We describe digital exclusion as a complex challenge that consists of 3 interconnected components, with inequalities evident in each component: (1) access to digital connectivity and infrastructure, (2) digital skills and literacy, and (3) engagement with digital platforms. Here we consider each of these elements, illustrated within the context of the key opportunities for digital technologies in the National Health Service (NHS) in the United Kingdom [11], with particular reference to the COVID-19 pandemic.

The digital era has transformed and continues to transform society rapidly, influencing our access to information, employment, working conditions, health and care services, and social connections, and in turn the conditions for good health. However, alongside advancements in digital innovation, digital exclusion is increasingly being recognized as an important factor potentially contributing to both health and social inequalities [44].

The evidence summarized here describes how digital exclusion is a complex factor that reflects underlying social and economic inequalities (such as economic barriers to accessing data and devices or variations in regional internet infrastructure) and contributes to social, economic, and health inequalities (as those less able or likely to engage with digital technology are also most likely to need support).

The digital inverse care law [7,45] reflects the direct impact of digital exclusion on health, for example by reducing an individual’s access to timely and reliable health information, services, and support (eg, from professionals or peer support groups) delivered on digital platforms (Figure 1). A focus on digitally delivered health systems, without due consideration of digital exclusion, has the potential to prevent “intervention generated inequalities” from reinforcing the underlying inequalities in health [46,47]. In a nationwide representative study of engagement with digital technology for health purposes in Wales, the use of digital technology was lower among groups of people who are likely to have greater health needs, including older people, those living in less affluent areas, those with poorer underlying health, and those reporting health-harming behaviors (ie, smoking, drinking, and physical inactivity) [48]. This has also been highlighted within the context of COVID-19, a period when having an internet presence is crucial to rapidly access not only health information but also digital health consultations and health monitoring apps. Furthermore, people who are at the highest risk of poor health outcomes are also those most likely to be digitally excluded, including older people and those living in more deprived areas [49].

Digital exclusion can also have an indirect impact on health outcomes, acting through the other social determinants of health [44]. For example, digital exclusion may reduce an individual’s access to social support services (eg, income support), education, and employment opportunities, all of which are recognized as underlying social and economic determinants of health (Figure 1). Within the context of the COVID-19 pandemic, during the first 2 weeks of the nationwide lockdown in the United Kingdom, approximately 1,000,000 people sought financial support through universal credit; however, this was only possible on the internet, thus posing a challenge to those who are digitally excluded. When schools, crèches, universities, offices, churches, shops, restaurants, and parks are closed owing to the COVID-19 pandemic, the usual institutions around which people structure their everyday lives and gain support were no longer physically accessible. Many people transitioned to web-based platforms, which was a supportive community inaccessible to digitally excluded people [50].

Digital exclusion is concurrent with Dahlgren and Whitehead’s definition of a social determinant of health [51], as a social and economic factor with the potential to increase or decrease social inequities in health. Recognition of digital exclusion as a social determinant of health by governments and systems would increase the visibility of the issue, highlight the need for routine measurement, report variations across population groups, and focus action on the factors contributing to digital exclusion across policy, health, and social systems. There are examples where health systems display different behaviors in light of increased awareness of digital exclusion, such as addressing data poverty by enabling mobile phone users toll-free access to the NHS and national government health websites [52]. Digitally excluded people would have been hugely impacted without access to rapidly changing and reliable public health advice during the COVID-19 pandemic.

In response to the COVID-19 pandemic, contact tracing digital apps were rapidly implemented in several countries [53] including the United Kingdom [54]. Engagement with the app is dependent on the individual owning a smartphone, having the skills to understand and download the app, and keeping their phone switched on with Bluetooth enabled. However, 6,500,000 adults in the United Kingdom cannot turn on a device, and 5,900,000 adults cannot open an app [28]. A survey carried out by IPSOS Mori for the Health Foundation [55] revealed a clear digital divide by age, occupation, and educational level in public readiness to download the NHS COVID-19 app. Public trust and confidence were also likely to be highly instrumental in adoption. One’s motivation to download and use the NHS COVID-19 app would reflect one’s views on risk and benefit within the context of a pandemic and, potentially influenced by the views of one’s family, social networks, and professionals.

Contact tracing apps are likely to be disproportionately taken up by younger, more affluent, and tech savvy populations. The benefit of the app extends to users and people around them; however, if digital exclusion prevents the most vulnerable people from participating in digital technologies, then the collective societal benefits are not equitable. This highlights the importance of digital innovation being accompanied by nondigital approaches, which are efficient and effective for people who cannot or choose not to engage with digital solutions, and to evaluate the outcome across population groups [56].

There is sufficient evidence to conclude that digital exclusion should be considered across sectors as a social determinant of health, with the potential to exacerbate health inequalities if progress does not consider the structural, economic, social, and behavioral factors contributing to digital exclusion.

The COVID-19 pandemic has clearly highlighted the levels of inequality in the ownership of digital devices and demonstrated that the access to fixed and mobile internet connections is a gateway to essential health information and care and many other key services including education, food delivery, employment, and social support, all of which indirectly impact health. As described here, the underlying inequalities in the access to digital technology and the internet are evident, and if left unresolved, have the potential to exacerbate health and social inequalities in a digital era.

In order to progress towards digital inclusion, one policy approach adopted by certain countries is to declare that access to the internet is a human right [57], whereas others consider technology and the internet as an enabler of rights rather than a right in itself [58]. Declaration of access to the internet as a right may enforce action at national and international levels to address a factor that indirectly contributes to health, placing it at par with other rights including the access to food and safe housing, all of which contribute to health equality.

This study shows that the underlying drivers of digital exclusion are complex, and as such the solution cannot be obtained through a single action or from a single organization. Addressing digital exclusion would require government, public, private, and nonprofit sector organizations to collaborate to ensure that progress toward a digital future does not inadvertently leave some people behind. This will require a comprehensive understanding of the factors contributing to digital exclusion in local populations and how these factors differ among groups, areas, and over time, along with effective engagement with digitally excluded people to coproduce solutions that adequately address the contributing factors such as poverty, access, skills, or motivation.

For example, a common viewpoint is that addressing barriers to internet connectivity will address digital exclusion; however, this is not the case. Inequalities would continue to persist after access is resolved if the complex underlying factors that contribute to the digital divide are not addressed. a crucial first step is to understand the extent and drivers of digital exclusion within and between different populations through data and qualitative insights. Followed by a multidisciplinary and multisectoral response to address structural barriers (such as the lack of infrastructure to support adequate internet access in all areas), financial barriers (eg, costs of devices and data plans), digital skills and literacy, and other barriers to engagement (eg, cultural, concerns regarding trust and data privacy) (Table 1). Embedded evaluation alongside digital innovation is essential to ascertain the impact of each of these actions against key outcomes including differences in uptake, engagement and effectiveness across population groups. For progress to occur, there is a need to bring together actions against these key components in a comprehensive and coordinated plan; for example, the New York City Internet Masterplan [59] is a comprehensive roadmap to close the digital divide through enhanced infrastructure, affordability, and inclusion. To ensure collective progress, it may also be necessary to ensure that the government takes the responsibility for digital equity.

Organizations in the United Kingdom have brought together practical calls to action to mitigate the risk of digital inequalities. The Good Things Foundation has issued a blueprint for a 100% digitally included United Kingdom for a post–COVID-19 economy, focusing on 3 key steps to fix the digital divide: the need to address the digital infrastructure, data poverty, and to develop an inclusive digital strategy [60]. The Carnegie UK Trust has reported 12 recommendations for policymakers, practitioners, academics, and industry professionals to learn from the lockdown and eliminate digital exclusion [61]. In Wales, the Older People’s Commissioner has urged investment in digital inclusion as one of the preconditions for an age-friendly recovery [62]. With an international focus, particularly on low-income countries, the Human Rights, Big Data and Technology Project has postulated 5 urgent principles for leaving nobody behind through technology in the COVID-19 response [63]: guaranteeing internet access as a human right and a public good, increasing the availability and acceptability of the digital infrastructure, increasing the accessibility and affordability of digital services, empowering people by addressing disinformation and hate speech without censorship, and ensuring that internet access is not a cause for more surveillance.

As health and care systems seek to deliver services through digital platforms, it remains important to monitor and evaluate the levels of engagement to ensure that a focus on digital approaches does not inadvertently reinforce underlying inequalities in health. Within the United Kingdom, the NHS provides a potential opportunity to monitor and address different levels of engagement with digital technology to help understand differences by clinical needs and population groups, an understanding of which could be shared to inform innovation elsewhere.

Based on theoretical frameworks and the lessons learned from the practical examples described here, the actions needed to ensure that digital health innovation helps address, rather than exacerbate, health inequalities are as follows:

Digital technology has the potential to revolutionize health and health care with growing interest among policymakers, researchers, and practitioners in exploring how digital technology can be harnessed to improve population health. Countries worldwide have been using digital technologies to respond to the COVID-19 pandemic, for applications including communication of public information, remote delivery of health care, and population surveillance [71]. Greater adoption of technology can also provide additional sources of valuable data for potential use by population health systems to understand health needs more clearly, monitor outcomes, and counter the factors underlying ill health.

Digital innovations should be accessible to everyone, empowering citizens to become active contributors to health and well-being. However, those likely to have the greatest health needs are also least likely to have access to digital platforms and the skills to use and navigate them, and they are hence less likely to engage with such digital platforms. This raises an important question of equity in population health in a digital era, an issue brought into sharp focus by the COVID-19 pandemic. This led us to ask how we can innovate through digital technology and transform population health, while leaving no one behind.

To address this question, we need to better understand who engages with digital technologies, the enablers and barriers, and the direct and indirect impact on health outcomes. The lack of consideration of these factors poses the danger that the pursuit of digital health solutions results in unintended consequences and reinforces existing social and health inequalities. Action is needed across government, public, private, and nonprofit sector organizations to ensure capitalization on the potential for digital technology to address health and minimize the risk of exacerbation of health inequalities.