Summary of the process
The work was implemented through virtual reality-based simulation of animated bedside and ward training scenarios for critical health workers. The training covered the handling of virtual reality equipment, donning and doffing of personal protective equipment (PPE), case management of people infected with COVID-19, and hand hygiene, while ensuring safety. against infections and improving access to training.
VR hardware and software
The VR training sessions were conducted at IDI’s ACE Viz Lab facilities. Each participant was equipped with an HTC Vive Pro head-mounted display (HMD) and two controllers. The system includes multiple wall-mounted base stations that emit safe infrared light that is detected by sensors on the HMD and controllers to provide highly accurate room-scale tracking. Each PC has a high-performance GPU to ensure an optimal high-resolution VR experience for the user.
Immersive content for the different phases was provided by: (1) Enduvo (Enduvo, Inc. Peoria, IL, US), a virtual reality training platform for the introductory phase with content created by the authors (https ://my.enduvo.com/course/6VuYTzTXY.); (2) Humulo (Humulo, Inc., Edgewater, MD, USA), a VR training platform for interactive donning and doffing training, using a custom module created by Humulo with requirements provided by the authors ; (3) SOMA, a platform for creating and consuming VR content in the form of 3600 videos, with content created by the authors.
Adaptation of the COVID 19 IPC curriculum and preparation of the platform
The COVID-19 IPC Virtual Reality Course and Curriculum has been developed using the Department of Health’s updated COVID-19 IPC Classroom/In-Person Course as a guide . Essential topics and skills were extracted and used to design a personalized preliminary curriculum map with particular attention to skills amenable to VR renderings. The program map was then enhanced with additional feasible simulations and interpretations, such as that of the COVID-19 virus particle and its anatomy. The resulting program map covered five modules; Introduction to VR equipment, Introduction to SARS-CoV-2 virus, Infection prevention and control, Disinfection and waste management, and COVID-19 case management (Table 1). The updated IPC training materials corresponding to each module of the program were then transformed into the VR space as artifacts through a stage of designing, coding and launching artifacts. Here, each artifact representing entities such as people, PPE gowns, disinfectants, beds, and other objects in the training environment was designed and implemented separately using VR software. Each artefact was then supplemented with required activities by adding dynamic movements and gesticulations. Narratives were then synchronously added to the dynamic artifacts through voice-over narrations and by adding text captions. The aggregation and alignment of these three channels of information: artifacts, activities and audio narrations, was made possible by the Enduvo Features of the VR software platform and therefore integrated into the VR platform. This content was then organized and arranged into different virtual reality modules or interactive scenes, which represent steps and practice protocols of IPC and PPE.
Using the developed VR content, pre-pilot training of expert staff from the Institute of Infectious Diseases (IDI) and MoH IPC was conducted to obtain pre-course feedback and evaluation. This helped refine the course before it was rolled out to clinicians in the field. It was conducted on 10 participants over a period of one week. Insights from this exercise included acknowledgment of the need to continually review and update content to align with current COVID-19 IPC best practices as WHO and CDC continued to update update the recommended preventive measures during the pandemic, integrate safety measures when using shared VR equipment and improve the quality of 3600 videos. To address these shortcomings, we reviewed and updated content, purchased fluid-resistant head nets, eye masks, and alcohol hand sanitizer for each VR station. We also redid some of the 3600 videos to improve quality.
After incorporating solutions to deficiencies raised at the pre-pilot level, the exercise was then rolled out to clinicians in the field as a background pilot training. The field clinicians were a multidisciplinary team including doctors, nurses, laboratory technicians, clinicians, pharmacists and epidemiologists. The pilot field clinician training took place in two phases over a two-week period with a ratio of 1 VR instructor for every two participants per day. The first phase consisted of the theory of the COVID-19 IPC course in the immersive Enduvo VR platform for a duration of one week and phase two included practical training using 3600 videos to complete the first phase – for a period of one week. The successful implementation of the second phases (Fig. 1) set the stage for a more fully immersive third phase of virtual reality.
The introductory phase of this course served to orient participants to VR platforms and equipment and used a didactic approach to cover the theoretical principles of IPC COVID-19. Each group of participants received a demonstration of the navigation process of the Enduvo VR platform showing how to load sessions, navigate between sessions, pause, do assignments and submit assignments. This phase was delivered by course instructors, who provided instructional and troubleshooting support to participants in the event of device or procedural issues during the course.
Once participants were comfortable with navigating and manipulating the platforms, they covered the didactic introductory sessions in the VR platform to orient them to the various theoretical foundations of COVID-19 IPC. This content is publicly available in the Enduvo app at https://my.enduvo.com/course/6VuYTzTXY and can be used with or without a VR headset. It consisted of eleven short, goal-driven topics, each with five minutes of running time. However, actual subject duration was variable as it was self-paced, with individual participants having the ability to pause, forward, rewind, and browse the embedded VR artifacts (Table 2) and videos for their own clarity. At the end of each objective session, a multiple-choice assessment was undertaken to assess knowledge acquisition. This phase lasted half a day for each trainee.
During this phase, participants covered practical aspects of COVID-19 IPC and case management. This was done through demonstrations of the procedures using 3600 videos in combination with a full-immersion virtual reality prototype for hands-on experience (Table 2). The 3600 the videos helped participants view pre-recorded IPC and case management scenarios in an enhanced 3D experience that gives them better, near-real engagement with the scenario (Supplementary File 1).
This phase involved health workers exploring the 3600 VR videos multiple times in the VR lab until a required level of confidence in performing the task is achieved. At the end of each session, an objective assessment in the form of a knowledge-based multiple-choice question was undertaken to assess knowledge acquisition. Finally, participants were treated to a demonstration of a full-immersion virtual reality prototype that guided them through the actual practice of IPC activities. This was implemented through a collaboratively developed virtual reality prototype in which each IPC procedure is repeated until the desired skill level is achieved. It included an immersive, dynamic and interactive VR environment that alerts participants whenever they take a wrong step with an audio cue and does not let them continue until they have perfected the procedure (Supplementary File 2) . This phase lasted half a day for each trainee. The lab is actively procuring equipment to support the full interactive delivery of this phase.
The phases were delivered as group sessions with a maximum of six participants each to adhere to social distancing standards. There were 10 groups of six such participants, with two groups attending the course each day. Each cohort participated in phase 1 for half a day during the first week and in phase 2 for half a day during the second week. All sessions were self-paced and each trainee was given the opportunity to iteratively work through each session until they understood the content. A VR technical assistant was present to guide participants through their sessions.
Our goal was to train 60 frontline health workers working in IDI-supported health care facilities in Kampala and Wakiso districts. Sample size and cohort were determined by the six training stations available within the laboratory, all of which are more than four meters apart. Since each group of six participants needed half a day of training, the lab could only handle 12 participants per day. As a result, in 5 days of the working week, the lab could handle a maximum of 60 participants with proper social distancing. Pairing content with 3D web technology and internet connectivity would scale training by orders of magnitude, especially in cases where potential participants have appropriate devices such as smart phones to receive 3D web content.
This virtual reality training was evaluated using three main approaches. In the first approach, learning outcomes in terms of participants’ knowledge acquisition and retention were assessed by comparing similar outcomes in previous cohorts trained using the classroom teaching model. The second approach was an experiential evaluation designed to assess participants’ experience, particularly because it is a new educational and technological approach that many were experiencing for the first time. In this evaluation, an individual online survey form was used to evaluate the course. The purpose of this survey was to obtain participant feedback on the overall implementation of the course, including strengths, weaknesses and recommendations for the various components in order to inform subsequent course improvements. In the survey, we used a Likert scale questionnaire with a number of choices ranging from strongly agree to strongly disagree. The survey consisted of a series of eight multiple-choice questions and four open-ended questions that allowed participants to express their views in detail.