Introduction
Among broader pollution impacts, healthcare contributes between 4 percent and 8.5 percent of global greenhouse gas (GHG) emissions (Reference Romanello, Walawender and Hsu1). Of healthcare’s GHG emissions and contribution to pollution derived from the life cycle of materials, 79 percent are linked to suppliers and value chains, which are made up predominantly of health technologies (pharmaceuticals, medical technologies, AI, digital health technologies, vaccines, rehabilitation technologies, diagnostic tests and tools, and surgery) (2). For example, health technologies contribute to the ubiquitous use of virgin fossil fuel plastics, pharmaceutical pollution in water systems, and the use of finite raw materials embedded in digital interventions (Reference Thornber, Bentham and Pfleger3;Reference Di Bidino, Dutta Majumdar, Pegg, Mahon, Papavero and Mueller4). This challenge is not confined locally, as a proportion of this waste enters global waste streams and may ultimately be exported to lower-resource countries with limited capacity for safe processing (5).
Developing more sustainable use of technologies across their life cycle, public health interventions, and system changes can support both healthcare sustainability and the capability to be more resilient and adaptable. This is key to ensuring current and future populations can access healthcare that is resilient and adaptable (6). Health technology assessment (HTA) policy making and guidance can play a pivotal role in this endeavor (Reference Holtorf, Pegg, Mueller and McMeekin7).
Policy momentum is growing to align health systems with the Paris Agreement and net-zero goals (8;9). International initiatives include the Lancet Countdown, Lancet MedZero Commission, and WHO Alliance for Transformative Action on Climate and Health, through which more than 38 countries have committed to net-zero, climate-resilient health systems (Reference Romanello, Walawender and Hsu1;10;11). National strategies include England’s NHS net-zero commitments and the Netherlands’ Green Deal on Sustainable Health Care (12;13). Nevertheless, no HTA agencies currently require environmental evidence within standardized evaluation.
A planetary health framing is increasingly relevant because health technologies depend on plastics, chemicals, and active pharmaceutical ingredients. These “novel entities” are a critical planetary-boundary concern because their production, persistence, and release exceed global capacity for monitoring and safety assessment (Reference Persson L, Collins and Cornell14;Reference Gupta, Bai and Liverman15). Their presence in people, food, and water contributes to ecosystem disruption and contamination of critical resources, increasing risks to water and food security and health-system resilience (Reference Persson L, Collins and Cornell14;Reference Dobrzyńska, Wasilewski and Pośniak16).
Therefore, it is important to look at developing pragmatic, acceptable, and feasible approaches and methods to incorporate environmental impacts into HTA. A revised definition of HTA that formally integrates environmental impact as a core domain of a technology’s life cycle value has been published, and approaches and methods of integrating planetary health measures into HTA are currently being discussed and developed (Reference Holtorf, Pegg, Mueller and McMeekin7;11;Reference O’Rourke, Oortwijn and Schuller17–20). A critical systems-level approach is required to capture the downstream and long-term consequences of technologies, enabling HTA to move beyond narrow, carbon-focused assessments, and better account for cumulative, irreversible impacts on biodiversity and ecosystem stability that underpin human health.
Adapting HTA to include environmental impact will require integrating multidisciplinary approaches to support decision making and address challenges such as trade-offs, capacity constraints, limited data availability, and gaps in environmental management knowledge (Reference Holtorf, Pegg, Mueller and McMeekin7;Reference Pegg, Gregg, Bishop and Bouttell21). Several scoping reviews have identified methods to adapt HTA from disciplines beyond health (Reference Antonazzo, Gribaudo and Ye22–Reference Williams, Bell, Morton and Dieng24). However, environmental analysis within HTA is further complicated by the perspective taken by the HTA agency, which may include a broader clinical, economic, ethical, and population context. Previous reviews have highlighted these challenges; for example, the absence of a shared methodological framework, lack of disaggregated data for individual technologies, and difficulties conducting environmental life cycle assessments (eLCAs) that consider a holistic view of the environmental effects of a technology (Reference Antonazzo, Gribaudo and Ye22).
These methodological challenges create a need to examine approaches from multiple disciplines. This article identifies and compares methods that could support assessment of the environmental impacts of health technologies and provides an exploratory ranking of their potential applicability to HTA as a basis for future methodological evaluation.
Multidisciplinary methods for incorporating environmental impacts into HTA
Identification and synthesis
A scoping review was conducted to identify published studies and grey literature sourced from HTA websites that applied frameworks to compare the environmental impacts of technologies across various broad disciplines (Reference Arksey and O’Malley25). The review question and methods were outlined in a protocol prior to commencing the review. Eligible studies included full-text published studies, environmental assessment frameworks, or mixed-methods studies. Further details on the scoping review methods are provided in the Supplementary Material.
The following information was extracted from the frameworks outlined within the included studies:
-
• The methods used to generate qualitative and quantitative data, including any methods to convert qualitative to quantitative data and, therefore, methods that might use a combination of both data types.
-
• The comprehensiveness of the frameworks to evaluate environmental impacts.
-
• The feasibility and transparency of the frameworks.
-
• Methods to support environmental scope and boundary evaluation.
-
• Methods to support multi-criteria decision making.
-
• Methods to improve environmental data analysis.
A formal quality assessment was not performed as it was deemed not relevant for this type of research. A narrative synthesis was performed in which the included studies and their methods were summarized.
To facilitate comparison across the diverse frameworks identified, an exploratory comparative assessment was undertaken to provide a heuristic synthesis of the literature rather than a validated assessment framework. The intention was to highlight the relative characteristics of each framework, identify promising directions for future research, and stimulate discussion regarding the incorporation of environmental impacts into HTA, rather than to establish a definitive hierarchy of methods. Drawing on five interconnected dimensions of sustainability – economic, human health, ecological, social, and animal welfare – each framework was assessed by a single researcher for its transparency, feasibility, and coverage of these dimensions. The selection of transparency and feasibility as appraisal criteria was informed by established principles for the conduct of HTA (Reference Drummond, Schwartz and Jönsson26). Indicative scores ranging from 0 to 1 were assigned to support an exploratory ranking of the frameworks’ potential applicability to HTA.
Overview of included studies
After removing duplicates, the search yielded 1,465 unique records. Of these, seventy-five full texts were assessed, and ten studies were included in the final review. A PRISMA flow diagram illustrating the study selection process is presented in the Supplementary Material.
The ten included studies highlighted a range of methods for incorporating environmental impacts and developing more sustainable assessments of technologies. In four of these studies, eLCA was a core methodology, often combined with complementary methods to overcome methodological challenges(Reference Paris, Falkenberg, Noethlings, Heinzel, Borgemeister and Escobar27)(Reference Onat, Kucukvar, Tatari and Zheng28)(Reference Vukelic, Budak, Tadic, Simunovic, Kljajic and Agarski29)(Reference Cappuyns and Kessen30). Alternative methods were identified across the remaining six studies (Reference Lukman, Brglez and Krajnc31)(Reference Chung and Kim32)(Reference Carolus, Hanley, Olsen and Pedersen33)(Reference Lomas, Schmitt and Jones34)(Reference Cubero Dudinskaya, Naspetti and Arsenos35)(Reference Elder, Phillips, Potter, Frumhoff and Rogers36).
Key frameworks and methods
There were six key frameworks outlined by the included studies: eLCA and enriched life cycle assessments (LCAs), economic-based approaches, resource optimization and localized assessment, health and public perception, multi-criteria decision analysis, and risk-based models. Across the 10 included studies, 4 reported on multiple frameworks. Full details of the methods and frameworks identified are provided in Tables 1 and 2.
Methods adopted in identified studies
Table 1. Long description
The table consists of four columns: Study, Methods supporting environmental scope and boundary setting, Methods supporting multiple criteria decision making (M C D A), and Methods supporting environmental data analysis.
Key entries include:
* One Health framework: Uses extended e L C A for life cycle impacts and SimaPro for midpoint impacts like climate change. No formal M C D A applied.
* Circular economy framework: Uses 9 R strategy for boundaries and A H P for weighting sustainability indicators.
* M C D A and T O P S I S for wastewater: Uses D P S I R framework for indicators and T O P S I S to rank alternatives.
* Knee support technology: Combines cradle-to-gate e L C A with P R O M E T H E E M C D A for stakeholder preferences.
* Brownfield remediation: Compares R E C and ReCiPe L C I A frameworks using expert weighting.
* Bottom-up C B A: Uses stakeholder engagement for scope and a deliberative process for decision-making.
* Health impact assessment: Based on N I C E principles using P M 2.5 and N O sub 2 exposure data. No M C D A applied.
* Consumer preferences (D C E): Attributes selected via literature review; uses choice modeling and econometric models.
* Passenger cars (L C S A): Integrates environmental, economic, and social criteria using Compromise Programming for optimization.
* Fire management: Focuses on carbon emissions and future projection scenarios without M C D A.
Abbreviations: AHP, analytical hierarchy process; CBA, cost–benefit analysis; DALY, disability-adjusted life year; DCE, discrete choice experiment; DPSIR, drivers, pressures, state, impact, and response; eLCA, environmental life cycle assessment; LCIA, life cycle impact assessment; LCSA, life cycle sustainability assessment; MCDA, multi-criteria decision analysis; NICE, National Institute for Health and Care Excellence; PROMETHEE, Preference Ranking Organization METHod for Enrichment Evaluation; QALY, quality-adjusted life year; REC, risk reduction, environmental merit, and cost; TOPSIS, Technique for Order of Preference by Similarity to Ideal Solution.
Summary of frameworks outlined by the included studies
Table 2. Long description
The table consists of two columns: Key study/method and Explanation of method, data inputs/outputs and potential relevance for H T A.
* Row 1: Life-cycle sustainability assessment and multi-objective optimization. Purpose: Identifies optimal distribution of alternative passenger vehicles. Inputs: L C A indicators for seven vehicle types and charging scenarios. Outputs: Optimal vehicle distributions. Relevance: Shows optimization across environmental, economic, and social objectives.
* Row 2: Extended L C A using a One Health framework. Purpose: Integrates environmental, human health, and animal welfare impacts. Inputs: Dietary consumption, food production, and disease risk. Outputs: Environmental impacts and human health losses. Relevance: Captures interconnected health and sustainability outcomes.
* Row 3: Integrated L C A–M C D A using IMPACT 2002+, PROMETHEE II and GAIA. Purpose: Compares alternative knee supports. Inputs: IMPACT 2002+ results and product performance. Outputs: Product ranking and trade-off visualization. Relevance: Combines environmental evidence with other H T A criteria.
* Row 4: R E C and ReCiPe L C A-based assessment tools. Purpose: Compares brownfield-remediation technologies. Inputs: Resource use, energy, emissions, and cost. Outputs: Single environmental scores. Relevance: Illustrates trade-offs between simple screening and data-intensive assessment.
* Row 5: Cost–benefit optimization using the social cost of carbon. Purpose: Identifies fire-management strategies to minimize climate damages. Inputs: Burned area, expenditure, and carbon emissions. Outputs: Projected management costs and carbon damages. Relevance: Monetizes carbon impacts for optimization.
* Row 6: Circular-economy assessment using the 9R framework and A H P. Purpose: Assesses progress toward circularity. Inputs: 9R hierarchy indicators and A H P weights. Outputs: Overall circularity values. Relevance: Framework for assessing reuse, repair, and recycling across life cycles.
* Row 7: Bottom-up environmental cost–benefit analysis 2018. Purpose: Evaluates solutions identified by stakeholders. Inputs: Stakeholder options and local knowledge. Outputs: Monetized net benefits. Relevance: Supports stakeholder-led valuation of environmental consequences.
* Row 8: Pharmacoeconomic evaluation of air-quality interventions. Purpose: Translates air quality health effects into health-economic outcomes. Inputs: Health-impact assessment and mortality data. Outputs: Q A L Y s gained and N H S cost savings. Relevance: Incorporates environment-related health effects into conventional H T A.
* Row 9: Discrete choice experiment. Purpose: Elicits consumer preferences for red meat attributes. Inputs: Hypothetical choices from 2,900 consumers. Outputs: Attribute-preference estimates and willingness to pay. Relevance: Quantifies public preferences for environmental and ethical attributes.
* Row 10: Fuzzy TOPSIS and decision-making under complete uncertainty. Purpose: Prioritizes locations for treated wastewater use. Inputs: Hydrological simulations and climate scenarios. Outputs: Rankings of locations and robustness assessment. Relevance: Supports prioritization under uncertain environmental conditions.
Abbreviations: AHP, analytical hierarchy process; HTA, health technology assessment; LCA, life cycle assessment; MCDA, multi-criteria decision analysis; NHS, National Health Service; PROMETHEE, Preference Ranking Organization METHod for Enrichment Evaluation; QALY, quality-adjusted life year; REC, risk reduction, environmental merit, and cost; TOPSIS, Technique for Order of Preference by Similarity to Ideal Solution.
LCA and its enhancements
Four studies reported on eLCAs. Of these, two utilized an enriched eLCA to conduct a broad value assessment beyond environmental impacts (Reference Paris, Falkenberg, Noethlings, Heinzel, Borgemeister and Escobar27;Reference Onat, Kucukvar, Tatari and Zheng28). One of these studies included a One Health approach, facilitating an extended eLCA (Reference Paris, Falkenberg, Noethlings, Heinzel, Borgemeister and Escobar27). While comprehensive in health and environmental impacts, it notably did not consider financial impact, which limits its sustainability approach. The other framework was a Life Cycle Sustainability Assessment (LCSA), designed to assess financial, social, and environmental impacts (Reference Onat, Kucukvar, Tatari and Zheng28). The remaining three studies utilized SimaPro eLCA software, which provides a centralized, up-to-date data repository and supports the simplification of environmental modeling (Reference Paris, Falkenberg, Noethlings, Heinzel, Borgemeister and Escobar27;Reference Vukelic, Budak, Tadic, Simunovic, Kljajic and Agarski29;Reference Cappuyns and Kessen30). This eLCA software was found useful for overcoming data, scope, boundary, and resource constraints often highlighted as key challenges to incorporating environmental impacts into HTA.
Economic-based approaches
Two studies reported on economic-based approaches (Reference Onat, Kucukvar, Tatari and Zheng28;Reference Elder, Phillips, Potter, Frumhoff and Rogers36). One study applied an economic input–output (EIO) eLCA, which converts financial flows to environmental impact, to report emission-equivalent values. This approach supports data acquisition and increases the accessibility of environmental assessment in HTA (Reference Onat, Kucukvar, Tatari and Zheng28). Additionally, another study used the societal cost of carbon (SCC) approach to generate a financial cost from carbon dioxide equivalent (CO2e) emissions (Reference Elder, Phillips, Potter, Frumhoff and Rogers36). This can help monetize the environmental impacts of these emissions, which may be useful for certain HTA frameworks.
Resource optimization and localized assessment
Two studies considered resource optimization and localized assessments (Reference Lukman, Brglez and Krajnc31;Reference Carolus, Hanley, Olsen and Pedersen33). One study identified a circular economy framework, which prioritized optimizing resource use directly proportional to environmental impact (Reference Lukman, Brglez and Krajnc31). Moreover, another study discussed an environmental cost–benefit analysis (CBA), which focused on localized environmental assessment (Reference Carolus, Hanley, Olsen and Pedersen33). This approach involves a multitude of local factors and uses baseline data to define scope and boundary.
Health and public perception
Two studies reported on frameworks that considered health and public perception (Reference Lomas, Schmitt and Jones34;Reference Cubero Dudinskaya, Naspetti and Arsenos35). One employed a health impact assessment (HIA), which utilized quality-adjusted life years (QALYs) as an environmental outcome, specifically for air quality interventions (Reference Lomas, Schmitt and Jones34). This method defined specific health events caused by air pollutants to calculate QALYs, looking at long-term health effects from exposure to particulate matter < 2.5 micrometers in diameter (PM2.5) and nitrogen dioxide. Another study used discrete choice experiment (DCE) methodology to assess the public’s willingness to pay (WTP) for an intervention, selecting credence attributes based on literature and qualitative studies (Reference Cubero Dudinskaya, Naspetti and Arsenos35).
Multi-criteria decision analysis (MCDA)
MCDA methods, such as the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) and a Preference Ranking Organization METHod for Enrichment Evaluation (PROMETHEE), were used in four studies (Reference Paris, Falkenberg, Noethlings, Heinzel, Borgemeister and Escobar27;Reference Chung and Kim32;Reference Vukelic, Budak, Tadic, Simunovic, Kljajic and Agarski29;Reference Onat, Kucukvar, Tatari and Zheng28). TOPSIS ranks options based on their distance from an ideal and worst-case solution, while PROMETHEE uses pairwise comparisons and preference functions to model the strength of preference between options, offering greater flexibility but increased complexity. These analytical models support decision making, allow for weighting of criteria, and can handle data limitations using fuzzy or crisp numbers.
Risk-based models
In one study, the risk reduction, environmental merit, and cost (REC) framework was compared to the ReCiPe life cycle impact assessment (LCIA) framework to evaluate brownfield remediation options, with weighting determined by an expert panel (Reference Cappuyns and Kessen30). Risk assessment was a dominant criterion in this framework.
Comprehensiveness and feasibility assessment (HTA suitability ranking)
The frameworks were scored by the researcher (zero to one) and ranked on their overall potential suitability for HTA based on transparency, feasibility, and sustainability (incorporating financial, social, human health, environmental health, and animal health domains). Table 3 presents the different frameworks and their rankings for potential HTA suitability.
Framework ranking in order of comprehensiveness for HTA
Table 3. Long description
The table consists of two columns: Rank and Study.
* Rank 1: One health extended e L C A framework.
* Rank 2: Circular economy framework.
* Rank 3: T O P S I S.
* Rank 4: P R O M E T H E E.
* Rank 5: R E C and ReCiPe life cycle impact assessment models.
* Rank 6: Cost–benefit analysis.
* Rank 7: Health impact assessment.
* Rank 8: Discrete choice experiment.
* Rank 9: Life cycle sustainability assessment using extended input–output.
* Rank 10: Societal cost of carbon.
Abbreviations provided in the footer include: e L C A for environmental life cycle assessment; P R O M E T H E E for Preference Ranking Organization METHod for Enrichment Evaluation; R E C for risk reduction, environmental merit, and cost; and T O P S I S for Technique for Order of Preference by Similarity to Ideal Solution.
Abbreviations: eLCA, environmental life cycle assessment; PROMETHEE, Preference Ranking Organization METHod for Enrichment Evaluation; REC, risk reduction, environmental merit, and cost; TOPSIS, Technique for Order of Preference by Similarity to Ideal Solution.
Discussion of rankings
The highest-ranked framework was the One Health extended eLCA framework, suggesting this framework has the most potential within HTA (Reference Paris, Falkenberg, Noethlings, Heinzel, Borgemeister and Escobar27). This model’s strength lies in capturing complex interdependencies between humans, animals, and the environment, providing cross-sector benefits for sustainable technology and global health security. While a One Health perspective is often adopted due to its clearer operational focus on the interconnectedness of human, animal, and environmental health, a planetary health approach is more holistic in scope, encompassing broader ecological and socioeconomic systems (Reference Lerner and Berg37;Reference Morand38). However, a planetary health approach is consequently more challenging to define and measure within evaluation frameworks.
The second-ranked framework was the circular economy framework (Reference Lukman, Brglez and Krajnc31). This model effectively supports the reduction of environmental harm from health technologies, while simultaneously optimizing health financing by focusing on sustainable supply chains and prolonging technology use.
The mid-ranked frameworks (third to fifth) were TOPSIS, PROMETHEE, REC, and ReCiPe models (Reference Chung and Kim32–Reference Cappuyns and Kessen30). TOPSIS and PROMETHEE are MCDA methods that have been identified as useful and feasible to support the ranking of alternative methods across multiple criteria, including cost-effectiveness and environmental impact in HTA, while explicitly enabling trade-offs between them. The REC and ReCiPe models were also highlighted as being feasible in terms of the data required to assess the technology and transparency in their methods.
The lower-ranked frameworks (sixth to tenth) were CBA, HIA, DCE, LCSA using EIO, and SCC, suggesting these are the least applicable for use within HTA (Reference Carolus, Hanley, Olsen and Pedersen33–Reference Elder, Phillips, Potter, Frumhoff and Rogers36). CBA was ranked sixth because, although it is an established method, the monetizing of environmental benefits can be controversial, raising questions about its feasibility and transparency in HTA (Reference Carolus, Hanley, Olsen and Pedersen33). In seventh place is HIA, which is useful for air quality outcomes linked to CO2e emissions but applies QALYs to measure environmental impact utility. The DCE model is ranked eighth because, despite its comprehensive capture of sustainability domains, it may pose challenges by forcing respondents to make complex, unfamiliar, and ethically loaded trade-offs across fundamentally different domains (health, money, and environment) (Reference Cubero Dudinskaya, Naspetti and Arsenos35). This can lead to cognitive overload, unstable and framing-sensitive preferences, ethical resistance to trade-offs, and challenges in integrating results into HTA without inconsistency and double counting (Reference Cubero Dudinskaya, Naspetti and Arsenos35).
LCSA using EIO was ranked ninth because, while comprehensive and able to simplify environmental evaluation into a single metric, calculating carbon emissions generated through converting financial costs is likely to underestimate true environmental impacts. This is due to the focus the method places on financial proxies (Reference Onat, Kucukvar, Tatari and Zheng28). Finally, the SCC model was ranked lowest (Reference Elder, Phillips, Potter, Frumhoff and Rogers36). Despite the framework’s usefulness in signaling society’s WTP to avoid future carbon damage, its scope is limited to carbon measurement, potentially underestimating broader environmental impacts. Moreover, SCC relies on a different economic modeling approach compared with the QALY framework (approach). This means its values are often not directly compatible with cost-effectiveness analysis frameworks in HTA. The SCC model has also faced criticisms for political drivers and ambiguity.
Discussion
This article provides an overview and critical assessment of recent literature applying multidisciplinary methods to integrate environmental impacts into HTA. The research aligns and expands on the findings from other scoping reviews, highlighting methods such as eLCA, MCDA, choice experiments, and risk assessment-based frameworks as potentially useful for HTA (Reference Pegg, Gregg, Bishop and Bouttell21–Reference Pinho-Gomes, Yoo, Allen, Maiden, Shah and Toolan23). A key distinction of this article is its emphasis on the advantages of using a combination of methods and software from across multiple sectors to strengthen the consideration of environmental evidence in HTA. Moreover, the current research provides an overview of methods that underpin the comprehensive assessment of environmental impact in HTA, which was raised by Pinho-Gomes et al. (Reference Pinho-Gomes, Yoo, Allen, Maiden, Shah and Toolan23) as a vital element to consider.
Table 4 maps the methods identified in this review to five stages of incorporating environmental impacts into HTA and summarizes their potential applications and limitations. It is intended as an exploratory methodological map rather than a prescriptive set of recommendations: methods should be selected proportionately according to the decision context, available evidence, and purpose of the assessment. This approach aligns with the HTAi Environmental Sustainability in HTA Working Group’s agenda to evaluate methods, define scope, and develop feasible value frameworks for international piloting and implementation (39). The findings can therefore support future stakeholder deliberation and consensus-building without prescribing a single framework.
Potential contributions of each framework to the HTA process
Table 4. Long description
The table is organized into five columns: Stage in the H T A process, Key question and proposed activity, Methods or frameworks identified in the review, Potential application in H T A, and Key limitations or barriers to adoption.
* Stage 1: Define the scope and assessment boundary. Focuses on life-cycle stages and sustainability domains. Methods include One Health, extended e L C A, circular economy, L C S A, C B A, and D P S I R with T O P S I S. Limitations include reduced feasibility due to broad boundaries and data requirements.
* Stage 2: Identify environmental impacts qualitatively. Focuses on mapping benefits or harms across the care pathway. Methods include One Health, circular economy, D P S I R, C B A, and D C E. Limitations include potential omission of impacts due to limited evidence and unmanageable outcome numbers.
* Stage 3: Measure impacts and gather data. Focuses on quantifying impacts and selecting indicators. Methods include process-based e L C A, L C I A, H I A, D C E, and S C C modeling. Limitations include unavailability of primary life cycle data and sensitivity to modeling assumptions.
* Stage 4: Determine how environmental evidence will inform H T A. Focuses on reporting, weighting, and handling trade-offs. Methods include P R O M E T H E E, T O P S I S, R E C, ReCiPe, D C E, C B A, H I A, and S C C. Limitations include subjective weighting and the risk of double counting with Q A L Y estimates.
* Stage 5: Undertake the analysis and interpret results. Focuses on comparing alternatives and testing uncertainty. Methods involve combinations of e L C A, P R O M E T H E E, R E C, A H P, C B A, H I A, D C E, T O P S I S, L C S A, and S C C. Limitations include increased analytical complexity and limited transferability across different healthcare systems.
Abbreviations: AHP, analytical hierarchy process; CBA, cost–benefit analysis; DCE, discrete choice experiment; DPSIR, drivers, pressures, state, impact, and response; eLCA, environmental life cycle assessment; HIA, health impact assessment; HTA, health technology assessment; LCIA, life cycle impact assessment; LCSA, life cycle sustainability assessment; MCDA, multi-criteria decision analysis; PROMETHEE, Preference Ranking Organization METHod for Enrichment Evaluation; QALY, quality-adjusted life year; REC, risk reduction, environmental merit, and cost; SCC, societal cost of carbon; TOPSIS, Technique for Order of Preference by Similarity to Ideal Solution.
Previous research has established that eLCA can compare environmental impacts across health technologies. As an integral phase of LCA, LCIA translates life-cycle inventory data into environmental impact indicators; ReCiPe is an LCIA method implemented within software such as SimaPro (40;Reference Huijbregts, Steinmann and Elshout41). Such software supports structured environmental modeling, although results remain dependent on the underlying inventory data and assumptions (Reference Rizan42).
The findings of this research also align with support for eLCA’s wide applicability for comparing a broad set of environmental impacts beyond the assessment of GHG emissions, across the life cycle of health technologies. For example, PAS 2090 provides a complementary framework for use within HTA, by establishing harmonized Product Category Rules (PCRs) for pharmaceuticals. These are standardized guidelines that specify how pharmaceutical eLCAs are conducted and reported for this product category (health technology). The framework provides consistency, transparency, and comparability of environmental impact results that can further operationalize environmental sustainability considerations within HTA (43). Moreover, this article suggests that eLCA is most useful when combined with other approaches and frameworks to overcome challenges such as data availability, defining appropriate scope and boundary, and managing trade-offs between financial, environmental, and social criteria. This research aligns with the suggestion by the HTAi Environmental Sustainability in Health Technology Assessment (ESHTA) Working Group that a planetary health/One Health approach along with eLCA may be the most suitable framework for use within HTA (39). The absence of an explicitly planetary health-based framework was notable, although this may reflect its broader scope and the challenges of operationalizing planetary-scale ecological and socioeconomic interdependencies within technology assessment. One Health offers a more immediately applicable framework, particularly for pandemic prevention – while encompassing important elements of planetary health through its recognition of the interdependence of human, animal, and ecosystem health. This transdisciplinary and multisectoral collaboration is important for global health security and improved development outcomes, supporting healthcare’s role in preventing future pandemics (Reference Sinclair44). Also identified as a key framework to support sustainable development in HTA, a circular economy approach is suggested to be the second most comprehensive instrument, offering a powerful model for reducing environmental harm and optimizing financing functions, proving both feasible and transparent (Reference Saha, Farhanj and Kumar45).
MCDA methods may support the structured consideration of environmental impacts alongside clinical effectiveness, costs, technical performance, and ethical considerations (Reference Hwang, Yoon, Hwang, Yoon, Hwang and Yoon46;Reference Brans and Ph47). TOPSIS and PROMETHEE can make criteria weights and trade-offs explicit, and TOPSIS has recently been proposed for environmental assessment of digital health technologies (Reference Di Bidino, Dutta Majumdar, Pegg, Mahon, Papavero and Mueller4). However, MCDA remains difficult to reconcile with conventional QALY-based cost-effectiveness frameworks and decision thresholds (Reference Marsh, Sculpher, Caro and Tervonen48). Environmental criteria should therefore complement, rather than be represented through arbitrary changes to, cost-effectiveness thresholds: increases not grounded in opportunity cost may increase displaced health and do not inherently incorporate environmental sustainability (Reference Vallejo-Torres, Edney and Espinosa49).
Other environmental assessment methods identified by this research include DCE, CBA, HIA, and a REC model. A pilot study by Pegg et al. (Reference Pegg, Mistry, Barker and Taylor50) demonstrated the utility of DCE methods for eliciting how members of the public might decide on trade-offs between population health and environmental outcomes. The study revealed that UK respondents were willing to forgo human health gains to achieve environmental benefits such as reduced carbon emissions and improved biodiversity (Reference Cubero Dudinskaya, Naspetti and Arsenos35). Correspondingly, choice modeling is valuable for exploring multiple conflicting criteria in more sustainable decision making, helping to quantify preferences where direct health, environmental, and economic impacts intersect. However, its application in healthcare decision contexts presents several challenges, including respondents’ limited familiarity with technical environmental concepts and metrics, such as the effects of carbon emissions on climate change and other planetary boundaries; hypothetical bias; and the risk of double counting when integrating health economic outcomes with broader environmental externalities in HTA. While DCEs present challenges in incorporating environmental impacts within HTA, we propose that there may be value in building on previous studies using a similar trade-off – based approach, particularly through the inclusion of a broad range of informed stakeholders. Incorporating diverse expertise, from low- to high-resource settings and spanning policy, regulation, health economics, HTA, climate change, planetary health, public representatives, and health technology developers, would enable comparison of priorities across key decision makers. This can support consensus building and framework development, provide quantitative evidence to inform HTA methods, and help develop crucial context-specific solutions.
Data limitations are a recurring challenge in this field and can be addressed through methods such as EIO analysis and the use of hybrid primary and EIO data. Together, these methods facilitate a more holistic and consistent evaluation of environmental impacts within existing HTA structures, while supporting deliberative decision making under uncertainty. The findings of this research suggest that these multidisciplinary methods each offer a unique perspective on incorporating environmental impacts into HTA. Combining multiple approaches has the potential to create a more holistic and comprehensive picture for HTA. This may help stakeholders, including policy makers, clinicians, the public, and environmental experts, to better understand the impacts and help make informed decisions throughout a technology’s assessment and deployment. These multidisciplinary methods may also support capacity building by providing structured, transparent, and transferable approaches that enable HTA stakeholders, decision makers, and public contributors to develop the knowledge, skills, and shared understanding required to incorporate environmental impacts into future assessment frameworks (Reference Holtorf, Pegg, Mueller and McMeekin7;Reference Pegg, Gregg, Bishop and Bouttell21).
There are some limitations to this research which should be considered. First, study selection and data extraction within the scoping review were undertaken by one reviewer. A second reviewer would provide further validation. However, supervisory input was sought throughout the process to support critical appraisal and ensure consistency. Another limitation is that the search strategy did not include the term “life cycle assessment” or “environmental life cycle assessment,” which introduces potential evidence selection bias. This decision was made to manage the volume of studies and focus on multidisciplinary frameworks. However, previous research has extensively reported on eLCA in healthcare environmental assessment, and its usefulness is well established. Another limitation is that the scoring mechanism used to rank the frameworks was conducted by a single researcher and did not follow a validated methodology. Instead, the scoring mechanism was used to highlight frameworks that may have the most potential to be adapted to an HTA context. Additional research is needed to establish a consensus on how best to evaluate such frameworks.
Overall, the findings of this review support the growing recognition that environmental impacts need to be considered a core pillar of health system sustainability and provide potential mechanisms for incorporating environmental impacts alongside traditional assessments of effectiveness and cost-effectiveness. For example, Braithwaite et al. (2024) outlined that, in response to climate change and ecological degradation, resilient health systems need to balance environmental aspects alongside existing clinical and economic priorities (Reference Braithwaite, Smith and Zurynski51). In practice, HTA agencies could begin by defining the decision context and relevant environmental hotspots, then select proportionate methods, including carbon footprinting, eLCA, or MCDA, and the boundary set by the care pathway chosen, for example, according to the technology, available data, and intended use of the assessment (Reference Holtorf, Pegg, Mueller and McMeekin7;Reference Pegg, Gregg, Bishop and Bouttell21). These approaches should be introduced iteratively through pilot assessments, multidisciplinary stakeholder deliberation and capacity building, with findings used to refine methods and develop context-specific guidance before routine implementation (Reference Pegg, Gregg, Bishop and Bouttell21;39;43).
To continue this discussion, further research is recommended to assess barriers to developing HTA frameworks and to explore stakeholders’ views to gain explicit information on appropriate methods for assessing environmental impact in HTA.
Conclusion
This novel research identifies and synthesizes multidisciplinary methods from various disciplines that could support the assessment of environmental impact in HTA. The framework rankings suggested are important to consider when facilitating the development of robust environmental sustainability frameworks that align with healthcare decision-making processes. The findings suggest that a combination of environmental assessment methods is likely required to develop comprehensive and credible sustainability frameworks for HTA. Crucially, the integration of environmental sustainability impacts within healthcare decision-making frameworks is not uniform, but will depend on the specific technology being assessed, the context in which it is used, and how environmental impacts are generated and valued. Together, these findings highlight the need for flexible, context-sensitive approaches to embedding environmental considerations within HTA, rather than a single standardized methodology.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/S0266462326104309.
Funding statement
This research received no specific grant from any funding agency, commercial, or not-for-profit sectors.
Competing interests
The authors declare none.
Open access
