Environmental sustainability evidence in critical care medicine: a scoping review

Abstract

Purpose

We sought to conduct a scoping review to summarize the existing literature on environmental sustainability in critical care medicine, methods for measuring intensive care unit (ICU)-related environmental impact, and interventions to improve ICU environmental sustainability.

Source

We conducted a systematic search of four databases to identify studies published before October 2024 pertaining to environmental sustainability and assessment of the carbon footprint of ICUs and manually screened selected journals.

Principle findings

The combined search strategy identified 1,967 potentially eligible articles. We conducted a full-text review of 149 articles, of which 99 met eligibility criteria. Articles were categorized into five domains: waste management (n = 31), carbon footprint (n = 20), sustainable medication practices (n = 9), energy delivery/consumption (n = 4), and other (n = 39), with four studies classified into two themes owing to overlap in content. Daily material waste in the ICU ranged from 4.9 to 15.2 kg per patient per day. Most studies reported poor waste sorting practices, and several studies reported reductions in waste generation with multicomponent interventions. Three carbon-footprint-themed studies calculated the carbon footprint of ICUs either through life-cycle assessments or material flow analysis. Eleven studies quantified carbon emission reductions using either prospective or theoretical stewardship interventions related to equipment, medications, or ICU care processes.

Conclusions

In this scoping review, we found a rapidly expanding yet fragmented body of evidence on ICU environmental sustainability. Important gaps exist in evidence to guide clinicians on practical bedside strategies and understanding how to implement them to drive meaningful, sustainable change.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12630-026-03142-2.

Human activities are the primary cause of the climate crisis, with more than a million deaths, four billion injuries, and numerous health-related consequences due to climate-related geophysical disasters over a 20-year period.¹ Nevertheless, the system tasked with responding to increased climate-related health care demands is also contributing to the problem with carbon emissions from health care systems, which contribute up to 5% of the net global carbon emissions.² The intensive care unit (ICU) has some of the highest per-patient resource use rates of any health care environment.^3,4 When compared with other acute care units, an ICU can generate 1.3 times more solid waste and 3.1 times more greenhouse gases per patient per day,⁵ providing the opportunity for ICUs to have a central role in initiating and improving sustainability in health care.^4,6

There have been several recommendations for sustainable practice in the ICU internationally.^4,7,8 Despite campaigns and recommendations, there is a paucity of data summarizing evidence for specific initiatives and actions that clinicians can take to decrease the environmental impact of health care. It is also unclear which initiatives can be applied to the ICU setting, creating the need for specialty-specific recommendations.⁶ To address sustainability practices in critical care, we sought to conduct a scoping review to summarize the existing research on environmental sustainability, methods for measuring ICU-related environmental impact, and interventions to improve ICU environmental sustainability. We selected the scoping review methodology to enable provision of a broad perspective of the ICU sustainability evidence.

Methods

This scoping review is reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses Extension for Scoping Reviews (PRISMA-ScR) framework.⁹ We did not publish our protocol.

Literature search strategy

The initial search strategy was developed in Ovid MEDLINE® (United States National Library of Medicine, Bethesda, MD, USA/Wolters Kluwer N.V., Alphen aan den Rijn, Netherlands) by a health sciences librarian (M. S.) in collaboration with subject experts (M. B. and S. M.). The search combined Medical Subject Headings (MeSH) and keyword terms relating to ICU settings and environmental sustainability. The search strategy was then adapted for each included database: Ovid MEDLINE, Cumulative Index to Nursing and Allied Health Literature (CINAHL; via EBSCOhost, EBSCO Information Services, Ipswich, MA, USA), Embase® (Elsevier, Amsterdam, Netherlands), and Scopus (Elsevier). The date of the most recent search was 4 October 2024. Full search strategies are tabulated in Electronic Supplementary Material (ESM) eTables 1–4. Database searches were conducted from inception to October 2024.

One author (M. B.) manually reviewed the online indexes of eight critical care and/or climate change and health journals (American Journal of Respiratory and Critical Care Medicine, CHEST, Critical Care, Critical Care Medicine, Journal of Climate Change and Health, Annals of Intensive Care, Critical Care and Resuscitation, and Pediatric Critical Care Medicine) from January 2020 to September 2024.

Eligibility criteria

We included all publications, regardless of the type of article, that assessed or discussed environmentally sustainable practices in ICU settings and/or the environmental footprint of ICUs. We did not apply language restrictions. Both adult and pediatric ICUs were eligible. We excluded articles that discussed critical care procedures without explicit mentions of ICU settings. For example, studies that included acute care interventions relevant to the ICU but performed in non-ICU settings such as the operating room or emergency department were excluded.

Study selection

Two authors (J. D. and K. H.) performed title and abstract screening in duplicate with data tracking in Covidence (Veritas Health Innovation, Melbourne, VIC, Australia).¹⁰ Disagreements were resolved via consensus, and if agreement could not be reached, the article was retained. Records deemed eligible by both reviewers were subsequently retrieved and underwent full-text screening in duplicate. One author (M. B.) identified articles through review of online journal publication indexes, and those judged to meet eligibility criteria were also included.

Data extraction

Data abstraction was performed by four reviewers (M. B., T. B., J. D., and K. H.) using a standardized extraction sheet. Extracted data included basic study information such as article type, journal name, publication year, and study design, as well as study objectives, main findings, and conclusions. Studies were categorized into five domains developed a priori: 1) waste management, 2) carbon footprint, 3) energy delivery/consumption, 4) sustainable medication practices, and 5) other/unclassified. Study data were reviewed by one author (M. B.) to identify intervention-based and implementation studies which were categorized by study type and domain. Non-English articles were translated by an author with fluency in the publication language or using Google Translate (Google LLC, Mountan View, CA, USA).¹¹

Results

Our literature search identified 1,955 articles. An additional 12 articles were included from manual review of selected journal issues for 1,967 articles. Review of titles and abstracts of the articles led to the subsequent screening of 149 full-text articles, of which 99 were included in the review (Fig. 1). The most common reason for article exclusion was non-ICU-specific setting followed by absence of sustainability-related content.

Fig. 1.

Flow diagram of article selection

Summary of articles

From 1993 to 2024, 99 articles describing the carbon footprint or sustainability initiatives within adult, pediatric, and neonatal ICUs were published (Table 1). Over the last decade, sustainability publications in critical care have increased nearly fivefold, with 69 studies published from 2020–2024 compared with 15 in the preceding 5 years (Fig. 2). Nearly half of all publications (48/99) were secondary research articles, with the largest single category (n = 36) being nonreview articles (i.e., perspectives, commentaries, or editorials).

Table 1.

Characteristics of the articles included in the scoping review

Characteristic	Articles, n/total N N = 99
Type of article
Primary research	51/99
Quality improvement	19/99
Cohort	8/99
Waste audit only	6/99
Lab/bench/modelling study	5/99
Survey	5/99
Qualitative study, interviews	4/99
Cross-sectional	2/99
Randomized clinical trial	1/99
Feasibility study	1/99
Secondary research	48/99
Literature and narrative reviews	9/99
Systematic, bibliometric, and/or scoping  reviews	3/99
Other (commentary, perspective, editorial, etc.)	36/99
Year of publication
Before 2001	1/99
2001–2005	1/99
2006–2010	4/99
2011–2015	10/99
2016–2020	19/99
2021–2024	64/99
Type of ICU
Not reported/unspecified	63/99
Pediatric/neonatal	14/99
Adult general	7/99
Multiple types	7/99
Neurocritical care	3/99
Surgical	2/99
Burn	1/99
Trauma	1/99
Cardiac	1/99
Continent of origin
Europe	46/99
North America	27/99
Oceania	14/99
Asia	9/99
South America	2/99
Africa	0/99
Various/multiple continents	1/99
Non-English language publications	5/99
Domain*
Waste management	31/99
Carbon footprint	20/99
Sustainable medication practices	9/99
Energy delivery/consumption	4/99
Other/not specified	39/99

^*Four articles were classified in two domains: Two were considered both carbon footprint and sustainable medication practices,^37,41 one was sustainable medication practices and waste management,²² and one was combination carbon footprint and waste management⁴³

ICU = intensive care unit

Fig. 2.

Publication year and domain of included articles

^*Four studies were classified in two domains resulting in 103 classifications of 99 included studies

The largest number of articles pertained to the theme of waste management (n = 31; 31%). Carbon footprint was the second most common theme, with n = 20 (20%) included articles. Medication practices and energy delivery/consumption were relatively uncommon, comprising only n = 9 (9%) and n = 4 (4%) of the total articles included, respectively. Thirty-nine articles (39%) did not fit predominantly into one of the themes. Four articles were classified into two themes owing to considerable overlap in content (Table 1). Five articles were non-English publications, four of which were amenable to translation by Google Translate,¹¹ and one was translated by one of the authors who reads Chinese as a native language (J. D.). Twenty-two (22%) articles evaluated interventions or implementation strategies, predominantly through quality improvement studies within the waste management and sustainable medication practices domains (Table 2). A summary of study characteristics and findings for all included articles was created (ESM eTable 5).

Table 2.

Summary of sustainability interventions and implementation strategies

Study type	Domain	Intervention
Quality improvement	Waste management	• Staff education12,22,23,25,26,79,96–98
		• Installation of recycling/domestic waste bins12,22–24,26,99,100
		• Improved signage and posters12,96–98
		• Initiation of a green team or sustainability leaders23,96,98
		• Optimizing infectious and general waste layout23 or bag sizes100
		• Implementation of new bedside-cart-stocking practices25,79
		• Implementation of new recycling practices100
		• Improvement plan developed through analysis of glove use and hand hygiene practices and staff survey101
		• Changing printers to default double-sided printing100
		• Prime and visual prompt nudge and social norm nudge on decreasing apron usage102
		• Lean Six Sigma and DMAIC methodology to implement changes through PDSA cycles to improve waste sorting practices27
		• Installation of water flow regulation devices22
	Sustainable medication practices	• Staff education48,49
		• Informational posters and publications49
		• Audit and feedback tracking spreadsheets49
		• Changing medication labelling practices to allow recycling of plastic bags22
		• Elimination of plastic bags from pharmacy deliveries22
		• Standardization of drug doses22
	Carbon footprint	• Staff education, audit, and feedback on unnecessary pathology tests35
	Other	• Implementation of digital charting, ordering, and documentation for CRRT103
Randomized controlled trial	Sustainable medication practices, carbon footprint	• Enteral vs intravenous phosphate replacement41
Cohort study	Other	• Implementation of reusable cloth gowns in place of single-use plastic gowns104
Laboratory study	Sustainable medication practices	• Ready-to-administer syringes prepared through aseptic batchwise filling in pharmacy vs ready-to-administer sterilized syringes produced in compounding pharmacy50

CRRT = continuous renal replacement therapy; DMAIC = define, measure, analyze, improve, control; PDSA = plan, do, study act

Waste management

Nearly one third of all articles included pertained to waste management, and of these, 29% (9/31) reported a waste audit alone (Table 3). Approximately half (16/31) of the included waste management articles evaluated interventions relating to waste management.

Table 3.

Summary of articles evaluating waste management

(A) Waste-audit-only studies
Reference	ICU setting	Country	Waste audit findings
Simbruner 199321	PICU, 5 beds	Austria	Over a 7-day period: • 6,500 L of waste, mainly medical products and packaging • 400 L glass • 200 L hazardous waste (sharps and infectious material) • 2,000 syringes and a similar number of gloves per week
McGain 200918	ICU, 10 beds	Australia	Over a 7-day audit: • Total waste was 540 kg, of which:
			° 401 kg (74%) was general waste, of which:
			▪ 57% recyclable
			▪ < 1% infectious
			° 139 kg (26%) was infectious waste, of which
			▪ 87% infectious
			▪ 6% noninfectious, nonrecyclable
			▪ 7% recyclable noninfectious
			° 240 kg (44%) was recyclable
			Total ICU waste constituted 4.7% of total hospital waste.
Kubicki 201517	ICU, 11 beds	Australia	Over a 7-day period: • total ICU waste, 505 kg • general waste, 222 kg (44%) • infectious waste, 138 kg (27%) • potentially recyclable waste, 145 kg (28%) Of the total waste, 76% was disposed of correctly. Of the potentially recyclable waste, 49% was actually recycled (14% of the total ICU waste). Infectious waste bins had 82% of waste which was truly infectious. Estimated cost of recycling program an additional ~AUD 1,000 annually.
Parkin 201619	Cardiac ICU	UK	Sharps containers were weighed and analyzed over a 1-week period:
			• 62 items weighing 123.82 kg • Estimated 6.5 tonnes per year Reported contamination of sharps bins with non-sharps objects.
Ghersin 202016	PICU, 14 beds	USA	76 kg of unused medical waste collected over 3 weeks, mostly at the time of discharge/transfer and mostly from isolation rooms. Most commonly unused disposed items were endotracheal tubes, diapers, disposable under-pads, and flexion suction catheters.
Cochius Den Otter 202214	Tertiary PICU, 28 beds	Netherlands	26.9 kg per day.
			Highest categories by weight:
			• food waste (19%)
			• protective clothing (14%)
			• medical product packaging (13%)
			6% of waste was unused items.
Corbin 202215	Neuro ICU, 20 beds	USA	• 4.9 kg of waste per patient per day.
			• 96.8% sent to landfill, generating 119.7 metric tonnes CO2e per year
			Of the total waste, 24.7% could be diverted from landfills. Of this divertible waste: • 47.9% could undergo industrial composting. • 28% could undergo mechanical recycling. • 22.2% could undergo advanced recycling. • 1.9% could undergo sterilization and reuse.
			Appropriate sorting/processing could reduce emissions by 7.7%.
Anstey 202313	51 ICUs	Australia and New Zealand	Less than 40% of items of paper, glass, intravenous fluid bags, medications cups, and metal instruments were recycled per patient per bed space over a 12-hr shift.
			Disposal per bed space per 12-hr shift included 10 gowns, 10 syringes, and 30 pairs of gloves.
			Of the patients studied, 53.2% of them had any item recycled from their bedspaces.
Sala 202320	ICU	Portugal	Compared SLED with CRRT: • Waste: SLED, 1,600 g vs CRRT, 2,700 g plus substitution volume • Water: SLED, 237 L at 8 hr/345 L at 12 hr vs CRRT, 43.2 L at 24 hr/129.6 L at 72 hr • Energy usage: SLED, 5.71 kW at 8 hr/8.43 kW at 12 hr vs CRRT, 12–14.4 kW at 24 hr

(B) Waste intervention studies
Reference	ICU setting	Country	Waste audit findings	Intervention	Intervention efficacy
Hsiang 200212	Surgical ICU	Taiwan	Amount of regular garbage produced by a single nursing member 15.18 kg·day−1, no recycling of paper waste.	Development of strategy and principles for resource reclamation, recycling, monitoring system, and employee education/promotion of using own eating utensils and shopping bags.	~25% reduction in garbage produced by single nursing member (decreased to 11.4 kg per day). • Increase in paper waste recycling to an average of 17.2 kg per week per 12 beds. • Recycling of zipper bags and subsequent reduction of use of vomit bags ( ~ 50% reduction). • Approximate savings per ward per month, ~NTD 395.
Wisdom 201227	PICU	USA	16% of waste in regulated medical waste containers met criteria for regulated medical waste.	Methodology based on Lean Six Sigma, using DMAIC method to identify variations in waste disposal practices, and streamline and standardize these processes.	32% reduction in regulated medical waste in the PICU with interventions implemented in other departments resulting in estimated hospital-wide savings of USD 118,000 annually.
Hames 201397	ICU, 8 beds	Australia	Over four 24 h periods total, clinical waste 42.3 kg (plus 1.8 kg from an infectious room). • 44% general waste • 14% sharps waste	Staff education with in-services and signage.	Large decline in sharps waste attributed to clarification and appropriate sorting of Albumex glass bottles.
Morrow 201379	ICU, 40 beds	Canada		• Staff education. • Development of infection control guidelines for bedside supply cabinets and linen supply areas. • Removal/relocation of surplus supplies. • Implementation of new nurse server/cart stocking guidelines.	Improved teamwork between nurses and service support workers. • Cost savings (not quantified).
Cockerham 201625	ICU, 51 beds	USA		• Standardization and updating of bedside cart with staff engagement and feedback. • Staff education. • Education poster on reprocessing.	63% decrease in amount of supplies wasted between patients (by cost). • Sustained for 3 years following implementation. • Projected to save USD 72,654.40 per year.
Furukawa 201622	ICU, 41 beds	Brazil	13.7 kg per patient per day total. • 8.76 kg per patient per day infectious and sharps • 0.001 kg per patient per day chemical • 3.76 kg per patient per day normal nonrecyclable • 9 kg per patient per day normal recyclable	Lean Six Sigma methodology used to analyze medication processes by nurses and to develop and implement interventions aimed at environmental sustainability including recycling.	38.6% total waste generation (reductions in materials used and infectious waste, sharps, and normal nonrecyclable waste). • Estimated waste reduction of 5,324.9 kg less waste in 1 month. • 37.1% reduction in plastic bag usage. • 67.1% reduction of hormonal anti-inflammatory drugs. • 32.2% improvement in proper waste disposal. • Increase in recycling of plastic bags.
Jouet 2017100	ICU, 16 beds	France	~8 kg waste·per patient per day. • 55% general waste • 45% of which was infectious waste	• Developed recycling program for polypropylene, colourless polyethylene, and coloured polyethylene. • Changed relative size of infectious and general waste bags to better reflect relative amounts of each type of waste generated. • Diversion of medical broken glass to general waste. • Transition to using large soap pump dispensers. • Reduction in plastic bag usage for linens. • Changing printer default to double-sided.	Estimated annual savings of at least GBP 6,950.
Shen 201798	ICU	Taiwan		• Educational posters. • Educational training sessions. • Establishment of an “environmental protection pioneer team.”	Waste weight reduction per person of 13.2%.
Prasad 201826	ICU	UK		• Introduction of more domestic and recycling bins at each patient bed space. • Staff education and communication (email, posters, teaching sessions).	Cost of waste bag disposal decreased by 22%, largely owing to reduction of disposal of clinical waste bags (compared with domestic waste bags). • Estimated annual cost savings of GBP 2,620.93.
Buxton 201924	ITU/HDU	UK		Installation of domestic (i.e., nonclinical) waste bin in every bay with explanatory poster signs.	Cost savings of GBP 205 per bed space per year.
Allen 202199	ICU (COVID-negative area, 4 patients)	UK		Installation of labelled recycling boxes.	3.2 kg of recyclable material collected over a 24-hr period. • 2.25 kg paper and card • 0.85 kg plastic
Chang 202396	ICU	Taiwan	Biomedical waste, 2.98 kg per person per day.	• Multimedia medical waste education training courses. • Water reduction teaching videos. • Waste classification pictures. • Environmental protection reminders. • Environmental protection leaders.	~12% waste reduction (down to 2.62 kg per person per day. • Improved accuracy of waste sorting (85.3%). • Increased rate of classification behaviour (95%).
Clark 2023101	Critical care unit	UK		Development and implementation of change plan.	Reduction of unnecessary disposable glove use from 73% to 48% of observations.
Martin 202323	ICU	France	Infectious waste represented 58.9% of total waste.	• Established Green Team. • Reorganized waste sorting and recycling channels and layouts in the ICU rooms. • Developed documents on eligibility of recycling of materials.	Decreased percentage of infectious waste from 58.9% to 29.9% of total waste. • Reduced the monthly cost of waste treatment from GBP 4,235 to GBP 2,930. • Reduced CO2 production from 5,533 kg to 3,438 kg monthly. • Enabled recycling of 651 kg plastic and 4,808 kg cardboard over 1 year.
Nguyen 202441	Medical, surgical, trauma ICU	Australia		In a prospective, randomized, parallel-group, noninferiority clinical trial: • Sodium phosphate 500 mg by mouth or gastric tube compared with sodium dihydrogen phosphate or potassium dihydrogen phosphate 10–20 mmol	Noninferiority of enteral/oral phosphate at the 0.2 mmol·L−1 noninferiority margin but lower weight of waste: 7.7 g (8.3 g) vs 217 g (169 g); difference, 209 g (95% CI, 168–250).
Van Der Zee 2024102	ICU	Netherlands		Implementation of two sets of nudges on behaviour change: prime and visual prompt nudge and social norm nudge.	Neither the prime and visual nor the social norm nudge decreased apron use by health care workers.

CI = confidence interval; CRRT = continuous renal replacement therapy; HDU = high-dependency unit; ICU = intensive care unit; ITU = intensive therapy unit; PICU = pediatric intensive care unit; SLED = sustained low-efficiency dialysis

Several studies quantified material waste generation in the ICU with reported rates of total waste between 4.9 and 15.2 kg per patient per day.^12–21 The most common types of waste were general and infectious (regulated medical waste). Many studies evaluated the accuracy of sorting practices and recycling potential, reporting that between 8.7% and 44% of materials were considered recyclable.^17,18,22 Four papers focused on specific aspects of ICU-related waste including sharps disposal, infectious waste, unused supplies, or waste generated with renal replacement therapy.^16,19,20,23

Of the n = 16 studies evaluating single or multicomponent waste management interventions, these included staff education (n = 7), implementation of recycling processes (n = 6), and improved waste sorting practices (n = 11) (Table 4).^12,23–27 Significant cost savings were reported in half of these articles (n = 8), with reported estimated annual savings ranging from approximately TWD 4,740¹² to USD 72,654.²⁵

Table 4.

Summary of articles evaluating carbon footprint

Reference	ICU setting	Country	Study design	Methodology	Findings
Estimating the carbon footprint of critical care
McGain 201829	Medical, surgical	USA and Australia	Cohort	LCA	• Average daily ICU carbon footprint for patients with septic shock was 178 kg CO2e in the USA vs 88 kg CO2e in Australia. • The reason for the difference was linked to energy consumption for HVAC services.
Sanchez 202036	Medical, surgical	USA	Cohort	LCA	• Reusable blood pressure cuffs have a theoretical 10-fold smaller carbon footprint compared with disposable cuffs in the ICU.
Prasad 20225	Medical	USA	Cohort	LCA	• ICU generated 7.1 kg of solid waste and 138 kg CO2e·bed-day−1. • Acute care ward generates 5.5 kg of solid waste and 45 kg CO2e per bed-day.
Hunfeld 202328	Trauma, medical, surgical	Netherlands	Cohort	MFA	• Consumption of materials in the ICU accounts for 12 kg CO2e·bed-day−1. • Material throughput is 17 kg mass, 300 L of water usage, and 4 m2 of agricultural land occupation per bed-day.
See 20234	Mixed	Multiple	Review	Various	• Review identifying key methods of evaluating the environmental impact of critical care: life-cycle analysis and material flow analysis.
Carbon footprint reduction or comparison of theoretical interventions
Cowboy 200932	Medical, surgical	USA	Cohort	Targeted carbon estimate	• Telehealth interventions may reduce gasoline consumption due to travel.
Anderson 202330	Medical, surgical	UK	Cohort	LCA	• Use of pea protein (plant-based) rather than whey protein (animal-based) reduced the carbon footprint of ICU care.
Cameron 202331	Medical Surgical	UK	Cohort	LCA	• 56 continuous renal replacement therapy days amounted to a dialysis-related carbon footprint of 3,255 kg CO2e. • Optimizing or reducing the dose of renal replacement therapy could reduce carbon emissions.
Hemberg 202333	N/A	Sweden	Correspondence	LCA	• Central venous line insertion kits with reusable instruments cost less and have a lower environmental footprint.
Touw 202339	N/A	Netherlands	Correspondence	LCA	• Carbon emissions from intravenous fluids mostly originate from production and disposal of the polymer bag. • Emissions range from 130 g CO2e per 50 mL bag to 580 g CO2e per 1,000 mL bag.
Wong 2023105	N/A	Multiple	Correspondence	Targeted carbon estimate	• Proposal of a carbon footprint calculator associated with educational events based on numbers of attendees and faculty, travel, and other activities.
Estimating the carbon footprint of critical care
Carrandi 202442	Mixed	Multiple	Scoping review	Various	• Four studies met inclusion criteria, though none of the health economic evaluations of critical care incorporated environmental impact into their assessments.
Gaetani 202443	Mixed	Multiple	Systematic review	Various	• 13 studies assessed and described the environmental impact of 36 adult or pediatric ICUs. • Estimated carbon emission from critical care ranged between 88 and 178 kg CO2e·per patient per day, with key sources being electricity and gas use as well as consumables.
Ozkan 202444	Mixed	Multiple	Review	Various	• Review including 15 studies related to carbon footprint in ICU, with the first dating to 2014.
Soong 202438	Neuro, surgical	Singapore	Cohort	LCA	• For 31 patients with an average length of stay of 5.9 days, the carbon footprint from sedative and analgesia usage was 2.2 kg CO2e, and from medication wastage (not used but disposed) as 0.3 kg CO2e.
Massart 202434	N/A	France	Correspondence	LCA	• In a 21-bed ICU in France, reusable bronchoscopes had a lower annual carbon footprint when used more than 50 times per year compared with single use.
Santander 202437	Neuro	France	Cohort	LCA	• Switching from intravenous to oral levetiracetam in a 15-bed neuro-intensive care unit could reduce carbon emissions by as much as 1.200 kg CO2e·year−1.
Wang 202440	N/A	USA	Correspondence	Targeted carbon estimate	• By eliminating faculty and learner travel, delivery of online critical care education courses significantly reduced carbon footprint compared with in-person formats.
Interventions reducing carbon footprint
Nguyen 202441	Trauma, medical, surgical	Australia	Randomized clinical trial	Modified LCA	• Oral phosphate compared with intravenous was noninferior for replacement of hypophosphatemia in patients who were critically ill and is 20-fold less carbon intensive.
Pilowski 202435	Neuro, trauma, medical, surgical	Australia	Prospective intervention/quality improvement	Modified LCA	• Multimodal education and audit/feedback interventions can reduce unnecessary blood testing in the ICU (e.g., 31.2% reduction in arterial blood gases). • Annual emissions savings across 4 ICUs was 1,800 kg CO2e annually.
Other

HVAC = heating, ventilation, air conditioning; ICU = intensive care unit; LCA = life-cycle assessment; MFA = material flow analysis; N/A = Not applicable

Table 3 and ESM eTable 5 summarize additional articles on this theme.

Carbon footprint

One-fifth of included articles (n = 20) related to the carbon footprint of ICU care (Table 4). Moreover, three articles used life-cycle assessment (LCA) or material flow analysis to evaluate the carbon footprint of critical care provision or processes,^5,28,29 11 articles compared carbon emission reductions from completed or proposed stewardship interventions,^30–40 and one randomized controlled trial (RCT) reported a reduction in carbon footprint.⁴¹ Four reviews, including one systematic review and one scoping review, summarized the carbon footprint of adult and pediatric/neonatal critical care.^4,42–44

Carbon emissions reported from the three hybrid LCAs undertaken in two included articles measured 88 kg CO₂ equivalents (CO₂e) per patient per day (Australian ICU),²⁹ 178 kg CO₂e per patient per day (American ICU),²⁹ and 138 kg CO₂e per bed per day.⁵ Differences in estimates were largely attributable to differences in energy sources due to different heating, ventilation, and air conditioning requirements as a result of geographical climates, in addition to some variation in the LCA system boundaries. In all three LCAs, the largest contributors to greenhouse gas emissions were energy (including electricity), gas, and consumables, underscoring the significant impact of heating, ventilation, and air conditioning particularly in hospitals with predominantly nonrenewable energy sources. Carbon emissions from material waste were measured in one article amounting to 12 kg CO₂e per patient per day primarily composed of disposable gloves (49,900 kg CO₂e), syringes (27,000 kg CO₂e), and disposable medical clothing (24,300 kg CO₂e).²⁸

Most of the carbon footprint articles (15/20) recorded (n = 4) or projected (n = 11) the reduction in carbon emissions (footprint) through prospective interventions and theoretical comparisons. One prospective multiyear intervention to reduce unnecessary blood testing across four ICUs resulted in an annual reduction of 18,000 kg CO₂e.³⁵ A theoretical 10-fold reduction in carbon emissions was reported when switching from disposable to reusable blood pressure cuffs,³⁶ while switching from intravenous to oral levetiracetam to treat seizures in a 15-bed ICU theoretically could save 1,480 kg CO₂e annually.³⁷ Finally, potential carbon savings were also identified by using smaller intravenous bags when possible.³⁹

Studies using experimental designs were uncommon. We identified only one (5%) well-designed noninferiority RCT that reported that oral phosphate (vs intravenous) replacement in hypophosphatemia was safe and cost-effective and reduced the estimated carbon footprint 20-fold.⁴¹

Sustainable medication practices

Nine (9%) of the included articles addressed sustainable medication practices, most of which (n = 6; 67%) were interventional, plus an additional theoretical comparison study. Classes of medications encompassed in sustainable medication practices included antimicrobials,⁴⁵ anticonvulsants,³⁷ vasopressors,⁴⁶ electrolyte replacement,⁴¹ inhaled nitric oxide,⁴⁷ and analgesics/antipyretics.⁴⁸ Four articles evaluated potential waste and cost savings of different drug preparation or administration practices,^22,46,49,50 and three studies evaluated enteral vs intravenous administration.^37,41,48

Electronic Supplementary Material eTable 5 summarizes articles relating to quality improvement implementation initiatives,^22,48,49 medication degradation with storage temperature,⁴⁶ preparation practices such as prefilled syringes,⁵⁰ and others.

Energy delivery/consumption

Four (4%) articles discussed power delivery and consumption. One study reported on the development of a model to accurately predict electricity consumption in an ICU in the UK on the basis of patient acuity, organ support, and workplace practices.⁵¹ Another study compared start-up, standby, and operating mode energy consumption and estimated annual electricity savings of as much as 1.2 tons of CO₂e savings by turning off ventilators 80% of the time when they were not in use.⁵² The remaining two articles described the use of solar rooftop photovoltaic panels in India to power an open care radiant warmer in a neonatal ICU⁵³ and modeled the integration of an “electric spring” to improve power system quality in an ICU under various conditions.⁵⁴

Other

Of the included articles, 39% did not clearly fit any of the prespecified domains. Of these, n = 34 (87%) were secondary research articles, including 28 commentaries, perspectives, editorials, or brief reports and 6 reviews, discussing the effects of critical care medicine on climate change and providing suggestions for sustainability initiatives. These covered a wide range of levels from bedside clinical care^7,55–63 to unit/organizational level,^64–66 and the infrastructure and physical design of ICUs.^67–70

Only five (13%) of the unclassified articles included original research. Three of these were qualitative studies examining definitions, perspectives, and sustainability change ideas of ICU clinicians (largely nurses).^71–73

Over a quarter of the unclassified articles were published in critical care nursing journals and/or focused primarily on ICU nursing practices,^{60,62,64,67,69,71–75} while five were published in pediatric/neonatology journals or from pediatric/neonatal ICU perspectives.^{62,68,69,76,77}

Electronic Supplementary Material eTable 5 provides summaries of all articles.

Discussion

In this scoping review, we identified 99 unique articles related to environmental sustainability in critical care medicine across more than four domains. Despite an increasing number of publications focused on sustainability, there is a paucity of high-quality evidence available to guide ICU practitioners on specific sustainability practices. Furthermore, only two articles were primary research studies completed in Canada, highlighting a national gap in this area of investigation.^78,79 This review underscores the need to prioritize intervention-based sustainability research in the ICU since existing work shows high resource consumption, high environmental impact of ICU-related activities, and a multitude of sustainability opportunities. Intervention-based studies would be well supported by research to identify behavioural change strategies to effectively incorporate sustainability interventions within the ICU.⁸⁰ Intervention and behaviour change studies identified in this review included a limited number of implementation strategies,⁸¹ perhaps related to feasibility, resource constraints, and the narrow locus of control of involved clinicians. Future work should focus on developing implementation strategies and specific and effective interventions to improve environmental sustainability in critical care.

Although previous reviews have been published on sustainability,^7,8 this scoping review summarizes the published literature specific to environmental sustainability in critical care. Our findings complement those of existing reviews,^4,43,61,66 including those that include resource use and cost in critical care.⁸² Our findings also complement prior reviews in other medical specialties including surgery/anesthesiology^83–85 and radiology,⁸⁶ by identifying similar sustainability domains and highlighting opportunities for intervention. Further, the present review provides a broad overview of environmental sustainability initiatives and a framework of sustainability domains to help guide future research in this area. Our findings support published statements and emerging guidance on sustainability initiatives in critical care medicine.^87–89

Our review shows a fivefold increase in the number of sustainability publications in critical care medicine over the last decade. As high-quality, quantitative data increase and our understanding of the carbon footprint of critical care services develops, practical and easily accessible platforms for information sharing will be imperative to drive meaningful change. There is already a living database of pharmaceutical and health care process life-cycle analyses which may be helpful in providing accurate assessments of environmental impacts;⁹⁰ nevertheless, a wider scope and better understanding of the limitations and generalizability of available data are required to implement sustainability initiatives.

Only half of all publications were primary research articles, with most of these reporting observational studies, mainly from Europe and North America. Furthermore, of all articles included, only 22 involved interventions. These observations highlight the need for further study into which critical care processes are modifiable and the efficacy of specific sustainability interventions. Life-cycle assessments, although highly valuable, are neither necessary nor feasible for each process performed or item used in ICUs. Future studies should focus on highlighting the potential environmental benefits of eliminating low-value care and environmental impact assessments of tested interventions integrated within clinical trials. Among the largest theme categories, waste and carbon emissions, a mere 52% and 10% of included articles were interventional, respectively. Much of the interventional work was performed within the realm of sustainable medication practices and initiatives. This finding may reflect the early stages of descriptive research in this field; nevertheless, it may also suggest the presence of institutional barriers that hinder larger-scale sustainability initiatives such as a lack of leadership support, de-prioritization of resources for environmental changes, lack of guidance/perceived expertise in the field, or even low buy-in within the specialty. Many interventions may also be constrained by national regulations and regional/local services and thus are typically beyond the direct influence of bedside ICU clinicians. Understanding and addressing these barriers will be fundamental to long-lasting sustainable change, in addition to multidisciplinary collaboration and broadening sustainability initiatives beyond the walls of the ICU to the hospital and health care system. The overlap between critical care medicine and anesthesiology offers an ideal opportunity for crossdiscipline collaboration on sustainability initiatives.⁶¹ For example, existing literature from other specialties can be applied in ICU settings related to environmental impact of reusable vs single-use laryngoscopes,⁹¹ pulse oximeters,⁹² anesthetic gases,⁹³ and procedure kits.^94,95

In this review, we identified a wide range of effective interventions varying in feasibility and impact, many of which have been proposed by others.^7,8 These include small-scale changes in daily bedside practice that can be easily applied by clinicians immediately, such as improving waste sorting, minimizing unnecessary supplies waste, powering off standby electronic equipment, choosing equivalent enteral/oral medications over intravenous, and establishing multidisciplinary green teams. Larger-scale, more impactful potential interventions identified include advocating for more efficient heating, ventilation, and air conditioning systems; renewable energy supply to hospitals; institutional/regional environmental sustainability strategies; and sustainable procurement practices. Nevertheless, our findings highlight a lack of studies comparing efficacy of different sustainability interventions or implementation strategies.

Finally, we identified several gaps in the literature. These include the impacts (and mitigation strategies) of commonly used pharmaceuticals and procedural equipment in the ICU. Few studies investigated reusable equipment or circular economy interventions. Collectively, these components of daily ICU care result in significant resource use and carbon emissions²⁸ and represent ideal processes for sustainability teams to address. Furthermore, the paucity of data from low- and middle-income countries represents a missed opportunity for the global critical care community to learn from innovations in low-resourced areas. A global crisis demands global solutions, with broad collaboration including clinical, educational, research, and organizational efforts to reduce the ecologic footprint of critical care. Success will depend on the collective efforts of individual health care clinicians and the leadership of international critical/intensive care societies.⁸⁸

Limitations

This scoping review has several limitations. First, we focused on critical care publications. As such, the search did not include studies conducted in other environments (operating rooms or emergency departments) that may also be relevant for the provision of acute care. Further, we did not include studies conducted in the ICU without explicit mention of environmental or sustainability concepts. Consequently, we may have overlooked articles that described environmental benefit, although not explicitly stated, such as studies specifically evaluating reduction of low-value care within ICUs, a topic of a recent review.⁸² The a priori-determined four-domain framework in this review has not been validated in other studies or health care settings; other frameworks may be appropriate, particularly for sustainability interventions as proposed by others.^7,8 Additionally, we did not appraise the risk of bias in the included studies, which would have necessitated use of different instruments and was beyond the scope of the current review.

Conclusions

In this scoping review, we identified several publications in different domains. Nevertheless, despite the rapidly increasing body of research in the field of ICU sustainability, important gaps exist in evidence to guide ICU clinicians and health care teams regarding specific sustainability practices to implement at the bedside and the role of implementation science to drive change towards greater sustainability. This underscores the need to apply and integrate environmental sustainability research in the ICU amid data supporting high resource consumption and carbon emissions and an increasing call to identify sustainability opportunities and practices across clinical, educational, research, and organizational domains. Future work should focus on sustainable interventions and identification of potential barriers to implementation.

Supplementary Information

Below is the link to the electronic supplementary material.