Keywords

1 Introduction

As with many other blanket terms, digital health has many definitions. In its broadest sense it refers to the use of information and communications technologies in medicine and other health professions to manage illnesses and health risks and to promote wellness (Ronquillo et al. 2017). This is a very wide scope and thus it is inevitable that digital health will have many different facets, and consequently will be ‘re-defined’ closer to the viewpoints of different authors or different disciplines within which it is used, such as health informatics. Digital health is not a new concept in healthcare and some of its earliest uses refer to the adoption of computerised methodologies for the systematic organisation of patient files (Krishnan and Neuss 2022). However, even in this limited context of electronic hospital records (EHRs), implementation was slow and highly fragmented, because the collection of data for public health records (and later on for surveillance) was affected by a historic lack of investment in digital technologies (Vandenberg et al. 2018). In the beginning of the twenty-first century the notion of digital health changed fundamentally as the advent of the internet and improved capabilities for international data transfer meant that digital health was viewed both as a local as well as an international need, with an increasing emphasis on the harmonisation and interoperability of extant systems. The more recent experiences from the outbreaks of infectious diseases (Ebola virus, Zika virus, and currently SARS-CoV-2 virus) have added yet another dimension to digital health: the responsiveness to public health emergencies inclusive of the end user and wider public (Knawy et al. 2020).

Indeed, as technology develops and its integration into routine clinical practice increases, it can be anticipated that consequently the plurality of digital health definitions will also increase, reflecting on the ever-widening user base. At the same time, it is important for a more consolidated taxonomy to be developed on digital health, consolidating digital health concepts/spheres of influence, for use in research, policy development, public health and clinical practice. To this end, there have been some recent, excellent attempts in shaping and understanding this existing ambiguity (Värri 2020; Iyawa et al. 2016), and these would need to be maintained in the future. At the very heart of this ambiguity in digital health definitions lie the many different needs, where it is anticipated that digitalization will improve health services across the board. The World Health Organization (WHO) shares this view, and considers digital health to be linked to the general support of introducing and emboldening universal healthcare systems globally. To this end, the key objectives of digital health for the WHO are: (i) the translation of existing and produced datasets into action, contributing to decision-making; (ii) the use of digital technologies to enhance connectivity and information transfer, including remote communication activities and (iii) the systematic assessment of national and regional needs in relation to emerging new technologies, including the support of technological co-development (World Health Organization 2021).

Therefore, reflecting on the many different needs and definitions, neither this chapter or this book, take a singular view on digital health. Instead, as the book includes over 80 authors from more than 20 different countries, each chapter contains a working definition for digital health so as to facilitate the authors in developing their arguments fully. While this entails the inherent danger of potentially conflicting views, the editors have considered that to be a desirable outcome. Indeed, the current lack of a bona fide definition for digital health indicates that there are, and will be, conflicting approaches to such a definition and as such it serves the wider scientific community and interest for those differences of opinion to remain visible.

The overall approach in constructing the current chapter is that of a narrative review, in which the most recent policies, guidelines, and publications from the last 5 years have been identified through a literature search, with further additional manuscripts being identified through ‘snowballing’, i.e., using reverse citation tracking to find articles that cited others already deemed relevant to the review (Callahan 2014). This narrative review approach was chosen because the aim of this chapter is to provide a broad perspective and explore the general debates and developments. A systematic review approach has been used in other chapters because they focus on unique and specific queries using explicit methodologies (Rother 2007).

2 Differences and Commonalities Between Needs in HICs and LMICs

One of the most pronounced differences in digital health viewpoints is the one that exists between High-Income Countries (HICs) and Low-and-Middle Income Countries (LMICs). This book provides a detailed example from both sides: a HIC implementation and success story is shared by authors from the Kingdom of Saudi Arabia and by Poland, where the former provides an example of a nearly completely integrated healthcare system, while the latter describes the significant strides and achievements in getting to this stage. They should be viewed as akin to maturity models (Wendler 2012) highlighting a path of evolution for systematic development and improvement. What is common for these two mature examples is that they are using digital health as the engine for growth of the healthcare reach and efficiency and are achieving healthcare services as part of routine practice that would have been difficult to imagine only a decade ago. Thus, the view of digital health is both in terms of individual services as well as of the wider healthcare system. By contrast, while there is a good theoretical understanding of the wider digitalization benefits in healthcare as a whole in many LMICs, most of the examples represent individual institutional success stories, indicative of what can be further achieved, though not representing a healthcare system developed in its entirety.

At this point it should be noted that digitalization need not follow the exact same path of emergence in HICs and LMICs. The implementation of digital healthcare is highly context-driven (Gjestsen et al. 2017), needing to adapt to the user-base, available infrastructure, and political and regulatory frameworks. As such, LMICs should be expected to develop distinctly different approaches in the digitalization of their healthcare (Rossman et al. 2021; Surka et al. 2014). For example, in HICs, digitalization is dominated by the discussion on improving tertiary healthcare capacity (Al-Kahtani et al. 2022) and/or the consolidation of existing services (Vandenberg et al. 2020), while in most LMICs, tertiary healthcare is available to only a small section of the population. In most LMICs, healthcare is provided by primary healthcare centres, e.g., in Nigeria this proportion is as high as 85% of the population (Ugo et al. 2016), as such the path to digitalization would be predictably different. Furthermore, in HICs digitalization in healthcare is often linked with insurance providers (Posselt and Kuhlmann 2020), with the potential to add value in existing systems. However, in LMICs a large portion of employment is in the informal sector; for example, in some areas of Kenya this can amount to over 80% of the labour force (Were 2020). Thus, the market forces that will define the context for healthcare digitalization are distinctly different, and this would inevitably be reflected in the implementation pathways. The next section will investigate some of those trends in digital health and innovation with a focus on LMICs, thus setting the background for the subsequent chapters that will provide a deeper insight into specific contexts across the world.

3 Trends in Digital Health, Trends in Digital Innovations

Ubiquity of digital environments: Digitalization has been increasing across all fields of human activity and the introduction of many new technologies have led to the ‘datafication’ of many routine public health/governmental activities, including in healthcare, albeit with different intensities (Redden 2018). The increased presence of digital environments, as well as increasing experience in implementing digitization programs, are likely to ensure a continued growth for digitalization in healthcare in HICs and LMICs alike. At the same time, digitalization has resulted in an increasing overlap between the production and consumption of data, e.g., patients are increasingly both creators and consumers of healthcare data, highlighting the need for greater accountability on data use, as well as a greater oversight of the ‘translation’ process, i.e., how this data is feeding into ongoing routine activities and decision making (Agostino et al. 2022).

Fragmentation and lack of uniformity: The possibility of data users to add their own comments and perspectives in healthcare or even physiological measurements via wearable devices has been viewed as a way of improving existing services (Minniti et al. 2016) if implemented well. However, this opening up of interaction possibilities has not always been as successful as originally anticipated. For example, in many sub-Saharan African locations, such an interaction was limited to existing vertical programs, e.g., HIV surveillance, and required substantial educational input for staff to be effectively implemented (Kwame and Petrucka 2020). Additionally, the increase of data production, means that a strict hierarchical data structure may not be adequate, as more horizontal platforms may be required, e.g., platforms where patients and doctors would interact directly. This inevitably can lead to the (limited) decentralization of information within healthcare systems, where local larger healthcare units would be intermediate holders of information/platform hosts, and then the data would be integrated at a higher level of structural hierarchy.

However, the multiple sources of data production, mean that there is a high heterogeneity inherent to the healthcare data, rendering the collected big data less informative using current conventional technologies (Dash et al. 2019). Thus, the anticipation is that eventually the data processing will be performed closer to the data producers, utilising distributed technologies, and that a greater data harmonization will be implemented by the necessity of data interoperability as well as the implementation of predictive machine learning technologies (Dash et al. 2019). The need for data harmonization is not new (Liu et al. 2010; Zisis 2016), however the ever-increasing digitalization maintains this need at the forefront. The implementation of Health Level 7 (HL7) Fast Healthcare Interoperability Resource (FHIR) standard (Braunstein 2019) has led to some uniformity allowing partial interoperability (Edoh 2020), although implementation in LMICs remains highly fragmented and requiring the adoption of LMIC-adapted implementation frameworks (Hussein et al. 2023; He et al. 2023).

Lack of regulations, lack of incentives: The complexity and heterogeneity of healthcare data, as described above, generates expected questions regarding the data regulation. In HICs, well-defined data regulatory frameworks have been adopted, e.g., the General Data Protection Regulation (GDPR) in the European Union (Voigt and Von dem Bussche 2017) that applies across the entire data processing workflow, from data generation, to storage, dissemination and use. Within LMICs, the development of data protection regulations remains an ongoing process (Akintola and Akinpelu 2021; Vodosin et al. 2021), with the COVID-19 pandemic having acted as an accelerator in terms of the need of enacting such regulations (Hussein et al. 2023).

While the lack of relevant regulations might be one of the barriers to digitalization of healthcare in LMICs, the lack of financial incentives presents a second, equally important, barrier. Digitalization in HICs was driven by a mix of private companies investing in the creation of digital health solutions, coupled by governmental incentives (Abernethy et al. 2022). For example, the driving force for the nation-wide adoption of EHRs in the US was the Health Information Technology for Economic and Clinical Health (HITECH) Act, enacted as part of the American Recovery and Reinvestment Act of 2009, incentivizing the adoption of EHRs (Institute of Medicine (IOM) 2004). This substantial public investment into the digitalization of healthcare was justified on the grounds of improved patient safety, operational efficiency and quality of care (Institute of Medicine (IOM) 2010). Indeed, by 2019, i.e., a decade after the enactment of the relevant legislation, approximately 86% of office-based physicians and 96% of non-federal acute care hospitals had adopted certified EHRs (Health IT Dashboard 2023). This represents the greatest adoption of digitalization in healthcare by data volume ever recorded. At the same time, multiple studies have documented improvements in care quality (Atasoy et al. 2019; Buntin et al. 2011), however, caution should be exercised in attributing the improvement of care quality to a singular parameter such as EHRs adoption, as it is likely due to the confluence of a number of factors.

In LMICs, the introduction of financial incentives across the board is unlikely due to the many competing financial pressures, however, incentivization programs have been successful within vertical sets of activities, e.g., physicians in Nigeria (Adedeji et al. 2022), CashAdvance schemes in Kenya for healthcare providers (de Wit et al. 2022), public health professionals during COVID-19 in Indonesia (Aisyah et al. 2022a) and others. As a result of the increasing digitalization of healthcare in LMICs, new models have been proposed for financial incentivization that are more appropriate for resource-restricted settings (Dohmen et al. 2022), including perspectives akin to micro-finance initiatives, where individual patients/professionals can be rewarded minimally for each completed digital interaction (Faulkenberry et al. 2022).

Lack of infrastructure: The lack of infrastructure is a consistent parameter of operating within LMICs, and equally for supporting digital health advancements (Chen et al. 2022; Nit et al. 2021), although the digital infrastructures are growing in many regions such as India, Vietnam (Winkie and Nambudiri 2023), Indonesia (Aisyah et al. 2022b), Rwanda (Chen et al. 2022) and others. The creation of digital infrastructure is not sufficient by itself, as digital literacy remains at low levels in many countries (Nit et al. 2021; El Benny et al. 2021). Hence the lack of infrastructure should not be regarded in isolation simply as the need for capital investment on technologies, but as part of the digital health environment inclusive of digital literacy. Finally, it should be noted that even when infrastructure is in place in LMICs, it is often required to perform in different ways and/or environments to the one it was originally created for. This need for ‘tropicalization’ of infrastructure, i.e., the adaptation of its performance to LMIC context is critical for the long-term performance and impact of such investments (Tran 2016; Coto-Solano 2020; Ombelet et al. 2018).

4 Applications in Healthcare: Adoption Versus Diffusion

At this point it is important to make a point differentiating between adoption and diffusion of digitalization. While there is a partial overlap of the factors that affect both, they do also contain distinct elements. Specifically, common elements include the availability of adequate training, policies and procedures for end users, and financial incentives at a system level (O’Donnell et al. 2018; Betmouni 2021). Differentiating factors focus on the scale of operations, the need to utilise prototypes to empower adoption, to engage social networks to affect diffusion, the interoperability of operations and the longer-term view that diffusion affords the digitization process (Mason 2015; Plum et al. 2020). There are currently many more studies published in the scientific literature regarding point adoption of digital solutions than studies on diffusion, owing to the fact that the latter require more time for investigation, but also suggesting that diffusion occurs at a slower rate (Omotosho et al. 2019). Having said that, the diffusion studies from rural India (Haenssgen and Ariana 2017; Schierhout et al. 2021), Cambodia (Nit et al. 2021), Kenya (Dohmen et al. 2022), and Nigeria (Adedeji et al. 2022), are informative in highlighting the need for context-driven implementations of digitalization and provide examples of how such a nuanced approach can be achieved in the field.

5 Opportunities

Taking the examples mentioned above into account—and the fact that they represent only a fraction of ongoing activities globally—it is clear that digitization of healthcare in LMICs is only at the first stages of this process. There have been many excellent reviews and books about the future opportunities that digitalization is likely to confer on healthcare in general (Menvielle et al. 2017; McKee et al. 2019; Glauner et al. 2021), but few focus on LMICs specifically, as it is a much more new and fragmented area and more difficult to assess and predict (Tambo et al. 2016). Therein also lie the many opportunities for the creation of new digital innovations, defining new pathways for implementation and impacting the population healthcare in ways that have remained nascent. The recent COVID-19 pandemic served as a proof-of-principle that digitization of healthcare in LMICs, at least some aspects of it such as surveillance and diagnostics, is indeed possible and can provide system-wide advantages to individual countries. Being able to link such initiatives to the introduced universal health coverage services provides the added-value proposition of a rich data resource that can be then used for data mining and inform decision-making of policy-makers and clinicians alike.

6 Conclusion

Digitalization of healthcare has gained momentum over the last two decades and the COVID-19 pandemic acted as an impromptu accelerator for the implementation of digital applications across the world. While this trend of increasing digitalization remains steadfast in the background, a number of differences exist in the digitalization trends and processes between HICs and LMICs. This chapter provided an overview of the main trends, i.e., ubiquity of digital environments, fragmentation and lack of uniformity, lack of regulations, lack of incentives and lack of infrastructure, inclusive of digital literacy. Specific examples have been highlighted from across LMICs, demonstrating that digitalization is indeed possible and can be successful, however it would need to follow a different pathway to that used in HICs and be adapted to the individual contexts in which it is being introduced.