortages and selecting available drug treatments. Utilizing information from multiple data sources to develop alerts facilitated treatment. Finally, the upscaling/optimization of telehealth and remote working capabilities was used to reduce the risk of nosocomial infection within hospitals. Discussion A number of the HIT-related changes implemented at these sites were perceived to have facilitated more effective patient treatment and management of resources. Informaticians generally felt more valued by hospital management as a result. Conclusions Improving integration between data systems, utilizing specialized alerts, and expanding telehealth represent strategies that hospitals should consider when using HIT for delivering hospital care in the context of the COVID-19 pandemic.

identifying approaches to appropriately and effectively use such technology to combat the burden of COVID-19 requires knowledge sharing. Therefore, the aim of this qualitative study was to identify HIT strategies used by digitally mature hospitals in direct response to COVID-19 and make recommendations based on these approaches.

METHODS
This paper forms part of a larger ongoing qualitative study investigating the best approaches to optimizing CPOE based on the insights of key personnel within digitally mature hospitals in the United Kingdom (U.K.), mainland Europe and the United States (U.S.). The methods employed are described in detail elsewhere [25], but the methodology is described briefly below.

Recruitment
Participants were healthcare professionals or informatics specialists drawn from digitally mature sites (defined using the latest HIMSS Electronic Medical Record Adoption Model criteria as hospitals classified as level six or seven) [26] that had been extensively involved in the implementation and optimization of HITs such as EHRs, CPOE, CDS and telehealth. Sites were identified through two distinct strategies: firstly, a scoping review of optimization strategies in electronic prescribing was undertaken [27], with sites featuring prominently in this literature contacted for participation.
Additionally, drawing on a network of experts in the field of electronic prescribing, potential sites were discussed as candidates for recruitment at roundtable events. Following the identification of six digitally mature sites, emails were sent to key contacts (such as pharmacy managers, or heads of research departments) to request permission to contact members of staff within the organisation.
Following this, a purposive sampling strategy was used to recruit key member of staff who had been involved in optimizing HIT within each site, with information and consent forms communicated via email.

Data collection
Semi-structured interviews were used in this study, with interviews conducted by two experienced qualitative researchers (CH and SM). The interviews primarily focused on general approaches to optimizing CPOE. However, as the COVID-19 pandemic arose immediately prior to commencement of data collection, questions regarding each site's response to the pandemic from a HIT standpoint were incorporated into the broader interview topic guide. Specifically, participants were asked whether any specific changes or strategies were put in place within the EHR or related technologies as a direct response to COVID-19. When asking questions relating to telehealth, this was defined as any form of healthcare that was provided remotely to a patient. Follow-up questions were then asked based on the participant's responses to gain further insights. All interviews were conducted via videoconferencing software. All interviews were audio-recorded and transcribed verbatim. The full interview topic guide can be viewed in supplementary file 1, with questions 5, 5.1 and 5.2 relating to this study.

Data analysis
Thematic analysis was undertaken on the dataset using an inductive approach [28], which involved the coding of transcripts to identify themes. In order to develop a coding framework, two researchers (CH and SM) first independently coded two transcripts before discussing any discrepancies between code allocations in the text. The framework was then applied to the remaining transcripts, before the researchers collectively grouped codes into themes and subthemes. In order to reduce the potential effects of researcher biases during data collection and analysis, the researchers employed prospective reflexivity when conducting the interviews and interpreting the findings [29]. Data analysis was conducted using NVivo 12 pro qualitative data analysis software. Further details on data collection and analysis are available elsewhere [25].

Institutional review board approval
Ethical approval for this study was granted by the Usher Research Ethics Group (University of Edinburgh) on 21/01/2020 (ref.1906). Further details of ethical considerations of the study are detailed elsewhere [25].

RESULTS
Forty-four interviews were conducted across six sites (two U.K. and four U.S.-based sites) all of which were teaching hospitals affiliated with academic institutions. A range of healthcare professionals, all with expertise in medical informatics, participated in the interviews. Interviewee roles included pharmacy managers, pharmacists, nurses, physicians, data analysts and chief information officers.
Further details of each site are presented in Table 1.

Managing time and resources
One major theme discussed by all participants was the need to manage an unusually high number of patients, coupled with the additional strain the pandemic placed on resources such as medications, ICU beds, and staff. This necessitated rapid and responsive organizational and technological changes to health systems.

Expediting governance processes
The pandemic was viewed across study sites as impacting on the ordinary governance procedures involved when seeking to change practices or alter HIT systems rapidly. This was viewed as largely positive by those seeking to move existing paper-based practices to digital, especially given the time In the following excerpt, the interviewee is describing changes to the system made in response to a government imperative to decrease the burden of documentation prompted by COVID- 19.
So, what we did was we took a look at the regulations and said, okay, well we can make some changes within our system. Some of it was just that we told people that they didn't have to get the patient to sign it any more. But then we were able to, you know, make some simple changes within our system to designate areas where nurses did need to document. It was apparent across study sites that in addition to streamlining governance procedures, informaticians were also given more autonomy and leveraged to develop HIT solutions to the issues created by COVID-19. For example, hospital management were perceived to have taken more interest in the potential of EHRs and the use of data to facilitate more efficient and targeted care and make changes rapidly. Some participants stated that they felt their work was more valued by management than before the pandemic.

Managing hospital beds
Participants highlighted that general and critical care beds were near capacity during the first wave of the pandemic, and measures were needed to control flow of patients between admission, ICU, and discharge. One such example was the incorporation of discharge assessment into the EHR system, which allowed discharging physicians to complete the necessary patient checklists that would ordinarily be completed separately by infection control; this significantly reduced discharge times and, in turn, increased hospital bed capacity.

Interoperability of systems and integration of data sources
Making all relevant information about a patient easily accessible and viewable to providers was highlighted as important when treating a high number of patients with COVID-19 and other complex morbidities. Examples of configuring provider computer screens to allow this information to be viewed easily were provided in two sites, and the capabilities of integrated EHR systems was exploited to allow for the combining of data from multiple sources such as labs and prescribing. Such capabilities were also extended to allow many routine procedures such as medication reconciliation and patient chart updates to be conducted remotely, often outside the hospital campuses via secure remote-working systems. This was viewed as an important aspect of risk deescalation for providers who would otherwise need to enter wards to conduct such tasks.
Additionally, staff who were required to self-isolate due to a positive close-contact, could still provide support remotely by conducting medication reconciliation or administration from home.

DISCUSSION
We assessed HIT-dependent approaches used to treat patients, manage resources, and mitigate inhospital infection risk during the COVID-19 pandemic within six digitally mature international sites.
All participating sites acted quickly in altering governance and normal procedural practices to allow rapid changes to be made to EHR systems and accompanying technologies. Changes to such organizational factors have been highlighted in the literature as integral to responding to the pandemic [14]. Additionally, making changes to increase interoperability of systems, ease of access to and presentation of relevant patient data, and upscaling of telehealth usage were all adaptations viewed as facilitating more effective practice in the present study. Importantly, the HIT-related responses to COVID-19 were perceived to have further highlighted the value of informaticians for improving care and responding quickly to emergent needs.
The optimization and addition of new alerts were utilized both to improve treatment effectiveness, and to reduce drug shortages experienced due to incorrect prescribing by introducing "hard stop" alerts in combination with decision support and order sets, directing prescribers to more appropriate medications. Similar strategies have been employed to decrease drug shortages prior to the COVID-19 outbreak. For example, Brokenshire and colleagues tested a combination of hard and soft-stop alerts to address opioid shortages [30]. As expected, the study found hard stops to be more effective at reducing opioid ordering with immediate effect, while soft stops led to more subtle decreases in medication ordering over time or had no effect. While the use of hard stop alerts must be carefully considered in routine medical care [31], their ability to abruptly stop inappropriate ordering of scarce medications may be of most benefit during pandemic situations where the strain on resources is unprecedented, as demonstrated by their adoption in this present study.
Another healthcare resource which participating sites used HIT to manage were hospital beds, both through the building of discharge procedures into the EHR and ensuring appropriate integration of systems and data sources to allow bed capacity to be managed in real-time. Both these strategies have also been described in the literature regarding responses to COVID-19 [32]. Additionally, other innovative technological approaches to ensuring patient flow and bed availability are evident in the literature. One study aimed to decrease the burden of occupied beds by fitting wearable devices to 40 patients at discharge to monitor vital signs [21]. While many patients required home oxygen orders, none were re-hospitalised in this study. Such initiatives, coupled with streamlined EHR-based discharge assessment and increased system integration, have helped make available badly needed hospital beds.
In addition, HIT was not only used to treat COVID-19 patients and manage the direct burden on resources, but also to mitigate the risks of infections occurring within hospital. Telehealth was utilized to some extent in all sites, with hospitals that already had some form of patient portal upscaling the use of such technologies. In many places, this was dramatic, with near overnight adoption of telehealth at a scale not previously considered. The potential for telehealth to protect both routine patients and healthcare personnel from infection is highlighted by Russi and colleagues [33]. Additionally, the rapid implementation of a patient self-triage tool within a EHR patient portal was effective at reducing non-urgent triage times, in-person contacts, and had 88% sensitivity for detecting emergency-level patients [34].
The present study resonates with the literature in highlighting the use of mobile devices and screens to limit face-to-face contact between healthcare staff and confirmed COVID-19 patients as part of which allowed specialists to see patients remotely, and removed the need to don PPE which is both time-consuming and costly [22]. However, such technologies must be adequately integrated with the relevant existing systems and patient data sources if an optimized outcome for care is to be ensured.
Furthermore, isolating patients can both be difficult psychologically and may also increase their safety risks [35].
This study has several limitations that should be considered when interpreting the findings. Firstly, we present early qualitative findings without accompanying outcome evaluations of the technological changes described by the participants in this study. It is therefore not possible to draw definite conclusions regarding the effectiveness of these interventions, and formal evaluations should be undertaken to determine efficacy. Secondly, this paper presents findings obtained from interviews that were primarily designed to investigate the wider optimizations of CPOE systems at the participating sites [25] rather than specific optimizations due to COVID-19, meaning the data obtained may not fully capture the full extent of the adaptations undertaken. Despite this, the present study included perspectives from over 40 experts in medical informatics and patient safety within digitally mature sites, who were all directly involved in the technological adaptations described to some extent. Furthermore, while similar studies have previously been published investigating the use of HIT to respond to COVID-19 [15,17,18,21,22,32,34], these studies are largely small-scale, single centre studies. In contrast, this study presents data from six digitally mature sites both in the U.S. and U.K; offering further lessons of the role informatics can play in combating COVID-19 across differing healthcare contexts. The present study also further highlights the value of informaticians, not only in how they can contribute to optimal routine patient care, but also how they can rapidly configure and implement relevant HITs in acute situations such as the COVID-19 pandemic.

CONCLUSIONS
This study presents the views and experiences of key experts in patient safety and medical informatics who have worked to improve their site's respective EHR systems during the COVID-19 pandemic. As might be expected, many systems made substantial and rapid changes to adapt which were facilitated by streamlined governance and the leveraging of informaticians to improve systems.
Key points identified include the utility of existing EHR systems in supporting rapid expansion of telehealth, the importance of interoperability and availability of patient data to the provider, and strategies for conserving resources. Although some of these lessons may already have been learned and adopted within digitally mature hospitals, this paper can indicate potential strategies for less digitally mature settings, and for responses to future pandemics of a similar nature. Our interview data point to the value of a combination of responsive governance, EHRs and accompanying technologies to reduce burden of care, decrease discharge time, drug shortages and health worker exposure. As COVID-19 continues to be at the forefront of global healthcare priorities, more health systems can begin to consider how best to utilize HIT in the fight against the disease.

Conflicts of interest
Dr. Bates reports grants and personal fees from EarlySense, personal fees from CDI Negev, equity from ValeraHealth, equity from Clew, equity from MDClone, personal fees and equity from AESOP, and grants from IBM Watson Health, outside the submitted work. Dr. Sheikh is a member of the Scottish Government Chief Medical Officer's COVID-19 Advisory Group. The remaining authors have no conflicts of interest to declare.

Data availability statement
The data underlying this article will be shared on reasonable request to the corresponding author.