Virtual ward platform technologies for acute respiratory infections

3.1 There is an increased demand on NHS services for respiratory conditions. The NHS has set up a number of ARI hubs and ARI virtual wards to relieve pressure on other parts of the local healthcare system. Virtual wards allow people to be cared for in their home or usual place of residence while also reducing the pressure on hospital inpatient care. Virtual wards could also potentially reduce pressure on other aspects of the care system, including primary care appointments and emergency hospital attendance. Clinical experts highlighted that although remote care has existed for a long time, the digitisation of virtual wards supports scalability.

3.2 The committee acknowledged that the admission criteria for a virtual ward for ARI should be based on NHS England's guidance on ARI virtual wards. This NHS England guidance includes the consideration of symptom severity, such as National Early Warning Score 2 (NEWS2) scores and oxygen saturation, as well as clinical trajectory and comorbidities. Clinical experts stated that population creep (such as people having treatment who would not usually need hospital care) could be a potential problem with virtual wards. The EAG highlighted population creep as a key area of uncertainty for the economic analysis. They reported that population creep can reduce the potential cost savings with virtual wards. Clinical experts stated this problem is likely to be limited if local services develop clear admission and discharge criteria using NHS England's guidance.

3.3 Virtual ward platform technologies comprise 3 key parts: a patient-facing app or website, associated medical devices and a digital platform for healthcare professionals. The committee acknowledged that in addition to the core functions of a virtual ward platform, the technologies offer a range of additional features. These include variation in the monitoring devices offered, such as spot monitoring devices and continuous monitoring using wearable devices. The companies also offer a range of additional services including the delivery and maintenance of the associated medical devices and in-house healthcare professionals to support implementation and operation of virtual ward platforms.

3.4 The committee agreed that key aspects of a virtual ward are a user-centred design, the ability to allow multidisciplinary team working and interoperability with electronic patient records. Associated medical devices also need appropriate regulatory approval in addition to meeting local testing standards and must be validated for use in a home or place of residence. The platform also needs to provide risk-stratified alerts to healthcare professionals for when readings go outside of the agreed range. These alerts can be based on the person's device-measured vital signs or responses to questionnaires they submit themselves. When using wearable devices for continuous monitoring, it was acknowledged that alerts should be trend-based to prevent over-notifying, which could otherwise increase the overall time healthcare professionals spend reviewing and responding to alerts. Because of the limitations in the clinical evidence, the committee acknowledged that there is no evidence to show whether one platform is better than another. The EAG noted that the variation in cost and resource use relating to the features offered by virtual ward platform technologies is a key area of uncertainty in the economic model.

3.5 Virtual ward platform technologies for ARIs are available in some NHS sites. Expansion of these wards is needed to support the increasing demand on hospital beds for treating ARIs. Clinical experts said that good communication with the community and local hospitals would help support implementation and awareness of virtual wards. They also stated that multidisciplinary teams and patient involvement are needed during the setup of a virtual ward.

3.6 Clinical experts and companies stated that a key barrier to implementation was interoperability of the virtual ward platforms with electronic patient records. They highlighted the importance of information being accessible to the people who need it, including the multidisciplinary team running the wards and any out-of-hours services providing support. Companies said that the platforms have software to access NHS systems such as electronic patient records, but funding availability to cover the cost of that connection is the main barrier. Electronic patient record suppliers would also need to work with the companies to support the interoperability. The committee concluded that interoperability with electronic patient records systems and associated medical devices would be a key feature needed in a virtual ward platform technology.

3.7 Clinical experts stated that training for staff was essential for implementing virtual ward platforms. Staff must be trained on the different features of the platforms and how to train patients to use the technologies (if training is delivered by NHS staff). Companies state that there are different training options, including online or physical face-to-face sessions, videos and user manuals. Technical support should also be made available to patients and their carers and healthcare professionals.

3.8 Virtual ward platform technologies can increase treatment options available to people with an ARI that needs hospital-level care. A patient expert said that being able to have care at home allows people to have their home comforts and freedom, and to interact with family, friends and pets. They also stated that you can get undisturbed rest and more easily get fresh air at home compared with a hospital. Clinical and patient experts also highlighted that being at home reduces the risk of getting a hospital-acquired infection. It also reduces the risk of deconditioning because the person can move about and exercise more easily.

3.9 People admitted to a virtual ward or their carers need to have clear information and support to use the technology and manage their condition. They should also be given time to ask any questions or express any concerns about using the technology before being admitted to a virtual ward. A patient expert noted a concern that people may not know what to do if part of the technology, such as the oxygen supply, stops working. Clinical experts agreed that people need to feel empowered to reach out for help and escalate care if needed.

3.10 Admittance to a virtual ward should involve shared decision making with patients and their carers. Patients and their carers need the confidence, motivation and skills to be able to use a virtual ward platform technology and the associated medical devices. Family support may be needed to help people who are being monitored at home. Carer burden should also be considered. Although some people may feel reassured by monitoring their readings, others may find this increases anxiety. Suitability of the person's usual place of residence should also be considered. For example, does the residence have a fixed or mobile telephone line, running water, heating, electricity and have access to meals. People with no fixed address, no privacy, or with a lack of physical space may find it hard to use a virtual ward. Other considerations may also be needed such as a person's level of frailty or whether they live alone, or have any cognitive impairment or learning disability that may make a virtual ward less suitable. The committee concluded that a range of factors need to be considered when offering a virtual ward. Any treatment offered should be patient centred and specific to their condition.

3.11 Virtual ward platform technologies are delivered through a patient-facing app on a smart device. People on a virtual ward need regular access to a device with internet access to use the technologies. Additional support and resources may be needed for people who are unfamiliar with digital technologies or who do not have access to smart devices or the internet. Companies state that people who do not have access can be provided with smart devices and mobile internet access for the duration of their virtual ward stay. Additional considerations in relation to connectivity would be needed in areas where there is limited internet access. Companies should use simplified patient interfaces to make it easier for people who are not familiar with using digital technologies.

3.12 Additional support and resources may also be needed for people with visual or hearing impairments, problems with manual dexterity, or who are unable to read or understand English. The companies said that they have taken steps to improve the accessibility of their technologies. This includes providing tablets or monitoring devices with large screens and buttons for people with visual impairments or problems with manual dexterity. Screen-reading software can also be offered to people with visual impairments. Some companies also offer translation services or provide the app in multiple languages for people with English as a second language. Simple user instructions in multiple languages should be made available.

3.13 The committee recognised that some pulse oximetry devices have been reported to overestimate oxygen saturation levels in people with black and brown skin. This may lead to their condition not being treated when treatment is needed. Clinical experts stated that limitations should be recognised for any test and that a person's condition would be treated using a range of outcome measures rather than relying on 1 parameter. Pulse oximetry devices should be tested and validated for use in people with black and brown skin. Companies noted that some devices, such as those that continuously monitor respiratory rate, are not affected by skin colour.

3.14 The EAG prioritised 19 studies, including 6,129 people. These included 2 randomised controlled trials (RCTs), 1 prospective cohort study, 11 prospective case series (including 2 cohort studies extracted as case series) and 5 retrospective case series. The prospective cohort study was done in the UK and compared a technology-enabled virtual ward with a telephone-based virtual ward or historical hospital control. But, the RCTs were done outside the UK, so the evidence considered was limited by a lack of UK-based comparative studies. The EAG also noted that the virtual ward admission criteria in the studies were unclear and 16 of the prioritised studies were limited to only people with COVID‑19.

3.15 The evidence reported on length of stay, admissions, readmissions and escalation of care. The results suggested similar outcomes to inpatient care. However, this evidence was limited because of the lack of comparative studies and heterogeneity in outcome reporting. None of the evidence suggested that virtual wards were unsafe and reported mortality rates were low. The evidence showed that there was a high level of adherence to use of the technology as well as high acceptability for both healthcare professionals and people admitted to virtual wards. Studies reported barriers that included digital literacy, technical issues or inadequate demonstration or explanation of the technology, language barriers and digital exclusion.

3.16 Clinical evidence on 8 in-scope technologies (Clinitouch, Inhealthcare, Doccla, Luscii, Whzan Blue Box, Virtual Ward Technologies, Huma, Current Health) was identified in 15 of the studies. One additional technology identified during consultation (Masimo SafetyNet) has 1 UK study on a chronic obstructive pulmonary disease virtual ward and 1 US study on remote monitoring of people with COVID-19. The UK study is limited by uncertainty in the relevance of the included population. There is no evidence comparing the different platform technologies with each other. The EAG concluded that the evidence could not distinguish whether different features of virtual ward platforms affected clinical or service outcomes. See the assessment report in the project documents for this guidance for more details.

3.17 A simple cost-comparison model showed that virtual ward platform technologies are potentially cost saving to the NHS. The technologies were cost saving by an estimated £872 per person compared with inpatient care, and by £115 per person compared with care at home without a technology-enabled virtual ward. This cost model was platform agnostic and so the range of savings could vary depending on the technology considered. The base-case savings were supported by sensitivity and scenario analysis results. The economic model used a 30‑day time horizon and included the costs associated with a virtual ward, such as:

3.18 The EAG acknowledged that because of limitations in the available evidence, the early economic model used simplifying assumptions. This meant there was some uncertainty in the extent of the cost saving. These assumptions included considering a virtual ward and its comparators as being equally effective and that training, implementation and treatment costs per person were directly scalable to any virtual ward size. An assumption around staffing of a virtual ward was also used due to the known variation in set up of these services in the NHS. The EAG noted that the main cost drivers of a virtual ward include the length of stay, hospital admission rates, number of alerts and the cost of the platform.

3.19 The committee acknowledged that the variation in features offered by virtual ward platform technologies could lead to differences in costs and resource use. The EAG noted that the incremental cost of a given platform will most likely be impacted by any additional features offered by the platforms (such as company-provided support for monitoring), the effectiveness of continuous monitoring, interoperability with other NHS systems and ease of use of the virtual ward platform. The committee concluded that there was enough evidence to recommend the virtual ward platform technologies while further evidence is generated. Evidence on measures of clinical effectiveness, service outcomes and resource use are needed to reduce uncertainty in the economic modelling.

3.20 This evaluation focused on ARI. The EAG concluded that no included evidence directly addressed transferability of virtual ward platform technologies from other settings to an ARI setting, or vice versa. The committee acknowledged that virtual ward platforms would be purchased for use in multiple populations and that usability across populations would need to be considered when deciding which platform to use.

3.21 The committee concluded that there was enough evidence of a potential benefit of virtual ward platform technologies for them to be used in the NHS while further evidence is generated. The main evidence gaps for these technologies are: