Rationale

The healthcare sector ought to promote human health and wellbeing. Yet, healthcare has a significant global ecological impact with harmful downstream effects on health.

[1–4] Greenhouse gas emissions (GHG) from the US health care system climbed to 8.5% of total US GHG in 2018. Pollution from the US health care sector including GHG, PM2.5, and ozone exacted a toll of 388,000 disability-adjusted life-years in 2018 alone [3]. Global health care sector pollution was associated with the loss of 4 million DALYs in 2020 [5]. These serious health harms are preventable and pose an opportunity for the health sector to demonstrate leadership in the sustainability transformation.

Healthcare providers and organizations are increasingly turning to quality improvement (QI) as the science of systematic and continuous improvement to healthcare processes in attempts to mitigate the ecological impacts of healthcare [6, 7]. Existing efforts, while still early, focus mostly on climate impacts [8]. However, well-intentioned interventions to mitigate climate change may trade off against negative impacts on other earth system processes [9]. Intravenous anesthetics can sometimes be substituted for inhaled anesthetics with high global warming potential, but the common intravenous anesthetic propofol is often improperly disposed of, having toxic effects on aquatic life that cannot be ignored [10]. In this way, the planetary boundaries literature makes clear that a truly sustainable healthcare sector must attend to its relationship with an array of earth system processes [9].

Unfortunately, we know little at present about how QI initiatives have identified and measured the full range of ecological impacts–to the climate and beyond [11]. Indirect evidence from the healthcare life cycle analysis literature suggests that non-climate ecological impacts may be neglected in QI initiatives. The HealthcareLCA living database has catalogued studies of environmental impact of healthcare, noting that 99% (151/152) of studies catalogued climate impacts but only 55% (83/152) of studies evaluated any other impact category (such as acidification potential, particulate matter formation, and freshwater ecotoxicity potential, among others) [12]. While suggestive of a relative neglect in QI initiatives, the evidence from the LCA literature falls short of supporting any strong claim about prevailing approaches to measuring ecological impact in the sustainable QI literature.

Given that robust and meaningful measures are a cornerstone of QI, and furthermore that unrecognized tradeoffs may stymie efforts to build sustainable health systems, we must better understand how existing QI efforts to address environmental sustainability in healthcare have attempted to identify and measure ecological impacts. The primary aim of the proposed scoping review is thus to investigate this question and determine the gaps and areas of emphasis across this emerging body of literature. By mapping existing approaches to measuring ecological impact in the sustainable QI literature, we can suggest ways to leverage strengths, address weaknesses and inform future clinical and scholarly work.

Eligibility criteria

Clinically-led quality improvement (QI) or clinical innovation efforts taking place in all human healthcare delivery settings (acute care, post-acute care, ambulatory care) will be eligible for inclusion. We define clinical innovation efforts as ideas, services, or treatments used in a clinical setting that display benefits over current practices [17, 18]. Interventional studies will be included. The concept of interest is the exploration of metrics used to measure environmental impact in QI and clinical innovation efforts in the healthcare setting. Studies attempting to address environmental sustainability and that provide a quantitative measurement of ecological impact will be included. Measurement of ecological impact may be done directly or indirectly. We define direct metrics of environmental impact to include metrics such as kilograms of carbon dioxide equivalent greenhouse gas or water utilization. We define indirect metrics of ecological impact as clinical metrics with environmental impacts, such as the number of medication vials avoided, or the weight of plastics captured for recycling. We expect to find a broad range of metrics, beyond those identified in key papers retrieved from our initial search.

Modelling studies, reviews, case reports, editorials, letters, commentaries, and other publications that do not report any primary data or present new analyses of existing data will be excluded. Studies describing QI or clinical innovation efforts taking place outside of the clinical activity in the healthcare setting (i.e., physical plant, exclusively operations or engineering) will be excluded. Studies that are not peer-reviewed or published in languages other than English will also be excluded.

Search strategy

We have designed a comprehensive search strategy in collaboration with experienced information specialists and a quality improvement research synthesis expert. An initial search was completed on MEDLINE to retrieve relevant articles. Subject headings and keywords used in titles and abstracts of the articles of interest were used to create a full search strategy. The terms capturing QI and clinical innovation efforts included variations of the terms “quality improvement” and “clinical innovation”. The terms capturing ecological impact included “carbon”, “waste”, “overuse”, “environment”, “sustainability” and variations of these terms related to both direct and indirect ecological impacts. The healthcare setting was captured using variations of the term “healthcare”. The strategy was validated through the retrieval of several key studies, determined a priori to meet inclusion criteria. The full MEDLINE search strategy is detailed in Appendix 1 (“S1 File”). The full search strategy will be translated into all included databases.

Included search results were limited to those published in English and on or after the year 2000.

We limited results to those published on or after the year 2000 because this was the year the influential To Err is Human report was published and QI emerged internationally as a field of study [19]. Furthermore, the environmental impacts of healthcare have not been widely published until the last two decades [6, 20, 21].

Information sources

MEDLINE, EMBASE, CINAHL, and Scopus will be searched during August 2023 with database-specific search strategies for peer-reviewed published literature. The MESH terms selected for the search in MEDLINE were adapted to corresponding headings in each database.

Grey literature

Patient Safety Network (PSNet), Healthcare Without Harm, Cascades, UK Sustainability in Quality Improvement (SusQI), Choosing Wisely Canada National Meeting Abstract Books, and Institute for Healthcare Improvement (IHI) will be searched using keyword combinations derived from our main literature search. These organizations and websites were selected based on group discussion and input from subject matter experts that they are subject domain leaders in quality improvement and sustainability.

Data management

Covidence will be used to manage duplicates and screen titles, abstracts, and full-texts. Piloted Microsoft Excel spreadsheets will be used to chart the extracted data.

Selection process

Two independent reviewers will conduct two-stage screening in duplicate using piloted screening forms at each stage of screening. The first stage is abstract and title review, with eligibility confirmed by two independent reviewers. Any disagreements will be resolved through discussion and consensus, involving a third reviewer if consensus cannot be achieved. The second screening stage will be conducted on full text articles, again in duplicate with disagreements resolved through discussion, consensus, and similarly involving a third reviewer if consensus cannot be achieved.

Data collection process

Data will be extracted in duplicate by two independent reviewers using a pilot-tested data abstraction form. The pilot-tested abstraction form will be run for a random sample of ~30 articles by reviewers to calibrate internal consistency among reviewers. Resulting changes will be made before continuing extraction for the remaining articles. Any disagreements will be resolved through discussion and consensus, involving a third reviewer if necessary. As the two reviewers proceed, the data charting form will be revised iteratively as needed.

Data items

Data items are described in Table 1. We expect that many studies will include multiple outcomes. In these cases, we will extract data for all reported outcomes.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Data abstraction table.

Items planned for abstraction from the fulltext results.

https://doi.org/10.1371/journal.pone.0309417.t001

Data synthesis

We will present data narratively. We will also present a measurement framework to organize the data items described in Table 1, a figure to illustrate publications over time, and a diagram to illustrate the overlap between potential metrics of environmental impact and the metrics of environmental impact utilized in our dataset. Potential metrics of ecological impact will be identified from key references in life cycle assessment literature [8, 18].

Institutional review board statement

Not required in the setting of a review of published literature.