Who is most at risk of digital exclusion within healthcare?

Abstract Introduction There has been a rapid growth in the use of digital technologies in healthcare to improve population health. However, due to digital exclusion (DE) (the unfair differences in motivation, access and use of these technologies), some demographic groups are unable to harness the health benefits of these technologies. Various frameworks have been designed to highlight groups vulnerable to health inequities, but none focus on DE. To provide better support for digitally excluded individuals, we must start by identifying demographic groups most at risk of DE. Aim Review the literature to identify demographic groups most at risk of DE within healthcare. Methods PubMed, Google Scholar and Scopus were searched on 22nd October 2022 in a non-systematic way to scope and examine emerging evidence surrounding groups at risk of DE within healthcare. English-language peer-reviewed articles and grey literature (e.g., government reports and regulatory organisation documents) identifying demographic groups at risk of DE and possible factors influencing exclusion were included. Search terms included words associated with DE and health inequity. Data extracted from the included articles included the different demographic groups highlighted and reasons for DE. Results Analysis of twenty-two included articles highlighted six common sociodemographic factors associated with DE that we compiled into six groups. These groups include Culture (ethnicity, language, religion), Limiting conditions (visual and hearing impairments), Education, Age, Residence (rural, deprived areas, homeless), and Socioeconomic status, forming the CLEARS framework. Cultural factors associated with DE included bias in artificial intelligence models used in healthcare technologies, low sensitivity in sensors when used across different ethnic backgrounds, and lack of consideration for different religious groups. Limiting conditions may make it difficult to use the technology, such as reading health information on small screens if visually impaired. Low educational attainment and old age (over 65) were frequently associated with poor digital literacy skills and poor understanding of the benefit of health technologies, thus reducing motivation and use of health technologies. Residential factors centred around barriers to accessing infrastructure to support connectivity (e.g. WiFi) and socioeconomic factors, including financial barriers to compatible devices and connectivity. Individuals who identify with two or more of these groups (intersectionality) are at even greater risk of DE. Conclusion This review identified six groups at risk of DE, the contributing reasons for exclusion, and the role of intersectionality. The CLEARS framework generated from this review can guide researchers and health technology developers in identifying groups that need to be involved in the co-design of digital health technologies to support inclusion. During the development of the CLEARS framework, groups in previous health inequities frameworks, such as the LGBTQ+ community, were considered, but later excluded due to the lack of representation within the literature. However, it is also important to recognise underserved groups are less likely to appear in the literature. Further work is needed to design tailored guidelines to support the development of inclusive digital health technologies that meet the needs of underserved groups.