Carbon footprint of hospital laundry: a life-cycle assessment

Abstract

Objectives

To assess greenhouse gas (GHG) emissions from a regional hospital laundry unit, and model ways in which these can be reduced.

Design

A cradle to grave process-based attributional life-cycle assessment.

Setting

A large hospital laundry unit supplying hospitals in Southwest England.

Population

All laundry processed through the unit in 2020–21 and 2021–22 financial years.

Primary outcome measure

The mean carbon footprint of processing one laundry item, expressed as in terms of the global warming potential over 100 years, as carbon dioxide equivalents (CO₂e).

Results

Average annual laundry unit GHG emissions were 2947 t CO₂e. Average GHG emissions were 0.225 kg CO₂e per item-use and 0.5080 kg CO₂e/kg of laundry. Natural gas use contributed 75.7% of on-site GHG emissions. Boiler electrification using national grid electricity for 2020–2022 would have increased GHG emissions by 9.1%, however by 2030 this would reduce annual emissions by 31.9% based on the national grid decarbonisation trend. Per-item transport-related GHG emissions reduce substantially when heavy goods vehicles are filled at ≥50% payload capacity. Single-use laundry item alternatives cause significantly higher per-use GHG emissions, even if reusable laundry were transported long distances and incinerated at the end of its lifetime.

Conclusions

The laundry unit has a large carbon footprint, however the per-item GHG emissions are modest and significantly lower than using single-use alternatives. Future electrification of boilers and optimal delivery vehicle loading can reduce the GHG emissions per laundry item.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• We analysed a comprehensive dataset allowing greenhouse gas emissions estimates for hospital laundry.

• The mass adjusted approach has facilitated estimates for a range of different commonly used hospital laundry items.

• The results are time and place specific, meaning they should be generalised with careful consideration.

• Our method of accounting for gas and electricity use by a combined heat and power unit is a recognised one, but might have overestimated the laundry unit’s greenhouse gas emissions.

Introduction

Healthcare is having a significant and ongoing contribution to the climate crisis, and health systems and governments are becoming increasingly aware of the need to address this.^1–4 Finding meaningful solutions requires data describing the nature and extent of the problem as a first step. Targeted life-cycle assessment provides us with a means of achieving this.⁵ It can demonstrate high-impact areas and how to reduce our environmental impact.

While we need to adopt a whole health system approach to address healthcare’s impact on the climate, it is particularly important to focus our efforts on greenhouse gas (GHG) emissions hotspots. Hospital laundry units use significant amounts of fossil fuels, electricity, water and detergents. The UK Department of Health recognises the need to address sustainability concerns across the National Health Service (NHS), including linen decontamination services.^{6 7}

To date, no life-cycle assessment has been published outlining the environmental impact of different healthcare laundry items, or a hospital laundry unit itself. Given the ubiquity of laundry items across healthcare settings, and the potential to generate understanding around mitigation strategies, further analysis is warranted.

Aim

To estimate the environmental impact of linen items reprocessed by a large hospital laundry unit.

Objectives

• Assess annual GHG emissions of a large laundry unit servicing multiple healthcare units.

• Perform a cradle to grave life-cycle inventory analysis (LCA) for linen items commonly reprocessed by a large hospital laundry service.

• Model different per-use GHG emissions outcomes depending on customer location relative to the laundry unit.

• Explore ways to mitigate the environmental impact of hospital laundry.

Methods

This is a cradle-to-grave LCA, following the GHG protocol product standard, for a large laundry unit (LDU) in Southwest England.⁸ The LDU services 12 care settings across the Royal Devon University Healthcare NHS Foundation Trust footprint in Devon. This includes the main acute hospital in Exeter, where the unit is based. Additionally, the unit provides a laundry service for 18 sites across three other NHS trusts in Southwest England. An operational description of the LDU is included in online supplemental appendix 1. A full LDU capital goods inventory and a continuous batch washer (CBW) schematic are included in online supplemental appendix table 1 and figure 1, respectively.

Unit of measure

Our unit of measure is the average cradle-to-grave GHG emissions associated with the use of an item of hospital laundry once. GHG emissions are presented as carbon dioxide equivalents (CO₂e) expressed in terms of global warming potential over 100 years (GWP100).

Boundaries

The unit of analysis begins with dirty laundry at the LDU site and ends with clean laundry at the LDU ready for return to the customer site. Additionally, we separately analyse GHG emissions due to transport to and from the LDU. The system boundaries are outlined in figure 1.

Figure 1.

System boundaries flow chart. GHG, greenhouse gas.

Data sources and assumptions

Linen

Items laundered through the LDU were weighed and their composite materials ascertained. These were added to an inventory of materials online supplemental appendix table 2. EcoInvent (V.3.9.1, 2022) emissions factors were used to calculate life-cycle GHG emissions for each item.⁹ Biogenic carbon was accounted for in biomass-derived materials (cotton and cardboard). A list of all emissions factors used is shown in online supplemental appendix table 3. Laundry items were assumed to have been manufactured in China. Transport emissions by container ship (Shanghai to Felixstowe, 22 052 km) and heavy goods vehicle (HGV) over land (Felixstowe to Exeter 495 km) were included. EcoInvent and UK Government Department for the Environment, Farming and Rural Affairs/Department for Business, Energy and Industrial Strategy 2022 (DEFRA/BEIS 2022) freight emissions factors were used for cargo shipping and HGV transport, respectively.¹⁰

Linen items were modelled as reused 100 times before disposal by recycling. This is a conservative estimate based on an internal LDU audit of reusable personal protective equipment gowns, which consistently lasted >100 laundry uses. Recycling-related transport emissions were included, and other recycling-related GHG emissions were not allocated within system boundaries. In scenario modelling, GHG emissions due to incineration were calculated using the contained mass of carbon in each item, and allocated within the system boundary. DEFRA/BEIS 2022 emissions factors for transport to a recycling and incineration facility were used. These data were used to derive GHG emissions for extraction, manufacture, transport and disposal of each linen item.

The LDU records all processed items. Data for two financial years (2020–21, 2021–22) were obtained. The type of item was accurately recorded in 99.5% of cases, belonging to one of 14 categories. The remaining 0.5% of items were either recorded by the LDU as miscellaneous, or categorised for our analysis as miscellaneous because they were laundered infrequently and did not fit into another category. An estimated mass for miscellaneous items was derived from a volume and mass adjusted average from all other laundry items. Dry cleaned items accounted for 0.025% of laundered items and were excluded from analysis. Item types, categories and their numbers laundered are outlined in online supplemental appendix table 4.

Gas, electricity and water

Gas and electricity use data for 24 months from April 2020 to March 2022 (inclusive) were obtained. Electricity supplied to the LDU comes from the national grid and an oil powered backup generator. Mains gas supplies steam, hot water, heat for laundry drying and building heating. This powers two boilers and a combined heat and power unit (CHP), the steam from which is used in the LDU. All LDU energy inputs are submetered. All CHP gas use was allocated to steam production, and the electricity produced and used elsewhere at the hospital site considered ‘free’ and outside the system boundary.¹¹ Supply and disposal of water via sewerage are metered. Relevant DEFRA/BEIS 2022 emissions factors were used to determine GHG emissions.

Detergents

Five main detergent products are supplied by Christeyns (Gent, Belgium); ‘Dual Ultra’, ‘Metajet Ultra’, ‘Bisoft classic’, ‘Sanoxy’ and hydrogen peroxide. Open-source information was used to determine the weight to volume composition of each proprietary product, and EcoInvent carbon factors used for each ingredient. Carbon dioxide and methane emissions due to organic detergent breakdown during wastewater treatment were estimated using data published by Saouter and van Hoof.¹² The average annual detergent volumes used were estimated using 13 weeks of data between November 2022 and June 2023. Online supplemental appendix table 5 outlines these calculation steps.

Staff travel

Across 24 hours the LDU is staffed by 60 laundry operatives and 8 office-based staff. We assumed that a 2020 trust-wide travel census of 822 staff was representative of LDU staff travel. Using these data, an average staff member’s GHG emissions travelling to and from work was estimated.

Transport

Laundry is transported between customers and the LDU by a fleet of HGV, with gross vehicle weights of 7.2, 12, 14 and 18 t. Laundry is transported on and off HGVs in wheeled pallet cages weighing 45 kg, and HGVs can transport either 12, 21, 24 or 30 cages depending on their size. HGVs are almost invariably filled to capacity with pallet cages when delivering and collecting laundry. The round-trip distances between each customer site and the LDU were calculated using Google Maps to determine the shortest HGV-appropriate route.

Different LDU customers request different numbers of laundry items to be loaded on to pallet cages based on their local manual handling protocols. Maximum and minimum pallet cage loading scenarios were acquired for each laundry item based on these requirements. We represented an hypothetical ‘average’ pallet cage by adjusting for the mass of each laundry item and the frequency with which each item was laundered annually. These representative minimum and maximum pallet cage loading scenarios were used to determine the typical ‘real-world’ minimum and maximum payload capacities achieved for delivery HGVs. The maximum achievable payload capacities for laundry are less than the true payload capacity of the HGVs. This is because even when pallet cages are maximally filled according to customer specification, and HGVs filled with the maximum number of pallet cages, the true HGV payload capacity is not reached.

Using the relative mass and numbers of different laundry items processed annually, we modelled the per-item delivery vehicle GHG emissions according to HGV size, the percentage of payload capacity transported and the round-trip distance travelled. DEFRA/BEIS 2022 freight emissions factors were used. These provide emissions factors for HGVs that are 0%, 50% and 100% filled with freight. Using these emissions factors for the different sized fleet HGVs, emissions factors at 10% increments of payload capacity were derived (online supplemental appendix table 6). We modelled up to 100% of true payload capacity even though the realisable payload capacity is less than this when using the current pallet cages.

Calculating the per-use carbon footprint

Using the activity data outlined above, the life-cycle GHG emissions for different laundry items were calculated. The LDU GHG emissions attributable to each laundry item were apportioned based on their mass because each CBW compartment has a maximum load capacity, and drying larger products typically requires greater energy.

Single-use alternative products

Mass and composition data were acquired for single-use alternatives to six commonly laundered items processed at the LDU (blankets, sheets, patient gown, scrub tops, scrub bottoms, curtains). Their life-cycle emissions were estimated to compare per-use GHG emissions against their reusable equivalents. Reusable laundry GHG emissions were modelled both excluding and including transport to and from the LDU. We additionally modelled the per-use GHG emissions for reusable items if incinerated at the end of their lifetime. For this, the incineration and waste transport emissions were both allocated to the respective product.

Data were managed, analysed and modelled using a series of custom databases and an author-built calculator in Microsoft Excel (Microsoft 365).

Patient and public involvement and engagement

Patients and the public were not involved in the design, conduct, reporting or dissemination of this study.

Results

Between 1 April 2020 and 31 March 2022, the LDU processed 26 168 258 laundry items. Table 1 shows these data as an annual average, and the activity levels and GHG emissions attributable to electricity, gas, oil, water, detergents and staff travel. The annual average LDU GHG emissions were 2947 t CO₂e. The average GHG emissions per laundry item were therefore 0.225 kg CO₂e, and 0.508 kg CO₂e per kilogram of laundry. These figures exclude transport to customer sites. Per item, 0.798 kWh of gas and 0.0593 kWh of electricity were used on average. Online supplemental appendix figure 2 shows the average weekly gas and electricity use per item across the 2-year period. There is a slight trend towards decreasing energy use per item across the time-period.

Table 1.

Overall activity data for the laundry unit financial years 2020–21 and 2021–22.

		Financial year		Annual average
		2021–22	2020–21	Activity	GHG emissions/t CO2e	% GHG emissions
Total laundry articles	Count	13 625 263	12 542 995	13 084 129
Main hospital site		3 546 899	3 206 084	3 376 492
Off-site hospitals		10 078 364	9 336 911	9 707 638
Mains electricity	kWh	796 113	816 021	806 067	210.83	7.2
Gas	kWh	10 629 130	11 047 413	10 838 272	2565.74	87.1
CHP and boilers	kWh	8 218 603	8 724 602	8 471 603	2005.48	68.1
Laundry drier	kWh	2 410 527	2 322 811	2 366 669	560.26	19.0
Oil	kWh	67 382	51 422	59 402	19.95	0.7
Generator	kWh	49 446	36 408	42 927	14.42	0.5
Boiler	kWh	17 936	15 014	16 475	5.53	0.2
Water supply	m3	37 021	36 566	36 793	5.48	0.2
Sewerage water	m3	30 727	30 350	30 539	8.31	0.3
Detergents	L	–	–	71 202	49.70	1.7
Staff travel	km	–	–	550 537	87.03	3.0
				Total	2947

CHP, combined heat and power; CO₂e, carbon dioxide equivalents; GHG, greenhouse gas;

Table 2 shows the different laundry items, the number laundered, overall GHG emissions and GHG emissions per use of each item. Figure 2 shows the different sources of GHG emissions per item use. Overall, the relative sources of GHG emissions were as follows: LDU gas (75.7%), LDU electricity (6.2%), LDU oil (0.6%), LDU water use (0.4%), LDU detergents (1.5%), LDU staff travel (2.6%), average laundry item manufacture (13.4%), average laundry item transport from manufacturing facility (0.8%), average laundry item disposal by incineration (−1.2%, due to recycling allocation and biogenic carbon).

Table 2.

Laundry item properties, volume laundered and GHG emissions

Item	Mass/kg	Material	Annual number processed	Total mass processed/kg	GHG emissions
					Annual LDU/ kgCO2e	Per item LDU/kgCO2e	Manufacture, freight, disposal per item use/kgCO2e	Total GHG emissions per item use/kgCO2e
Curtains	4.200	100% polyester	41 794	175 533	89 174	2.134	0.2424	2.376
Blanket	1.305	100% polyester	1 450 046	1 892 310	961 335	0.663	0.0695	0.732
Sheet	0.573	70% cotton, 30% polyester	3 480 334	1 994 231	1 013 113	0.291	0.0502	0.341
Mop head	0.454	90% cotton, 10% LDPE	295 616	134 209	68 181	0.231	0.0498	0.280
Slidey sheet	0.298	90% polyester, 10% silicone	83 153	24 780	12 589	0.151	0.0528	0.204
Linen bag	0.320	100% polyester	183 306	58 658	29 800	0.163	0.0185	0.181
Patient gown	0.296	50% cotton, 50% polyester	894 403	264 743	134 495	0.150	0.0230	0.173
Towel	0.271	100% cotton	2 216 172	600 583	305 109	0.138	0.0277	0.165
Scrub trousers	0.191	65% polyester, 35% cotton	795 799	151 998	77 218	0.097	0.0135	0.110
Pyjama top	0.184	50% cotton, 50% polyester	142 908	26 295	13 358	0.093	0.0143	0.108
Hand towel	0.168	100% cotton	73 937	12 421	6310	0.085	0.0172	0.103
Pyjama bottom	0.152	50% cotton, 50% polyester	157 243	23 901	12 142	0.077	0.0118	0.089
Scrub top	0.153	65% polyester, 35% cotton	819 683	125 411	63 712	0.078	0.0108	0.089
Pillow case	0.119	70% cotton, 30% polyester	2 385 866	283 918	144 237	0.060	0.0104	0.071
Miscellaneous	0.501	N/A	63 872	32 016	16 265	0.255		–
		Total	13 084 129	5 801 007	2 947 039

CO₂e, carbon dioxide equivalents; GHG, greenhouse gas.LDPE, Low density polyethylene; LDU, large laundry unit; N/A, not available;

Figure 2.

Greenhouse gas (GHG) emissions per laundry item. CO₂e, carbon dioxide equivalents; LDU, large laundry unit.

Scenario modelling and sensitivity analysis

Transport of laundry

Figure 3A–D model a scenario where each HGV type is progressively filled from 20% to 100% of its maximum payload. Each HGV is modelled as being filled with its maximum number of pallet cages across the model, with pallet cages progressively filled with more laundry items. The typical minimum and maximum degrees of filling are plotted to demonstrate the range of likely real-world HGV filling rates. This is based on the maximum degree to which pallet cages are permitted to be filled by LDU customers. There are lower GHG emissions per item when the rate of HGV filling with laundry items is increased. GHG emissions are particularly sensitive to HGV filling between 20% and 50%.

Figure 3.

Laundry unit transport emissions. CO₂e, carbon dioxide equivalents; GHG, greenhouse gas; HGV, heavy goods vehicle.

The typical minimum (worst) case scenario for pallet cage filling as a proportion of maximum HGV payload was 36.9%, 37.0%, 33.9% and 32.3% for 7.2, 12, 14 and 18 t HGVs, respectively. The GHG emissions per item kilometre for HGVs in this scenario were 0.238, 0.170, 0.147 and 0.176 g CO₂e/km, respectively. The typical maximum (best) case scenario for pallet cage filling as a proportion of maximum HGV payload was 60.9%, 61.2%, 56.0% and 53.2% for 7.2, 12, 14 and 18 t HGVs, respectively. The GHG emissions per item kilometre for HGVs in this scenario were 0.126, 0.0920, 0.0794 and 0.0959 g CO₂e/km, respectively. This represents a reduction in GHG emissions due to transport of 45%–47% per item if pallet cages are filled at the upper limit of what is commonly practised at the LDU.

Gas substitution

If gas-powered boilers and dryers were replaced with national grid electricity-powered alternatives, this would increase annual LDU GHG emissions by 269 t CO₂e based on the UK’s energy mix in 2022. This would increase boiler and drying emissions by 10.5%, and overall LDU emissions by 9.1%.

By linearly extrapolating the recent carbon intensity of UK electricity to 2030 using DEFRA/BEIS emissions factors (online supplemental appendix figure 3), one would expect emissions of 0.150 kg CO₂/kWh. In this 2030 scenario, using electricity-powered boilers instead of gas-powered boilers could result in a reduction in annual LDU GHG emissions of 940 t CO₂e. This would reduce boiler and drying emissions by 36.6% and overall LDU emissions by 31.9%, excluding transport. This assumes comparable energy performance of gas and electricity boilers and driers, whereas industrial electricity powered boilers are typically more efficient.¹³ It also assumes that national grid electricity could be acquired at the average carbon intensity.

Single-use alternatives

Table 3 compares per-use GHG emissions for six frequently used reusable laundry items with single-use alternatives. A range of round-trip HGV journey distances are modelled up to the maximum customer distance (342 km). The typical least and most efficient transport scenarios are modelled based on the variation in the degree of pallet cage filling.

Table 3.

Comparing the per-use life-cycle GHG emissions for reusable laundry items reprocessed at the LDU with single-use alternatives

	Item	Curtains	Blanket	Sheet	Patient gown	Scrub trousers	Scrub top
	Mass/kg	4.200	1.305	0.573	0.296	0.191	0.153
	Material	100% polyester	100% polyester	70% cotton, 30% polyester	50% cotton, 50% polyester	65% polyester, 35% cotton	65% polyester, 35% cotton
	Annual number processed	41 794	1 450 046	3 480 334	894 403	795 799	819 683
	Total GHG emissions per item use/ kgCO2e	2.376	0.732	0.341	0.173	0.110	0.089
Single-use alternative	Item	1.650 kg SB PP curtain	0.300 kg SB PP blanket	0.080 kg SB PP sheet	0.040 kg SB PP gown	0.0569 SB PP scrub trousers	0.0431 SB PP scrub top
	Per-use emissions/ kgCO2e	15.282	2.650	0.712	0.358	0.507	0.385
	Relative GHG emissions for single-use items compared with reusable equivalent
	On-site LDU (no transport)	643%	362%	209%	206%	459%	435%
Including median delivery round trip (141 km)	7.2 t HGV min filled	634%	346%	190%	173%	352%	315%
	7.2 t HGV max filled	638%	353%	198%	187%	395%	362%
	12 t HGV min filled	637%	350%	195%	181%	377%	342%
	12 t HGV max filled	640%	355%	201%	192%	411%	379%
	14 t HGV min filled	638%	352%	197%	184%	386%	352%
	14 t HGV max filled	640%	356%	202%	194%	417%	386%
	18 t HGV min filled	637%	350%	194%	180%	375%	339%
	18 t HGV max filled	640%	355%	201%	191%	409%	377%
Including maximum delivery round trip (342 km)	7.2 t HGV min filled	622%	326%	168%	140%	264%	227%
	7.2 t HGV max filled	632%	342%	185%	165%	330%	292%
	12 t HGV min filled	628%	335%	178%	154%	300%	262%
	12 t HGV max filled	635%	347%	191%	175%	357%	321%
	14 t HGV min filled	630%	338%	182%	160%	315%	277%
	14 t HGV max filled	636%	349%	193%	178%	368%	333%
	18 t HGV min filled	627%	334%	177%	153%	297%	259%
	18 t HGV max filled	634%	346%	190%	173%	354%	317%

^‘Min filled’ and ‘max filled’ refer to the typical minimum and maximum expected filling rate of pallet cages delivered to and from the laundry unit in current real-world practice.

GHG, greenhouse gas; HGV, heavy good vehicle; LDU, laundry unit; SB PP, spunbond polypropylene.

No modelled scenarios favoured any single-use item. If all six items were changed to single-use alternatives, the aggregate relative GHG emissions range from 244% (including 342 km round-trip in a minimally filled 7.2 t HGV) to 300% (on-site LDU with no HGV transport). Based on the number of LDU items reprocessed annually, this is equivalent to between 4723 and 5332 t CO₂e of additional GHG emissions.

Online supplemental appendix table 7 shows how reusable laundry items compare against single-use equivalents if reusable items were incinerated at the end of their lifetime. Incineration instead of recycling increased the per-use attributable GHG emissions by between 3.1% and 4.1% for reusable laundry. All modelled scenarios favoured reusable products, with relative single-use product GHG emissions varying between 137% and 618% of the reusable equivalent product’s emissions.

Discussion

This life-cycle inventory analysis describes the GHG emissions associated with a large hospital laundry unit, and makes estimates per laundry item type and per-unit mass. By presenting a breakdown of the relative GHG emission contributions from different processes, we identify where the greatest improvements can be made. The three foremost areas are:

1. Boiler and dryer electrification as renewable electricity generation increases.

2. Optimally loading delivery vehicles.

3. Avoiding single-use linen.

While the annual GHG emissions from the LDU are high, the per-item and per-unit mass GHG emissions are modest. For external comparison, our modelled GHG emissions per average item are comparable to the life-cycle emissions of four single-use protective hospital gowns, watching television for 1 hour or to the adult human metabolism over 7 hours.^14–16 Hospital laundry services support a broad range of healthcare activities. The estimates herein are therefore likely to be of utility to many parties interested in understanding how best to reduce healthcare’s considerable contribution to GHG emissions.

Natural gas use was the largest contributor to LDU GHG emissions and demonstrated the greatest potential for emissions reduction. We demonstrated that, based on the UK’s 2022 electricity supply, switching to electricity powered boilers and dryers would currently cause an increase in overall GHG emissions. This changes substantially as we look to the future, however, as the UK’s electricity supply is decarbonised in line with UK government targets.¹⁷ This demonstrates how forward planning and capital investment could help to reduce the NHS’ medium-term GHG emissions in line with its 2045 net-zero target.¹⁸ It could also save the NHS significant long-term financial costs as lower cost renewable energy becomes available.¹⁹ Steam produced using fossil fuels is commonly used across a wide range of different industries, and the technology already exists to efficiently generate steam at scale using electricity.²⁰ Renewable energy generation capacity and its transmission and distribution networks require ongoing development to support this predictable additional future demand.²¹ There is also potential to support further on-site renewable electricity generation as outlined by the Greener NHS and at overseas healthcare facilities.^{18 22} Energy communities might also augment access to non-national grid renewable energy, as seen in the European Union.²³

We demonstrated that the per-use life-cycle emissions for reusable items are substantially less than single-use alternatives. This exemplifies how the single-use economy often fails to meet the requirements of 21st century healthcare and society, as demonstrated in other studies.^{24 25} By changing six types of laundry item to single-use alternatives in Southwest England, the excess GHG emissions would cause predictable human harm.²⁶ Besides a public health duty to operate sustainably, heavy reliance on single-use items introduces supply chain vulnerability. This was witnessed particularly during the COVID-19 pandemic, but is a longer standing ongoing issue.^27–29 This analogously applies to the LDU’s heavy reliance on natural gas, as demonstrated by recent supply constraints and price inflation caused by Russia’s invasion of Ukraine.³⁰

The per-use GHG emissions for different laundry items can be used to estimate the degree to which changes in the way that hospitals use linen could reduce GHG emissions. Examples include avoiding the use of linen for alternative purposes, such as positioning patients for surgery.^{31 32} Positioning patients using reusable purpose-built devices that can be wiped clean might result in lower GHG emissions over time. A further source of variation in practice relates to the frequency of hospital bedding changes. No clear national guidance exists relating to this practice, and for patients occupying a hospital bed for >1 day unsoiled bedding might be changed anywhere between daily and weekly depending on local practices and protocols. There were >21 million hospital admissions involving two or more nights in hospital in England between 2020 and 2022. With GHG emissions to launder one blanket, sheet and pillowcase exceeding 1 kg CO₂e, it is evident that rationalised bed-changing practices could produce significant GHG emissions reductions.

Our sensitivity analysis incorporated a range of scenarios involving laundry transport to and from peripheral customer sites. This tested whether single-use items could be justified in locations very distant to a LDU, however we found that all modelled scenarios favoured reusable items. This is the case in scenarios where reusable items are incinerated at the end of their life, as well as when they are recycled.

There is slightly less than a twofold difference in the per-item transport-related GHG emissions depending on customer preference about how pallet crates are filled. Standardisation of pallet crate filling to an agreed safe maximum could therefore reduce GHG emissions by decreasing the frequency of HGV journeys required to certain units. Sufficient inventory of linen should therefore be held by each hospital to prevent hospitals running low on their linen supplies between deliveries. This approach would likely improve energy efficiency within the LDU as well, since linen is grouped according to type and customer, and each CBW compartment can hold 50 kg of dry linen. In instances where there is insufficient linen to fill a compartment, it will run underfilled. Stakeholder engagement and education is important in addressing GHG emissions, and these data might help those throughout the laundry value chain to understand their role.³³ Additional transport-related GHG emissions reductions might be realised through information technology facilitated route optimisation, and practising fuel-efficient ‘eco-driving’ .^{34 35}

Strengths and limitations

As for any LCA, we have used reasoned assumptions and boundaries to produce our estimates. We have presented results with a focus on transparency, such that our findings can inform action outside of Southwest England. While we explored major emissions hotspots, our scenario modelling did not incorporate different CBW units. Our impression is that the 14 and 16 container units used in Exeter probably confer favourable energy efficiency, however we have not demonstrated this through lack of external comparison. Caution therefore needs to be exercised when applying our findings to smaller LDUs, and we would welcome similar analysis of an LDU with smaller CBWs. LDUs with different energy mixes will also have a different environmental impact to that which we have outlined. Given that we allocated 100% of CHP gas use to steam production, and CHP electricity was not used at the LDU, we are likely to have modestly overestimated the GHG emissions that might reasonably be attributable to the LDU. Despite this, we have found reusable laundry processed at this LDU to result in lower GHG emissions compared with using single-use alternatives. Our mass-adjustment approach to estimate GHG emissions for different items is likely to be reasonably accurate, given how laundry is processed according to its mass.

We did not account for HGVs delivering to multiple destinations in nearby locations. We also estimated per-item transport emissions rather than per unit weight, so we could compare per-item use life-cycle estimates against single-use items. We took a conservative approach to modelling transport-related emissions by assuming that HGVs performed one journey leg without cargo. This is often not the case for the laundry unit studied, and therefore our transport-related emissions presented are likely an overestimate. Nonetheless, we have provided a sense of the magnitude of transport emissions, which is useful for LDUs and healthcare suppliers more broadly.

Lastly, our study period includes the COVID-19 pandemic, during which LDU throughput reduced transiently by around 27%. This will therefore underestimate LDU GHG emissions compared with different years, but it should not substantially alter per-item GHG emissions estimates.

Conclusion

This life-cycle assessment estimates the per-use GHG emissions for items reprocessed at a large laundry unit supplying numerous hospitals in Southwest England. We have demonstrated the potential for significant GHG emissions reductions by optimally filling transport vehicles, and switching from gas to electricity power as national grid electricity supply decarbonises. Substantial excess GHG emissions associated with single-use alternatives have been demonstrated.

Data availability statement

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

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

Not applicable.