Measuring the carbon footprint of clinical research activities: a scoping review of measurement tools and methods

Abstract

Background

Healthcare contributes substantially to global greenhouse gas emissions; however, the environmental impact of clinical research activities remains poorly understood. Quantifying emissions is a necessary first step toward reducing the carbon footprint of research.

Objective

To map and synthesise literature measuring carbon emissions associated with clinical research activities, with a focus on research domains assessed, tools and methods used and units of measurement reported.

Methods

A scoping review was conducted following the Arksey and O’Malley methodology and Preferred Reporting Items for Systematic Reviews and Meta-Analyses for Scoping Reviews guidance. PubMed, Web of Science, CINAHL, Scopus, EconBiz, GreenFile and ProQuest were searched from database inception to June 2026. Eligible studies reported measurement of carbon emissions related to clinical research activities. Data were charted and synthesised narratively.

Results

Twenty-five studies met the inclusion criteria, most published between 2019 and 2025 and primarily from Europe and the UK. Studies used a wide range of tools and resources, including life cycle assessment databases, online calculators, international standards, government emission factor datasets and healthcare-specific sustainability frameworks. A clear trade-off emerged between methodological rigour and accessibility: comprehensive life cycle assessment tools required expertise and licensing, while simpler calculators enabled rapid but less precise estimates. Carbon dioxide equivalent and global warming potential over a 100-year time horizon were the dominant reporting metrics, although variation in functional units limited comparability across studies.

Conclusions

The methodological landscape for measuring emissions in clinical research is fragmented and lacks standardisation. Development of consensus guidance and reporting standards is needed to support consistent measurement and reduction of research-related emissions.

STRENGTHS AND LIMITATIONS.

• This review provides the first comprehensive mapping of tools and methods used to quantify carbon emissions in clinical research and includes both peer-reviewed and grey literature across multiple disciplines.

• Only English-language studies were included, which may have introduced geographic bias.

• The search has been updated in June 2026 to include additional relevant literature.

• This study did not assess the methodological quality of included studies.

• Heterogeneity in methods and reporting limited direct comparison across studies.

Background

Clinical research is essential for generating new medical knowledge and evaluating the safety and effectiveness of drugs, treatments and healthcare innovations.¹ Clinical research encompasses a range of study designs, including observational and experimental and both interventional and non-interventional studies.¹ This provides vital evidence for the safety and evaluation of treatments and therapies, advancing medical knowledge and improving our understanding of both communicable and non-communicable diseases, ultimately improving patient outcomes.²

Periods of rapid scientific advancement, such as the COVID-19 pandemic, have highlighted the importance of promptly generating knowledge to inform disease prevention and control.³ Studies of this nature, particularly clinical trials, represent a key cornerstone of medical research, requiring significant investment of time and resources.⁴ Collectively, a thriving clinical research environment not only improves patient outcomes and advances medical knowledge but supports innovation and drives economic growth.¹

Global concern regarding greenhouse gas (GHG) emissions has intensified in recent years. These increasing emissions have been linked to sharp climate change impacts that have resulted in more frequent occurrences of weather events such as forest fires, droughts and storms.⁵ While clinical research and healthcare are vital, they have been linked to contributing large amounts of these GHG emissions.^{6 7} Healthcare systems are estimated to contribute approximately 4% of global GHG emissions.⁷ Approximately 350 000 clinical trials have been estimated as contributing upwards of 27 million tonnes of GHG emissions across their lifecycle.⁸

These findings highlight the urgent need for action. The environmental burden of clinical research activity requires systematic assessment, with particular focus on identifying the activities generating the most significant emissions, as well as the ways in which emissions are measured and accounted. In accordance with commitments outlined in the Paris Climate Agreement and the United Nations Framework Convention on Climate Change,⁹ it is essential to develop robust methods for quantifying emission generation across clinical research activities and to identify effective mitigation strategies to reduce those emissions.

While previous studies have quantified emissions associated with healthcare systems and individual clinical trials, no review has systematically mapped the tools and methods used across clinical research activities. Understanding this methodological landscape is essential to support consistent measurement, enable comparison between studies and guide development of lower-carbon research practices. This scoping review addresses this gap by synthesising the available evidence on how emissions associated with clinical research are currently measured and reported.

Aims and objectives

This scoping review aimed to map and synthesise literature measuring carbon emissions associated with clinical research activities. Specifically, the review examined:

1. Types of research activities assessed.

2. Tools and methods used to quantify emissions.

3. Domains and subdomains of measurement.

4. Units used to report emissions.

Methods

Patient and public involvement

No patient or public involvement was included in the design, conduct, reporting or dissemination of this study.

A scoping review methodology was chosen to enable rapid identification and synthesis of evidence relating to carbon emissions from clinical research activities. Scoping reviews provide a rigorous and transparent approach for addressing broad research questions and mapping emerging evidence.¹⁰

The scoping review framework underpinning this study by Arksey and O’Malley (2005),¹¹ as well as the scoping review guideline developed by the Joanna Briggs Institute (JBI),¹² informed the conduction of this review. The review followed the Preferred Reporting Items for Systematic Review and Meta-Analysis for Scoping Reviews (PRISMA-ScR) checklist (online supplemental file 1).¹³

Search strategy

A specialist librarian (DS), based at University College Dublin, assisted the research team with the development of the review search strategy, including identification of key search terms and phrases. Several databases were searched, including PubMed, Web of Science, CINAHL, Scopus, EconBiz, GreenFile and ProQuest. This ensured both health- and non-health-related databases were systematically searched. The full search strategy for each database can be located in online supplemental file 2.

The search aimed to identify literature measuring GHG emissions generated from research activities in clinical settings. As such, terms related to and including clinical research, carbon emissions and healthcare were included. Key terms and phrases were examined from pilot searching preliminary sources of literature to generate appropriate search terms. Key terms were adapted for each database and included relevant subject headings, for example Medical Subject Headings (MeSH). A descendancy and ascendancy literature-searching approach was undertaken, in line with JBI guidance, to systematically identify any adjacent relevant literature sources.¹² Searches were conducted from database inception to June 2026.

Eligibility criteria

Inclusion criteria are outlined in table 1 and are specified according to the domains of the PCC framework: population, concept and context. The PCC framework was used to inform the research questions, search strategy and eligibility criteria.¹⁴ Studies that described the measurement of carbon emissions related to research activities or domains, including units of measurement, as well as tools or methods of measurement, were eligible for inclusion. The review excluded articles or materials that described carbon emission measurement unrelated to research activities, including measurement of clinical activities or equipment that is not associated with research. Additionally, studies that were not associated with clinical or healthcare settings, or materials that provided quantification measures but did not describe the measures, were also excluded. Many study designs were eligible for inclusion if the data adhered to the aim of the review. There was no restriction on geographical location or timeframe for studies. Due to a lack of resources needed to support translation, only papers published in English were included.

Table 1. Inclusion and exclusion criteria.

Inclusion criteria

Primary research studies

Peer-reviewed grey literature sources, which may include dissertations and theses, government reports, policy statements, clinical/environmental reports or guidelines

Case studies reporting on the measurement of carbon emissions

Frameworks developed that detail carbon emission measurement

Exclusion criteria

Secondary research sources (ie, reviews; systematic, scoping etc.)

Full texts unavailable in English

Study selection and data screening

Studies selected after database searching were first exported to EndNote (V.21), a reference management software for organisation, before being uploaded to Covidence, a systematic review software tool.¹⁵ Two authors (DK, LB) completed title and abstract screening of all identified articles against the eligibility criteria. The majority of conflicts that arose between both authors were discussed and resolved. Any remaining conflicts were resolved by a third author (LON). Full-text articles were then selected, and screening was completed by two authors (DK, LB). Similarly, any conflicts were resolved through discussion.

Data charting

Using a pilot data charting form, the first author (DK) charted the data from a selection of included articles (n=5). The second author (LB) randomly selected a sample of articles (n=2) from this selection for charting. The first and second authors then conducted a peer review of the data charted to ensure it matched the pilot form and there were no discrepancies. Both authors discussed the fields of data charted to identify any additional data relevant for collection. In line with PRISMA-ScR guidelines, information on study funding was also extracted. Data were charted for the following domains: country, research type, study aim, research activities evaluated for carbon emissions, tool/method of measurement, validity of tool/method, main domain(s) of carbon capture, sub-domain(s) of carbon capture, method of carbon capture, unit of carbon capture, type of study (clinical trial, non-clinical trial), setting (of research activities) and funding.

Data synthesis

A narrative approach was taken to synthesise the data. Narrative synthesis is an approach that involves first, a preliminary synthesis of the literature data to determine patterns, followed by assessment of the evidence to identify strengths and weaknesses and lastly using the evidence to form conclusions and recommendations.¹⁶

The narrative synthesis approach was used to extract the tools and methods of measurement, as well as the various study types and research activities. The main domains and subdomains of the tools and methods were highlighted where possible, with identification and examination of relationships or patterns between domains.

Protocol and registration

The protocol for this study is published on Health Research Board (HRB) Open, a peer-reviewed open and accessible online research platform that can be located via https://doi.org/10.12688/hrbopenres.14092.2 or in online supplemental file 3.

Results

Search results

Twenty-five studies met the inclusion criteria and were included in the final synthesis. Database searches identified 3089 records across the seven included databases. After removal of duplicates (738), 2351 records were screened for title and abstract. After this step, 2259 articles were excluded as they did not meet the inclusion criteria (figure 1).

Figure 1. PRISMA-ScR diagram. PRISMA-ScR, Preferred Reporting Items for Systematic Reviews and Meta-Analyses for Scoping Reviews.

The remaining 92 articles were eligible for full-text screening. A total of 25 studies met the inclusion criteria outlined in table 1. The study characteristics of the included studies are outlined in Table 3 (online supplemental file 4). The remaining articles (n=67) were excluded for several reasons: duplicate (n=4); wrong context, primarily calculating emissions for clinical care activities rather than research activities (n=21); wrong study design (n=25) and wrong outcomes (n=17).

Backward citation searching

Backward citation searching was performed on the 25 included articles. This process involves searching the reference lists of each included article to identify additional articles of interest. After screening titles, seven studies were deemed of interest for further screening. All seven were then excluded after full-text screening, as they did not meet the eligibility criteria.

Forward citation searching

Forward citation searching was then performed on the 25 included articles. This process involves identifying studies that have cited an included article to identify additional articles of interest. After searching was performed, 331 articles were identified. After duplicates were removed (n=108), 223 remained. Title screening was performed, and 33 articles were deemed relevant for further investigation. Of the 33 articles, 10 were duplicates of the original included articles. One additional study was a duplicate of a previously excluded article. The remaining 22 studies were included for full-text screening. All studies were then excluded as they did not meet the eligibility criteria.

Stakeholder consultation

After citation searching was completed, the research team decided that stakeholder consultation to identify additional articles was unnecessary.

Characteristics of included studies

All included studies were primary investigations. Of the included studies, the majority (68%, n=17) were retrospective analyses, mostly of clinical trials (88%, n=22). One study performed a retrospective analysis of research activities at a multiple sclerosis clinic.¹⁷ Another study performed a current analysis on an ongoing clinical trial.¹⁸

Nearly all studies (84%, n=21) included researchers from Europe, with the majority (64%, n=16) including researchers from the UK. The US (12%, n=3),^{17 19 20} Australia (8%, n=2)^{21 22} and Canada (16%, n=4)^{21 23–25} represented a small number of studies, while a separate Australian study examined the adaptation of their clinical trials programme due to the COVID-19 pandemic²⁶ and included researchers from a range of low- and middle-income countries (Bangladesh, Ethiopia, Indonesia, Nepal, Pakistan and Thailand).

The majority of studies (84%, n=21) were published between 2019 and 2025. Studies pre-2019 were all UK-based studies, which highlights a large early regional drive from the UK research community examining the environmental impact of conducting clinical research. Three studies from that period were well cited in the remaining literature (Burnett et al,²⁷ n=10; Lyle et al,²⁸ n=12; Subaiya et al,²⁹ n=8).

The overall aim of the studies varied, but the key aims were variations of measuring GHG emissions, environmental impact of research activities, the carbon footprint of clinical trials, primary drivers affecting emissions, hot spots for emissions, carbon dioxide equivalents (CO₂e) for activities and carbon reduction strategies.

Tools and methods of measurement

Table 4 (online supplemental file 5) summarises the tools and methods used across the included studies to quantify carbon emissions associated with research studies. A wide range of resources were identified. Most studies (72%, n=18) used more than one tool or data source to estimate emissions. Five studies (20%)^28–32 used four different tools or methods, while five (20%) additional studies used three tools or methods each.^{20 22 33–35}

The tools and methods represented a diverse set of resources used to calculate environmental impacts in clinical research. These resources varied across several key dimensions, including resource type, domain coverage, intended user base, maturity and accessibility.

Resource type categories

Nine categories of tools and resources were identified:

Databases were among the most comprehensive and data-intensive resources. Tools like Exiobase,³⁴ SimaPro,^{20 23 30 31} Ecoinvent, ^{20 23 30 31 33 36 37}ABPI (Association of the British Pharmaceutical Industry),^{19 20} ADEME (Agence de la transition écologique),^{37 38} NegaOctet³⁰ and HealthcareLCA (life cycle assessment)³¹ provide extensive reference data spanning countries, industries, emissions categories, systems and life cycle stages. These databases form the foundational infrastructure that supports more specialised calculators and assessment methods.

Online calculators were widely used to estimate emissions. The Paper Calculator,³⁹ SHC Care Pathway Calculator,^{33 36} Online Carbon Calculator,¹⁸ Climate Impact Tool,²⁸ VCCAT (virtual care carbon accounting tool)²⁴ and distance calculators (Webflyer, World Airport Codes)²⁹ enable quick assessments without requiring technical expertise in LCA methodology.

Online resources provided supporting information for emissions calculation. For example, Conservatree³⁹ offers paper-to-tree conversion estimates, while Google Maps^{22 28} was used to estimate travel distance.

International environmental standards, including ISO 14040 and 14044,^{19 20 34} were used to guide the development of methodological frameworks for LCAs.

Government resources provided region-specific emission factors, conversion tables and statistical baselines essential for accurate carbon footprinting and activity data. Examples include UK departments (DEFRA (Department for Environment, Food and Rural Affairs),^{21 28 32} Department of Transport, BERR (Department for Business, Enterprise and Regulatory Reform),²⁹ Australia’s Bureau of Statistics,²² Singapore’s Department of Statistics,²³ European Eurostat¹⁹ and South Africa’s Department of Forestry, Fisheries and the Environment³⁵). Resources cover various categories, including travel, energy, waste, material production and consumption.

Report and guideline resources, including the GHG Protocol,^27–29 Care Pathways Guidance,³⁴ IPCC (Intergovernmental Panel on Climate Change) reports,^{19 20 27 29} National Health Service (NHS)²⁸ and the University of Cape Town’s Carbon Footprinting Report,³⁵ were used to guide broad reporting and categorisation of emissions.

Research articles reported topic-specific emissions, including from dermatological surgery,⁴⁰ hospital waste streams,⁴¹ landfill emissions⁴² and NHS-related patient travel⁴³.

Consultancy resources were uncommon and focused on metrics related to area and energy consumption of office spaces.²⁸

Finally, several studies used life cycle impact assessment methods. These approaches allow environmental impacts to be quantified across multiple impact categories, including climate change, material production and consumption.^{23 30}

Main domains

The scope and focus of domains varied considerably across tools. Large databases such as Exiobase,³⁴ SimaPro,^{20 23 30 31} Ecoinvent^{20 23 30 31 33 36 37} and HealthcareLCA³¹ covered multiple sectors and life-cycle stages. ADEME^{37 38} and ABPI^{19 20} covered direct and indirect product categories, while NegaOctet³⁰ covered digital systems/services. This multi-domain structure enables more rounded assessment but can require significant expertise to navigate.

Healthcare-specific tools used domains based on clinical activities and impact categories related to healthcare fields. This structure allows users to estimate emissions associated with specific healthcare research pathways.

Online calculators were typically organised around activity categories. The Climate Impact Tool includes domains such as flights, road transport, energy and accommodation,²⁸ while the Online Carbon Calculator categorises emissions across household energy use, travel and consumption.¹⁸ The VCCAT tool focuses on virtual care emissions across visit and travel categories.²⁴

Governmental resources highlighted geographic and sectoral specialisation. Geographic specificity provides accuracy for regional studies but limits transferability unless data regarding emission factors are aligned. To note, the use of these resources was specific to the researchers and institutions applying them. As such, the focus of each resource was applied with location restrictions based on each of the studies’ objectives.

Sub-domains

The level of detail in sub-domains reveals different design approaches. Highly granular technical databases like Ecoinvent list ten major categories (see Table 4, online supplemental file 5) with further subdivision into resources, emissions, impact categories and specific indicators.^{20 23 30 31 33 36 37} This granularity supports precise assessments but creates steep learning curves for new users.

Detailed operational sub-domains appear in the SHC Care Pathway Calculator.^{33 36} This operational detail enables accurate modelling of specific healthcare research pathways.

Impact-focused sub-domains characterise tools like the Paper Calculator.³⁹ This impact-focused structure helps users understand effects across multiple environmental dimensions.

Simpler tools, such as Google Maps, report minimal sub-domain specificity.^{22 28} While distance calculators (Webflyer, World Airport Codes) have no meaningful sub-domains.²⁹ This simplicity reduces barriers for use but limits deeper analysis.

Intended user base

Target audiences varied from highly specialised technical users to the general public. Technical specialists in LCA represented the primary users of databases (Exiobase,³⁴ SimaPro,^{20 23 30 31} Ecoinvent,^{20 23 30 31 33 36 37} ADEME^{37 38}) and international standards (ISO 14040/14044),^{19 20 34} engaging in large-scale carbon accounting and formal LCAs, requiring deep methodological knowledge.

Healthcare stakeholders form a distinct multidisciplinary user community, unified by healthcare context rather than technical expertise.^{33 36}

Sector-specific practitioners used specialised tools, such as pharmaceutical users employing ABPI for drug emissions calculations,^{19 20} paper users accessing the Paper Calculator and Conservatree³⁹ and digital service users employing NegaOctet,³⁰ while small businesses seeking carbon offsets turned to the Climate Impact Tool²⁸.

Online carbon calculators serve the general public and researchers without requiring specialised training, democratising carbon footprinting but potentially sacrificing precision.

Geographically defined users access region-specific government resources. UK,²⁸ Australian²² and Singaporean transport data users²³ each require local emission factors and activity patterns for accurate regional assessments.

Maturity

The maturity spectrum reveals both established infrastructure and emerging innovations. Established resources dominated the landscape, offering multiple versions/updates.

An established but static resource includes the ReCiPe (RIVM and Radboud University, CML, and PRé Consultants) method,^{23 31} which has not been updated since 2016, potentially reducing relevance as emission factors and methodologies evolve. Whereas novel resources represent recent innovations responding to growing sustainability demands. The NHS Carbon Footprinting Report²⁸ and Conservatree³⁹ introduce new frameworks and data sources that may not yet have appropriate updates applied.

Established frameworks with version ambiguity include tools like the Climate Impact Tool²⁸ and Online Carbon Calculator¹⁸. Specific version information is not reported, suggesting either continuous updating or the use of informal versioning practices.

Access

Freely accessible resources predominate in government data (UK,²⁸ Australian,²² Singaporean,²³ European²⁰ and South African³⁵), online calculators (Paper Calculator,³⁹ SHC Care Pathway Calculator,^{33 36} Online Carbon Calculator,¹⁸ distance calculators,²⁹ VCCAT²⁴), online resources (Conservatree,³⁹ Google Maps^{22 28}), standard-based methods (Impact World+,²³ ReCiPe,^{23 31} PEF (product environmental footprint)³⁰) and guidance documents (GHG Protocol,^27–29 Care Pathways Guidance,³⁴ IPCC reports,^{19 20 27 29} NHS guidance²⁸). This open access model supports widespread adoption and democratisation for environmental assessment.

Licensed databases create a two-tier system. SimaPro requires licensed plans without further specification.^{20 23 30 31} Ecoinvent offers licensed plans but provides guest users basic access to documentation for datasets, which creates a model that allows exploration before purchase.^{20 23 30 31 33 36 37} ABPI remains free despite its pharmaceutical specialisation, which may reflect industry commitment to transparency.^{19 20} World Airport Codes calculator offers additional data with a one-time fee; however, this does not impact access to distance calculations.²⁹

Paid international standards position ISO 14040 and 14044 behind paywalls, though previews allow some access.^{19 20 34} This creates potential barriers to methodological consistency, as users may struggle to access definitive, rigorous guidance.

Consultancy fees restrict access to Carbon Trust resources, creating barriers and limiting transparency around methodologies and data sources.²⁸

Units of measurement

Table 5 (online supplemental file 6) details units of measurement across all tools/methods. CO₂e was the dominant unit, enabling different GHGs to be expressed using a common metric. Commonly reported scales included grams (gCO₂e), kilograms (kgCO₂e) and tonnes (tCO₂e), with some databases reporting megatonnes (MtCO₂e) and gigatonnes (GtCO₂e).

Global warming potential over a 100-year time horizon was the standard conversion methodology. Emissions were reported relative to functional units (per km travelled, per kWh, per monetary unit or per unit of production), with the use of mg for pharmaceuticals and ft²/m² for area.

Online calculators offered varied unit inputs, converting results into a singular standard unit (kgCO₂e, tCO₂e, MtCO₂e).^{18 28 33 36} While widespread CO₂e use supports comparability, variation in scale and functional unit limits cross-study comparison, highlighting the need for standardised reporting.

Activities measured

Table 6 (online supplemental file 7) outlines activities measured across studies. Some provide a comprehensive breakdown, while others offer a partial assessment.

Travel was the most consistently measured activity (76%, n=19), encompassing patient travel, staff commuting, delivery of equipment, consumables and drugs and site-related activities, such as monitoring visits.

Centre-specific contributions, including energy and fuel use and general utilities were measured. Study-specific activities such as data management, laboratory sample processing and drug product management were also reported.

Discussion

This review mapped the tools, methods and reporting practices used to quantify carbon emissions associated with clinical research. The findings demonstrate a fragmented methodological landscape characterised by variation in tools, reporting approaches and measurement units. Table 2 summarises the findings for each study aim.

Table 2. Summary of study aims.

1	Travel seems to offer a relatively easy measurement and emerged as the most consistently measured activity. Other contributors (energy, utility use, data management and laboratory processing) were captured less consistently across studies, with some emission data lacking.
2	Many studies used several tools to quantify emissions for various activities. The range of resources varied considerably in type, domain coverage, target user, maturity and accessibility, reflecting a fragmented landscape rather than a standardised approach to quantify emissions.
3	Domains vary in both scope and granularity. Large multi-sector databases cover the full life cycle of activities, healthcare-specific tools organise emissions around clinical pathways, online calculators focus on everyday activity categories and government resources offer geographic and sectoral precision. The more specific a tool, the seemingly greater the complexity for use.
4	While kgCO₂e is the near-universal unit for expressing emissions, the diversity in scale and functional units (e.g., per km, per kWh, per monetary unit) makes comparison across studies challenging, underscoring the need for standardised reporting.

Cross patterns

Several patterns emerged across these dimensions. There is an inverse relationship between comprehensiveness and accessibility. The most comprehensive databases (SimaPro,^{20 23 30 31} Ecoinvent)^{20 23 30 31 33 36 37} require licences and technical expertise, while simpler calculators remain free and user-friendly. This creates a methodological trade-off between rigour and practicality. Additionally, there is sector specialisation versus transferability, whereby healthcare-specific tools (SHC Care Pathway Calculator,^{33 36} Care Pathways Guidance³⁴) offer immediately relevant criteria but limited applicability beyond healthcare, while generic tools require users to adapt them to specific contexts but support a broader spectrum of use. Another key pattern is geographic specificity versus global applicability. Government resources provide accurate regional data but do not necessarily transfer across borders, while international databases and standards enable global studies but may lack local precision.

There is a correlation between maturity of tools/methods and accessibility. There are many established resources that show mixed access models (free government data vs licensed commercial databases), while novel resources tend toward free access, possibly reflecting a range of origins, for example, academic versus commercial development. Finally, the expertise of users is a key pattern that emerged, whereby the diverse ecosystem of tools/methods supports multiple expertise levels, ranging from the public/lay persons (online calculators)^{18 24 28 36} through sector-specific professionals (healthcare, pharmaceutical, business) to technical assessment specialists (LCA practitioners), with the complexity of tools being scaled accordingly.

Implications for research practice

This diverse ecosystem requires researchers to navigate trade-offs. For researchers that require comprehensive, rigorous assessments that align with established peer-reviewed standards, investment in licensed databases (SimaPro,^{20 23 30 31} Ecoinvent^{20 23 30 31 33 36 37}), application of formal international standards (ISO 14040/14044),^{19 20 34} and employment of established life cycle impact assessment (LCIA) methods (Impact World+,²³ ReCiPe,^{23 31} PEF method³⁰) are recommended. Applying the above approaches will maximise the credibility and comparability of analyses and results but requires expertise, time, financial resources and likely training.

For researchers that require rapid, practical assessments to inform decision-making, using online calculators combined with relevant government emission factor databases can provide actionable estimates without extensive training or cost. This approach prioritises accessibility and speed over precision. However, the reliability and rigour of using government databases can lead to a stronger result.

For researchers that require healthcare-specific applications, specialised tools and frameworks (SHC Care Pathway Calculator,^{33 36} Care Pathways Guidance,³⁴ NHS Carbon Footprinting Report,²⁸ HealthcareLCA³¹) offer contextualised criteria that improve relevance and reduce adaptation requirements compared with more generic approaches.

For researchers that require regional accuracy, combining international databases with local government resources (DEFRA for the UK,^{21 28 32} Australian statistics,²² Singapore data,²³ South Africa data³⁵) balances using a recognised global methodology with geographic precision.

The choice among these resources depends heavily on research objectives, available expertise, budget constraints, desired rigour and contextual considerations (i.e., specific sectors or geographies). No single tool optimally serves all aforementioned purposes. To conduct an effective environmental impact assessment requires strategic selection and often integration across multiple resource types. This review explores just some of the tools and methods available in a small selection of literature.

Implications for policy and research funders

As research funders and institutions increasingly prioritise sustainability, consistent approaches for measuring research-related emissions will become essential. The lack of standardised approaches identified in this review suggests an urgent need for consensus guidance and reporting standards for research-related emissions. Incorporating environmental impact assessment into funding requirements, ethics applications and trial design guidance may accelerate the adoption of lower-carbon research practices.

Limitations

A comprehensive search strategy was employed that included a wide variety of databases, as well as a descendancy and ascendancy approach to ensure a rigorous search. While there was variability in study location, only English-language studies were included in the review, which may introduce geographic bias. The field is emerging rapidly, and new literature is in development. To address this, an updated search was performed in June 2026 prior to publication to ensure the most current results. Due to the nature of the review, methodological quality was not appraised. Similarly, while the results were extensive, there was strong heterogeneity which limited cross-study comparison.

Conclusion

This scoping review demonstrates that while a wide range of tools and methods exist to quantify the carbon footprint of clinical research, the methodological landscape remains fragmented and lacks standardisation. Researchers must currently navigate trade-offs between methodological rigour, accessibility and contextual relevance when selecting approaches to measure emissions.

Developing consensus guidance and standardised reporting frameworks will be critical to support consistent measurement and meaningful reduction of research-related emissions. As clinical research continues to expand globally, integrating environmental sustainability into research design and funding priorities is increasingly urgent.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.