The environmental sustainability of data-driven health research: A scoping review

Abstract

Data-Driven and Artificial Intelligence technologies are rapidly changing the way that health research is conducted, including offering new opportunities. This will inevitably have adverse environmental impacts. These include carbon dioxide emissions linked to the energy required to generate and process large amounts of data; the impact on the material environment (in the form of data centres); the unsustainable extraction of minerals for technological components; and e-waste (discarded electronic appliances) disposal. The growth of Data-Driven and Artificial Intelligence technologies means there is now a compelling need to consider these environmental impacts and develop means to mitigate them. Here, we offer a scoping review of how the environmental impacts of data storage and processing during Data-Driven and Artificial Intelligence health-related research are being discussed in the academic literature. Using the UK as a case study, we also offer a review of policies and initiatives that consider the environmental impacts of data storage and processing during Data-Driven and Artificial Intelligence health-related research in the UK. Our findings suggest little engagement with these issues to date. We discuss the implications of this and suggest ways that the Data-Driven and Artificial Intelligence health research sector needs to move to become more environmentally sustainable.

Environmental impacts associated with the manufacturing of personalised medicines: a scoping review

“…It also includes questions about who would pay financially and environmentally for these medicines versus who will benefit. Often when improved health is the driver of research, those most at risk from the financial and environmental costs, are those least likely to benefit from the health innovation (111,112),raising environmental justice type concerns. Furthermore, investment in personalised healthcare solutions could also risk squeezing resources away from more preventative, low tech, cheaper health innovations (112,113).…”

“…Often when improved health is the driver of research, those most at risk from the financial and environmental costs, are those least likely to benefit from the health innovation (111,112),raising environmental justice type concerns. Furthermore, investment in personalised healthcare solutions could also risk squeezing resources away from more preventative, low tech, cheaper health innovations (112,113). These approaches are more likely to improve the health of those who cannot access personalised medicines as they often address the social determinants of health.…”

The current state and potential evolution of carbon emissions in the healthcare sector: a narrative review article

“…Curtailing the disposal of viable pharmaceutical products is a feasible indirect approach (Tong et al, 2011;The Academy of Medical Royal Colleges, 2014). In addition, NHS should ensure that products are being prepared under environmental sustainability regulations (NHS Supply Chain, 2012;Samuel and Lucassen, 2022). Public awareness sessions are equally important to convey the extent of environmental severity (Edwards and Roberts, 2009;Capstick et al, 2015;Rahimi, 2020).…”

The environmental impact of data-driven precision medicine initiatives

“…However, poor practices also continue (Bilham, 2021 ), often attributed to gaps in the regulation (Magallón Elósegui, 2020 ) 17 or to the fact that initiatives are often developed by powerful companies who shape the discourse and neglect important stages of the mining life cycle (Phadke, 2018 ), and who outsource responsibility ‘at a distance’ (Calvão et al, 2021 ; Deberdt and Billon, 2021 ), disregard complexity (Ayeh and Bleicher, 2021 ) and do not engage with the social and cultural context of the industry (Hecht, 2012 ; Mantz, 2018 ; Smith, 2022 ). While health-related and other adverse mining-associated impacts are context specific and will vary depending on the type of mining, the mineral being extracted, as well as the economic, political and cultural context (Bilham, 2021 ), Samuel and Lucassen ( 2022 ) have argued that those working in precision medicine must become more aware of these issues in order to mitigate them as much as possible.…”

“…Rather, the literature has largely focused on promised benefits and increased patient autonomy (Samuel and Farsides, 2017 ; Birk and Samuel, 2020 ). One exception is Samuel and Lucassen’s ( 2022 ) recent mapping of the literature exploring specific environmental impacts of data-driven health research, some of which included research associated with precision medicine. These authors show how most studies have focused on developing software and hardware solutions using green IT, that is, an approach to IT that produces minimal waste during its development and operation and promotes recyclability, with less focus on a consideration of the need to think about changes in data practices (Samuel and Lucassen, 2022 ).…”