Prototype of an organising framework for healthcare decarbonisation research: an exploratory classification study Authors: Mila Petrova, Fanny Burrows, Jan W van der Scheer, Timoleon Kipouros, James N Smith Published: 2026 ABSTRACT Abstract AbstractObjectivesTo develop an organising framework for healthcare decarbonisation research which goes beyond classification schemes based on scope 1, 2 and 3 emissions or lists of loosely connected themes and which is intended to support the coordination, funding and application of research into policy and practice. The organising framework was developed with a focus on the National Health Service (NHS) in England but enables application to healthcare systems more broadly.DesignAn exploratory classification study of over 160 research questions derived from a review and data extraction of nine systematic reviews, 13 stakeholder documents, two research priority exercises and four research funder sources. A further eight systematic reviews and 14 stakeholder documents, which were not used for direct data extraction, were used to test the emerging framework and specify thematic gaps.SettingPrimarily high-income healthcare systems, with a focus on the NHS in England.ParticipantsNot applicable.Primary outcomeA multilevel thematic framework representing current and missing areas of research in healthcare decarbonisation.ResultsThe framework comprises six top-level themes: Natural resource use and sources of carbon; Healthcare settings and workflows; Solutions; Stakeholders; Organisational levers for change; and Scientific measurement and theory (the ‘NHS-SOS framework’). At levels two and three, there were 39 and 86 subthemes, respectively.ConclusionsThis framework offers a structured, empirically derived representation of the emerging field of healthcare decarbonisation research. It is intended as a living tool to support shared understanding, prioritisation and action and to foster coherence in a currently fragmented research landscape.Keywords: Organisation of health services, Health policy, Research Design, Climate Change STRENGTHS AND LIMITATIONS OF THIS STUDY The study used a transparent and structured process to derive themes from over 160 research questions, sourced from a diverse set of systematic reviews, stakeholder documents, research priority exercises and funding calls. The study used a transparent and structured process to derive themes from over 160 research questions, sourced from a diverse set of systematic reviews, stakeholder documents, research priority exercises and funding calls. The inductive approach respected the complexity, breadth and multiple perspectives inherent to healthcare decarbonisation research. The inductive approach respected the complexity, breadth and multiple perspectives inherent to healthcare decarbonisation research. The study drew on a wide range of sources selected for conceptual and perspectival breadth but was nonetheless small relative to the volume of publications in the field. The study drew on a wide range of sources selected for conceptual and perspectival breadth but was nonetheless small relative to the volume of publications in the field. The potential for bias arising from the need to derive many research questions through interpretive analysis, as opposed to directly extract them, was mitigated through a clearly documented and transparent process outlining how interpretations were made. The potential for bias arising from the need to derive many research questions through interpretive analysis, as opposed to directly extract them, was mitigated through a clearly documented and transparent process outlining how interpretations were made. Introduction Healthcare systems are major contributors to global greenhouse gas emissions, accounting on average for an estimated 4.9% of national emissions, with a reported range of 1.5%–9.8%.1 If healthcare were a country, it would rank as the fifth largest emitter on the planet.2 Recognising the urgency of addressing this impact, the WHO and its member states have called for action ‘at pace and scale’, leading to the formation of the Alliance for Transformative Action on Climate and Health.3 The National Health Service (NHS) in England has taken a leading role in healthcare decarbonisation by setting ambitious targets to reach net zero by 2040 for its direct footprint and 2045 for its extended footprint, including that of its supply chain.4 A 2024 systematic review found that among six countries with available time series data, only NHS England and NHS Scotland had demonstrably reduced their healthcare-related emissions.1 While peer-reviewed literature may not fully reflect the progress of many countries, the progress made by the NHS stands out as sector-leading. Decarbonisation of healthcare is a complex, system-wide challenge involving a wide array of stakeholders, including government bodies, healthcare commissioners and providers, non-profit organisations, businesses, professional communities and patients. The research community has a critical role in supporting healthcare decarbonisation efforts by generating the evidence to inform action through policy and practice. Both decarbonisation activities and the research which supports them have typically been organised either by emission sources or broad thematic areas without a unifying structure. Examples of organisation by emission sources include Greenhouse Gas Protocol’s three emission scopes5 6 and the NHS Carbon Footprint and Carbon Footprint Plus used in England.4 Examples of organisation by broad themes include the nine inductively generated themes in Braithwaite et al’s comprehensive systematic review of 205 studies on strategies to decarbonise healthcare7 and the 19 themes of the Resource Library of the Centre for Sustainable Healthcare.8 Controlled vocabularies and formal ontologies for databases have also made little headway. For instance, the Medical Subject Headings thesaurus of the National Library of Medicine, which represents the fields of biomedicine and health for the purposes of cataloguing information, does not include the term ‘decarbonisation’ and only includes three concepts in its tree for climate change.9 Without a broadly shared representation of the field, the research community risks fragmented efforts, missed synergies and overlooked gaps. A common frame of reference—here termed an ‘organising framework’—could enable more systematic identification of research needs, support system-level, coordinated research and action across programmes of work and facilitate knowledge exchange. This could be of particular importance as new evidence is generated and knowledge gaps are filled. While a fully standardised, finalised consensual map of the field may be neither feasible nor desirable, a shared, evolving framework can offer structure to key efforts such as capturing and synchronising practical evidence-based action, guiding impactful research, informing debate and forging new links between currently disconnected or weakly connected programmes and stakeholders. It may also support greater consensus on how research in this area is conceptualised, organised and reported, helping to build a more coherent and cumulative evidence base. The aim of this work was to develop an organising framework for healthcare decarbonisation research, focused on the NHS in England but applicable to healthcare systems more broadly and intended to support the coordination, funding and application of research into policy and practice. Methods We developed an organising framework for healthcare decarbonisation research by abstracting shared features of and classifying research questions identified through a review of relevant systematic reviews, stakeholder documents, research priority exercises and funding calls. A distinctive feature of our approach was the highly transparent process of identification and formulation of research questions, including many that could not be directly extracted from the source documents but were derived through an interpretive process necessitating subjective judgements, which we aimed to render as auditable as possible. ‘Transformations’ of the contents of source documents to derive research questions from them were often necessary where documents addressed broader, narrower or different contexts than our focus, namely the UK healthcare system (NHS). Further detail on our methodological approach is provided below. The study required no ethics committee approval. It was a desk review of peer-reviewed and grey literature, with no access to human participants or patient data involved. Source documents The pool of source documents was selected to maximise conceptual breadth across both research and policy domains (see online supplemental file 1 and online supplemental figure 1 for details on the selection process). Nine systematic reviews17 10,16 (out of 118 screened for relevance), 13 stakeholder reports and web sources24 17,27 (identified from 35 key stakeholder websites searched), two research priority exercises28 29 and four research funder sources30,33 (out of over 430 funding themes, calls and associated projects screened) were used to derive research questions. A further eight systematic reviews and 14 stakeholder documents, which were not used for direct data extraction because they did not meet fully the inclusion criteria but were of high thematic relevance or met the inclusion criteria but did not yield research questions, were used to generate ‘placeholder themes’ (see below) and test the emerging organising framework. See online supplemental file 1. Research questions The process by which research questions were derived from the source documents is described in detail in onlinesupplemental files 2 3 and is summarised below. Seven types of source contents were used to extract (word-by-word) or formulate (through an interpretative process) research questions: Research gaps or recommendations: statements which explicitly identify a research gap or need, or make an explicit research recommendation. Research gaps or recommendations: statements which explicitly identify a research gap or need, or make an explicit research recommendation. Research limitations: in a systematic review, statements about the limitations of the systematic review and/or source studies, or statements about risk of bias assessments. In policy documents, these were typically statements that describe limitations of national data. Research limitations: in a systematic review, statements about the limitations of the systematic review and/or source studies, or statements about risk of bias assessments. In policy documents, these were typically statements that describe limitations of national data. Research availability: statements that certain topics have been researched to a limited degree which are, however, unaccompanied by an explicit research recommendation. Research availability: statements that certain topics have been researched to a limited degree which are, however, unaccompanied by an explicit research recommendation. Research-policy-practice collaborations: statements which refer to an interaction (ongoing or intended) between researchers, policymakers and practitioners. Research-policy-practice collaborations: statements which refer to an interaction (ongoing or intended) between researchers, policymakers and practitioners. Intentions to fund or commission research — associated problem statements: statements about problems or research areas which are targeted by a current funding initiative/call for proposals; statements about proposals which have just received funding; statements expressing an intention or recommendation to commission a study, with no specific detail of a funding call. Intentions to fund or commission research — associated problem statements: statements about problems or research areas which are targeted by a current funding initiative/call for proposals; statements about proposals which have just received funding; statements expressing an intention or recommendation to commission a study, with no specific detail of a funding call. Recommendations for guideline development: statements concerning the development of evidence-based guidance in healthcare. Recommendations for guideline development: statements concerning the development of evidence-based guidance in healthcare. ‘Verbal actions’ requiring new, specialised knowledge: statements creating high expectations of practical action, such as statements of commitment, requirement or directive, which also refer to research, data, evaluation or other forms of new knowledge generated by experts. ‘Verbal actions’ requiring new, specialised knowledge: statements creating high expectations of practical action, such as statements of commitment, requirement or directive, which also refer to research, data, evaluation or other forms of new knowledge generated by experts. Where possible, contents of these seven types were used to directly extract research questions. In many cases, however, research questions had to be formulated from the source contents through an interpretative process that applied one or more of the following eight transformations: Attributing a focus on decarbonisation. Attributing a focus on decarbonisation. (Re)specifying boundaries (the context to which the research question pertains). (Re)specifying boundaries (the context to which the research question pertains). Splitting or combining in new ways elements from complex statements. Splitting or combining in new ways elements from complex statements. Raising the level of abstraction. Raising the level of abstraction. Articulating further concepts or contents. Articulating further concepts or contents. Adding a research perspective. Adding a research perspective. Adding a meta-science perspective. Adding a meta-science perspective. Editing for standalone clarity (see online supplemental file 2 for detail). Editing for standalone clarity (see online supplemental file 2 for detail). Approach to developing the organising framework Research questions were grouped together and/or a category (class, theme) was formed and labelled by identifying and abstracting the core characteristic(s) of one or more questions. Often, the process of comparing questions to determine lines of similarity and difference in order to group them together or keep them apart coincided with the process of forming and labelling a category. Categories could, however, also be formulated from single questions. In some cases, we also added ‘placeholder themes’ to the framework, which did not have a corresponding research question. Placeholder themes were generated in two ways. Some reflected areas of limited knowledge highlighted in source documents whose content was not specific enough to support a well-formulated research question. Others corresponded to a research gap identified in dated sources, which we read as background information due to their high relevance to the topic. In both cases, the placeholder theme captured a conceptual gap in the evolving organising framework. Placeholder themes are clearly labelled as such to indicate that a theme needs further evidence and debate to articulate current or emerging research questions and determine its relative priority. The above process was anchored in ideas from ontology development in information science and philosophical frameworks such as promiscuous realism and pluralistic realism.34,37 In information science, ontologies provide a shared vocabulary and structure for organising domain knowledge, making assumptions explicit and enabling sharing, reuse, integration and comparison of domain knowledge. The philosophical frameworks of promiscuous and pluralistic realism (which are highly influential in the philosophy of biology but remain largely unfamiliar in the philosophy of evidence-based medicine) emphasise, among other issues, that there are multiple valid ways to classify species and objects, as opposed to a single best and/or most objective way. These philosophical concepts informed much of our approach to labelling and grouping themes, recognising the value of representing multiple perspectives and avoiding overly rigid or reductionist structures. A set of further principles guided our approach to labelling and grouping themes based on research questions and developing the overall structure of the framework (in discussing the framework below, we will prioritise the term ‘themes’ as opposed to ‘categories’ or ‘classes’). Altogether, these principles were: Taking multiple perspectives: lifting the variety of perspectives to decarbonisation into high-level categories as opposed to committing to a dominant perspective (such as that of sources of carbon emissions). Taking multiple perspectives: lifting the variety of perspectives to decarbonisation into high-level categories as opposed to committing to a dominant perspective (such as that of sources of carbon emissions). Giving weight to the mental models of policymakers: giving some priority to the mental models and needs of policy and decision-makers, primarily in recognition of their power to drive change at the level of national healthcare systems. Giving weight to the mental models of policymakers: giving some priority to the mental models and needs of policy and decision-makers, primarily in recognition of their power to drive change at the level of national healthcare systems. Revising dynamically, while allowing for periods of stabilisation: aiming for a framework that is regularly updated yet, for certain periods, retains stable features so as to serve its organising functions. Revising dynamically, while allowing for periods of stabilisation: aiming for a framework that is regularly updated yet, for certain periods, retains stable features so as to serve its organising functions. Accommodating variable levels of commitment to decarbonisation, including resistance to it: aiming for a framework which, in its formulation of categories, is open to different levels of endorsement of the agenda for healthcare decarbonisation, including opposition to it. Accommodating variable levels of commitment to decarbonisation, including resistance to it: aiming for a framework which, in its formulation of categories, is open to different levels of endorsement of the agenda for healthcare decarbonisation, including opposition to it. Prioritising intuitive use versus technical language, strict hierarchies and brevity when conflicts arise: priority was given to user-friendliness through plain language and clarifying detail added to the labels of categories. For topics receiving significant attention, tensions between hierarchical level in logical terms and hierarchical level in terms of visibility were resolved in favour of visibility. Prioritising intuitive use versus technical language, strict hierarchies and brevity when conflicts arise: priority was given to user-friendliness through plain language and clarifying detail added to the labels of categories. For topics receiving significant attention, tensions between hierarchical level in logical terms and hierarchical level in terms of visibility were resolved in favour of visibility. Incorporating opportunities to align with external conceptual models: some category labels were informed by popular conceptual frameworks in healthcare, with the vision to incorporate them more explicitly in future iterations. Incorporating opportunities to align with external conceptual models: some category labels were informed by popular conceptual frameworks in healthcare, with the vision to incorporate them more explicitly in future iterations. Attending to conflicts and discrepancies: complex interactions, trade-offs and dilemmas were lifted to higher hierarchical levels to facilitate awareness and resolution. Attending to conflicts and discrepancies: complex interactions, trade-offs and dilemmas were lifted to higher hierarchical levels to facilitate awareness and resolution. Balancing transferability with domain specificity: seeking to combine the use of categories that emphasise more formal, abstract, cross-cutting features and categories reflecting thematic domains. Balancing transferability with domain specificity: seeking to combine the use of categories that emphasise more formal, abstract, cross-cutting features and categories reflecting thematic domains. Colocating topics to enable syntheses rather than reinforcing silos: bringing together themes that reflect approaches with common goals but are generally considered contrasting, conflicting or too different, to enable the exchange of ideas. Colocating topics to enable syntheses rather than reinforcing silos: bringing together themes that reflect approaches with common goals but are generally considered contrasting, conflicting or too different, to enable the exchange of ideas. Accommodating blurred boundaries between decarbonisation, climate change and sustainability: recognising the grey zones between healthcare decarbonisation and broader climate and health issues; formulating grey zone categories that avoid artificial splits in how the practical work and dominant debates have been configured. Accommodating blurred boundaries between decarbonisation, climate change and sustainability: recognising the grey zones between healthcare decarbonisation and broader climate and health issues; formulating grey zone categories that avoid artificial splits in how the practical work and dominant debates have been configured. Further details on each principle, including examples, are provided in online supplemental file 3. Derivation of research questions, abstraction of themes from them and restructuring of the framework continued until the latter reached a point of stability, at which no major new themes or structural revisions were introduced. Patient and public involvement There was no direct patient and public involvement in this study. Patient and public involvement organisations’ websites were examined but did not generate sources of research questions. Furthermore, patient perspectives were sought within the sources reviewed but none of those included in the data extraction represented this perspective in detail. It is recognised that patients are key stakeholders in the healthcare system and as such their input will be sought in future developments of the framework, building on the prototype generated through this study. Results We derived over 160 research questions from across nine systematic reviews, 13 stakeholder documents, two research priority exercises, and four research funder sources. The process of classifying these research questions, using 10 key principles, resulted in the proposed organising framework for healthcare decarbonisation research: the NHS-SOS framework, as per the initials of the six top-level themes: Natural resource use and sources of carbon. Natural resource use and sources of carbon. Healthcare settings and workflows. Healthcare settings and workflows. Solutions. Solutions. Stakeholders. Stakeholders. Organisational levers for change. Organisational levers for change. Scientific measurement and theory. Scientific measurement and theory. These six ‘level one’ themes (represented in figure 1) include 39 more specific ‘level two’ subthemes (box 1), with the latter broken down further into 86, even more granular, ‘level three’ subthemes. A small number of ‘level four’ subthemes were also generated. The full framework, including level three and four subthemes, is presented in online supplemental file 4. Organising framework for healthcare decarbonisation research (NHS-SOS) with top-level (‘level one’) themes. NHS-SOS, Natural resource use and sources of carbon; Healthcare settings and workflows; Solutions; Stakeholders; Organisational levers for change; and Scientific measurement and theory. Organising framework for healthcare decarbonisation research (NHS-SOS) with levels one and two themes A. Natural resource use and sources of carbon A1. Energy sources and water use in healthcare. A2. Medicines, equipment and consumables as decarbonisation targets; associated production, procurement and supply chains. A3. Waste, reuse, recycling and circularity in healthcare. A4. Food and nutrition in healthcare contexts and for health-related outcomes. A5. Transport, travel and mobility for healthcare purposes and in healthcare contexts. B. Healthcare settings and workflows B1. Decarbonisation of healthcare spaces (buildings, facilities, estates, sites, etc). B2. Decarbonisation of healthcare services (primary care, secondary care, community care, etc) (placeholder). B3. Decarbonisation of clinical specialities and/or health conditions. B4. Decarbonisation of healthcare supply chains. B5. Sustainable reorganisation of healthcare systems in contexts of emergencies, disasters and other stresses and shocks. C. Solutions C1. Solutions for decarbonising healthcare (big picture, cross-cutting issues). C2. Digital solutions. C3. Innovative materials. C4. Innovative designs. C5. Capture and removal of greenhouse gases (placeholder). C6. New models of care and decarbonisation. C7. Combining innovation and ‘back to basics’ approaches. C8. Improved healthcare efficiency as a route to decarbonisation. C9. Education, training and information provision. C10. Behaviour change-focused solutions for decarbonising healthcare. C11. Complex interventions for decarbonising healthcare. C12. The carbon footprint of adaptations to climate change and decarbonisation solutions themselves. D. Stakeholders D1. Stakeholders in healthcare and/or decarbonisation (big picture, cross-cutting issues). D2. The healthcare workforce and the decarbonisation of healthcare. D3. Types of healthcare workers and the decarbonisation of healthcare. D4. Patient populations and groups and the decarbonisation of healthcare (placeholder) E. Organisational levers for change E1. Policy and governance for the decarbonisation of healthcare. E2. Funding and financial mechanisms for the decarbonisation of healthcare. E3. Practical guidance for the decarbonisation of healthcare. E4. Certification and accreditation schemes for healthcare decarbonisation (placeholder). E5. Visions, models and frameworks of healthcare provision which incorporate decarbonisation principles. E6. Networks, communities of practice and platforms for collaborative work on healthcare decarbonisation. E7. Managing (perceived) trade-offs and conflicts of values in decarbonising healthcare. F. Scientific measurement and theory F1. Measuring the carbon footprint and the decarbonisation of healthcare. F2. Concepts, frameworks and theories concerning the decarbonisation of healthcare. F3. Healthcare decarbonisation in the social conversation. F4. Needs for quality improvement in research on healthcare decarbonisation. F5. Systematic approaches to identifying research and gaps in research on healthcare decarbonisation. F6. Turning knowledge into action on healthcare decarbonisation. Discussion Main findings We developed an organising framework for healthcare decarbonisation research (NHS-SOS) based on research questions derived from systematic reviews, stakeholder documents, research priority exercises and research funder sources. The framework comprises six top-level themes, each representing a distinct perspective for examining research questions about healthcare decarbonisation, supported by three further levels of subthemes. These six themes function as both lenses for inquiry and domains of action. The framework provides a starting point for constructing a shared representation of the field of healthcare decarbonisation to enable systems thinking and systemic action. Study strengths and limitations To our knowledge, this is the first detailed, multilevel mapping of the field of healthcare decarbonisation research. The inductive approach underpinning the generation of themes captured how researchers, policymakers and practitioners are engaging with the challenge of decarbonisation. It respected the complexity, breadth and multiperspectival nature of the landscape of healthcare decarbonisation research and avoided imposing narrow conceptualisations, for example, solely carbon-footprint based or unrealistically rational pre-existing ones. The logically disciplined, even if flexible and user-friendly structure, in turn, avoided the downsides of lists of loosely connected themes, often combining themes of different levels of generality and using different lenses. The principles used in individuating, labelling and classifying themes were highly unusual as a combination but, we argue, have significant heuristic potential when jointly applied. They bring together ideas coming from the philosophy of science, particularly theories of classification; basic rules for designing domain ontologies; attention to the needs of policymakers and a commitment to user-friendliness, among other conceptual influences. These principles and their combination do, however, entail value judgements, which merit further debate and articulation in the process of future coproduction work on the organising framework. A limitation of the initial development of this prototype classification framework was that, as a desk-based exercise, this process did not include patient and public participation. It is planned that patients will be involved in the next stage of development including reviewing the prototype, critiquing the emerging structure and themes, advising on usefulness and potential uses and informing its future development. While the framework was developed with a focus on the NHS in England, its research questions—and the themes and subthemes they informed—are likely broadly applicable across healthcare system levels and transferable to healthcare systems in other countries, particularly higher income ones. In addition, the method developed in this work for building an organising framework may hold relevance for other complex systemic challenges in healthcare research and beyond. The data source of research questions on which the framework was based was obtained through a highly transparent and structured process from a diverse set of strategically selected sources. Nevertheless, biases may have arisen from the limited number of included sources and the fact that many research questions could not be directly extracted but were formulated through interpretive processes. It is also important to acknowledge that there is a level of tension between the nature of the data source we used—research questions indicating research gaps—and the goal of representing a field of research. Even if research gaps signal strongly the themes and phenomena in a research field, a process of mapping and classifying existing research may have generated a different outcome. It is recognised that this field is rapidly evolving, supported by increasing research funding, and relevant research has been published since this work was undertaken.38,43 Implications for research and practice We envisage this framework being used in organising co-production processes around the prioritisation, design, conduct, dissemination and uptake of research to advance the decarbonisation of the NHS in England. With appropriate adaptation, the framework may also meet research-into-action needs of other healthcare systems, particularly in high-income countries. A fully comprehensive framework—based on an exhaustive review of documents available in electronic databases and grey literature—was beyond the scope of this study but may also not be the most effective way forward. The field of healthcare decarbonisation is rapidly evolving while there is no shared representation of it. Sharing an initial framework now, intended to serve as a living tool for continued cocreation processes, is likely of more value than a fully polished framework later—not least given the urgency of the climate crisis. Further development of the framework will require new empirical inputs and engagement and involvement processes. New empirical inputs may include not only emerging research, additional systematic reviews and stakeholder documents, but also content which is unconventional in academic research. For instance, submissions to start-up business competitions and award schemes in the climate innovation space may highlight emerging, cutting-edge research and implementation areas of work. The analysis of ‘green plans’ of healthcare organisations and of project evaluations conducted in healthcare settings may offer themes grounded in operational realities. The development of the NHS-SOS framework as a prototype of a classification system of research questions contributes to the field of health-related and research-related classification, taxonomy and ontology development. Biomedical ontologies are numerous, with 1275 included in BioPortal, the most comprehensive repository of biomedical ontologies, as of March 2026.44 In the field of research classifications, the Health Research Classification System, developed by the UK Clinical Research Collaboration on the basis of the WHO International Classification of Diseases and the Common Scientific Outline, has been used both by UK and international health research funders to classify and strategically assess award portfolios.45 The rise of information technology and AI requires increasingly complex classification systems and allows for non-manual methods of generating these. Bibliometric analysis of technology in sustainable healthcare is an example of the use of non-manual methods to describe the rapidly evolving field of technology in healthcare.46 Growing and refining the NHS-SOS organising framework requires its mapping against and partial or full integration with relevant domain ontologies. The work would also benefit from engagement with research and expertise concerning user interactions with classification systems and new developments in classification theory. Importantly, we see the ongoing development of the NHS-SOS framework as inseparable from the practical work of enabling system-level action on the decarbonisation of healthcare. The framework is not intended primarily as a theoretical model or cognitive heuristic, but as a practical tool to support a growing collective understanding of the field, foster systems thinking and enable coordinated and contextualised action. It is envisaged as one tool, out of many, needed to improve the flow of research evidence into practice and policy in the field of decarbonisation and to ensure that practice and policy needs shape new research. We invite stakeholders to contribute to an open discussion about what a new generation of research-into-action infrastructure for healthcare decarbonisation should look like. Conclusion This work presents an organising framework for healthcare decarbonisation research, developed through a rigorous inductive process and designed to accommodate multiple perspectives and points of entry into the field. The framework offers a foundation for structuring this emerging field and supporting the coproduction of research infrastructure needed to accelerate research in response to the climate emergency. Supplementary material Footnotes Funding: The project was directly funded by the University of Cambridge and NHS England. In addition, JWvdS is based on The Healthcare Improvement Studies Institute (THIS Institute), University of Cambridge. THIS Institute is supported by the Health Foundation, an independent charity committed to bringing about better health and healthcare for people in the UK (grant number RHZF/001, RG88620). The funders did not influence the results/outcomes of the study despite author affiliations with the funders. Prepublication history and additional supplemental material for this paper are available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2025-111213). Provenance and peer review: Not commissioned; externally peer-reviewed. Patient consent for publication: Not applicable. Ethics approval: Not applicable. Data availability free text: Underlying data extracted and used to generate framework may be available on reasonable request. Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research. Data availability statement Data are available upon reasonable request. References Supplementary Materials Data Availability Statement Data are available upon reasonable request.