Characteristics of health systems participating in the U.S. health sector climate pledge

Introduction

Background

Health systems, which have organizational missions to protect human health, are increasingly recognized as essential leaders in reducing carbon emissions and supporting community resilience. In recent years, growing attention has been paid to the human health risks from climatic changes associated with these emissions, which the World Health Organization has described as the single greatest threat to global health in the 21^st century (1). For example, reports estimate that health systems account for approximately four to five percent of global emissions (2), and the United States (U.S.) healthcare sector accounts for about eight to ten percent of national greenhouse gas emissions (3). In the U.S. alone, the pollution resulting from healthcare operations contributes to over 500,000 disability-adjusted life years lost (4), further highlighting how healthcare activities can undermine public health.

In 2021, the U.S. joined more than 50 other countries in agreeing to pursue the development of low-carbon, climate-smart health systems with the overarching goal of achieving sustainable, environmentally-friendly healthcare delivery (5). Furthermore, starting in 2022, the U.S. Department of Health and Human Services (HHS) began encouraging private-sector health systems to join federal health systems in pledging to reduce their greenhouse gas emissions at a pace consistent with Paris Agreement goals (6,7). In signing the voluntary HHS Health Sector Climate Pledge (henceforth, Climate Pledge), organizations made a non-binding commitment to cut greenhouse gas emissions by 50% by 2030 and achieve net-zero emissions by 2050. They also agreed to designate an executive leader or team to oversee these actions and develop a climate resilience plan to cope with future climate-related events that threaten public health and wellbeing. As of November 2024, health systems representing a total of 960 U.S. hospitals had signed the Climate Pledge, the final count that was publicly available before the dissolution of the program in 2025. According to Rogers’ Diffusion of Innovations theory (8), this total would be associated with the end of the early adopter stage and the beginning of the early majority stage—a time when adoption should accelerate to other organizations, particularly those viewing themselves as most similar to the early adopters.

In this study, we draw on Diffusion of Innovations theory as a conceptual lens to frame trends in Climate Pledge adoption in the U.S. healthcare sector. This theory provides a valuable perspective on understanding how innovations, such as commitments to emission reduction goals, can spread across health systems. The Diffusion of Innovations theory proposes that the adoption and subsequent diffusion of a new activity occurs across five dimensions: relative advantage (the perceived superiority of an innovation), compatibility with adopters, complexity of the innovation, trialability (the degree to which the innovation can be tested and modified), and observability (the degree to which the innovation may be viewed by third parties) (8). These underlying dimensions inform how groups of individuals or institutions evaluate an innovation’s scalability and diffusion. Practically speaking, this framework may help explain why certain healthcare organizations make public commitments to reducing greenhouse gas emissions, especially in relation to differing political and socioeconomic contexts.

Rationale and knowledge gap

Understanding underlying drivers for adopting emissions reduction goals could help with identifying organizations most likely to adopt these goals in the future. This knowledge could, in turn, help ensure efforts to spread sustainable practices are targeted for greatest efficiency. Researchers and practitioners have previously noted that most of the discourse surrounding strategies to reduce and potentially eliminate greenhouse gas emissions has focused primarily on how individual physician practices and hospitals can achieve lower emissions (9-12). However, it is unlikely that significant progress will be made on achieving net-zero emissions unless coordinated efforts across hospitals and health systems can help stakeholders find alignment on strategies and goals to reduce and mitigate emissions (13). Furthermore, studying health systems that adopt emissions reduction targets could provide useful information about the system-wide changes across hospitals and localities necessary to foster buy-in and remove barriers (14) that inhibit health systems from making progress towards decarbonizing their operations.

Objective

The purpose of our exploratory study was to examine the relationship between health system, socioeconomic, and Climate Pledge signatory status to better understand the factors associated with current signatory status, signaling health systems’ public commitment to decarbonization efforts. As healthcare leaders and executives continue to learn about the impact of healthcare operations on the environment (15), research that raises awareness about the factors relating to organizational action on reducing greenhouse gas emissions can also provide information that enables leadership to mobilize resources and support to achieve net-zero operations. This current study takes that initial step and explores the characteristics of early health system Climate Pledge signatories. We present this article in accordance with the STROBE reporting checklist (available at https://jhmhp.amegroups.com/article/view/10.21037/jhmhp-25-27/rc) (16).

Methods

Study design and data sources

We conducted a retrospective, cross-sectional study analyzing U.S. health systems. Given our focus on the health system-level of analysis, we used the Agency for Healthcare Research and Quality (AHRQ) Compendium of U.S. Health Systems (17), which provided information about the structure, staffing, and characteristics of 640 U.S. health systems as of 2022. The AHRQ Compendium dataset, based on the IQVIA OneKey and American Hospital Association (AHA) Annual Survey databases (source year 2021), operationalized a health system as having at least one non-Federal acute care hospital, at least 50 physicians (in total), and at least 10 primary care physicians. Additional information regarding how organizations are aggregated into systems can be found at the AHRQ Compendium website (17).

Variables were selected a priori based on our exploratory interest in organizational and sociopolitical characteristics that may influence climate pledge adoption. We aimed to assess a diverse set of potentially relevant factors informed by prior research and sector-specific considerations, rather than relying on statistical selection techniques to determine inclusion in the multivariate model. Health system characteristics including multi-state system status, bed size, total patient revenue as reported in the Healthcare Cost Report Information System (HCRIS, source year 2022), major teaching status, having at least one hospital with a high disproportionate share hospital (DSH) patient percentage (i.e., in the top quintile of DSH patient percentage among hospitals paid under the Medicare inpatient prospective payment system (IPPS), and system ownership type were collected from the AHRQ Compendium dataset.

The HHS Health Sector Climate Pledge website provided a list of health system signatories as of November 2023, which was used to create our dichotomous dependent variable of signatory status (18). We cross-referenced the list of signatory organizations that were “health systems, hospitals, and other providers” from the HHS Climate Pledge website with the AHRQ Compendium dataset. A one-to-one merging process was conducted by matching organization names across the two datasets. When discrepancies in naming conventions or organizational identifiers arose, two co-authors manually reviewed and verified organizational affiliations using publicly available sources (e.g., health system websites) to ensure the accuracy of the dataset merge.

The U.S. News and World Report (USNWR) website provided data on hospital honor roll status, as of January 2024, which is based on patient outcomes, the quality of care, and hospital staffing (19). The U.S. Cooperative for International Patient Programs (USCIPP) website provided data, as of January 2024, on U.S.-based academic medical centers, hospitals, and health systems that offer international patient programs and care for patients who travel to the U.S. for care (20), which provided a proxy measure for the presence of international commercial activity. More specifically, if one or more hospitals affiliated with a specific health system were identified as being a USNWR Honor Roll hospital, we attributed the Honor Roll status to the health system. Similarly, if a USCIPP member hospital, based on the information provided on the website mentioned above, we attributed the USCIPP membership to the health system. The carbon intensity of the economy, by state as of 2021, was gathered from the U.S. Energy Information Administration (EIA) (21). Political ideology mix, by state as of 2021, was collected from the Pew Research Center (22), and categorized according to whether the largest proportion of respondents was conservative, moderate, or liberal. State-level characteristics were assigned to each health system based on where the health system’s headquarters were located.

Statistical analyses

The individual datasets described above were merged using Excel Power Query (23) and then uploaded to R statistical software (24) for analysis. We conducted descriptive and bivariate statistics including Wilcoxon Rank Sum Test, Pearson’s Chi-squared test, and Fisher’s Exact Test. We assessed multicollinearity among covariates by calculating the Generalized Variance Inflation Factor (GVIF). We used the adjusted GVIF^[1/(2*Df)] transformation to account for variables with multiple degrees of freedom. For reference, the variables that exceeded a conservative threshold of 1.5 were carbon economy (1.52), system bed size (5.21), and hospital net revenue (5.23), which suggests the presence of multicollinearity. However, we retained all predictors in the final multivariate model due to their relevance in supporting our exploratory objectives. We also applied a Bonferroni correction (adjusted significance threshold of P<0.00357, based on 14 variables) as a sensitivity check for potential Type I error inflation. Binary logistic regression analyses were conducted to examine the probability of a health system signing the HHS Climate Pledge (1= yes, was a HHS Climate Pledge signatory; 0= no, was not a HHS Climate Pledge signatory). McFadden’s R² and Log Likelihood were calculated to assess model fit. We sought to minimize potential sources of bias by including all eligible health systems that met the study criteria and conducting multivariate analyses to adjust for potential confounding factors. We report both significant and non-significant findings to reduce reporting bias. The Institutional Review Board of Rush University Medical Center reviewed our study protocol and determined it did not constitute human subjects research.

Results

Descriptive and bivariate findings

Our final analytical sample included 635 health systems. Five health systems were excluded from the final sample due to missing data on key variables [i.e., system beds (n=5), system teaching hospital status (n=5), DSH status (n=5), hospital net revenue (n=5), and system ownership status (n=1)]. Table 1 summarizes the characteristics of the 67 health systems that signed the Climate Pledge (i.e., pledged health systems) and the 568 that did not (i.e., unpledged health systems).

Our bivariate analysis identified several statistically significant differences between pledged and unpledged health systems (Table 1): Compared to Non-Signatory health systems, a greater proportion of Signatory health systems had international commercial activity, as measured by USCIPP membership (19.0% vs. 4.2%, P<0.001) and were named to the USNWR Honor Roll (21.0% vs. 2.5%, P<0.001). Similarly, a greater proportion of multi-state health systems that spanned three or more states had signed the pledge compared to unpledged health systems (16.0% vs. 5.5%, P=0.005). Health system bed sizes also differed significantly, as pledged health systems had a median of 1,141 beds (IQR =1,827) and unpledged health systems had a median of 365 beds (IQR =604.8) (P<0.001). Furthermore, compared to unpledged health systems, it was more common for pledged health systems to have at least one major teaching hospital (P<0.001) and at least one high DSH patient percentage hospital (P<0.001) and higher median hospital net revenue (P<0.001). There was no statistically significant difference between pledged and unpledged health systems with respect to system ownership (P=0.82). Lastly, in regards to regional and socio-political factors, pledged health systems tended to be headquartered in states with lower carbon economies (196.3 vs. 300.4 metric tons of energy-related CO₂ per chained 2012 million dollars of GDP; P<0.001) and in states with a higher percentage of residents identifying as politically liberal or moderate (P<0.001).

Multivariate findings

The results of the binary logistic regression analysis, found in Table 2, indicated that having at least one major teaching hospital [odds ratio (OR) =2.3753; 95% confidence interval (CI): 1.1590, 4.9925; P=0.02] was associated with higher odds of being a pledged health system, and thus, major teaching health systems had nearly 2.38 times higher odds of pledging compared to non-major teaching systems. Also, hospital net revenue and system bed size were significant predictors of pledge status: For every ten million US$ increase in net revenue there was a 0.52% increase in signatory status (OR =1.0052, 95% CI: 1.0022, 1.0086; P=0.001), while each additional system bed was associated with a 0.06% decrease in the odds of pledging (OR =0.9994, 95% CI: 0.9989, 0.9998; P=0.005). Although larger health systems appeared more likely to pledge in bivariate analyses, this association reversed in the multivariate model after adjustment for other organizational characteristics. This pattern reflects confounding and collinearity among predictors (e.g., system bed size VIF =5.21), a situation consistent with Simpson’s paradox. In practical terms, this means the overall trend (larger systems pledging more often) masks a different underlying pattern once other factors are considered. Furthermore, being headquartered in a state in which the majority of the population identified as politically liberal was associated with a statistically significant higher odds of being a pledged health system (OR =7.8354; 95% CI: 2.1525, 28.1049; P=0.002). However, other factors, for example, USCIPP membership, USNWR Honor Status, multi-state system status, being categorized as a high DSH system, and carbon economy were not significantly associated with Climate Pledge signatory status. Furthermore, it is worth noting that the cross-sectional nature of the data inhibits causal inference, which we discuss further in subsequent sections.

Discussion

Key findings

Our exploratory study examined the relationships between organizational and socioeconomic factors related to U.S. health systems’ signing of the HHS Health Sector Climate Pledge. Our multivariate analysis findings underscore the importance of the health system types and contexts in predicting voluntary commitment to the Climate Pledge. For instance, the findings from our analysis revealed that teaching-intensive and higher patient revenue-generating health systems had higher odds of committing to the decarbonization goals set forth by the Climate Pledge relative to their peers, while larger health systems—measured by bed size—were slightly less likely to be signatories. Furthermore, health systems headquartered in liberal majority states had higher odds of being a Signatory system. We explain below how our findings contribute to the broader literature on health systems’ adoption of decarbonization efforts and our understanding of the organizational factors that may influence the widespread commitment of health systems to reduce greenhouse gas emissions and improve climate resiliency. We also acknowledge that the cross-sectional design of our study precludes causal inference, creating an opportunity for more rigorous longitudinal research on voluntary commitments to improving climate resiliency.

Comparison with similar research

As noted previously (25,26), early adopters of innovations or new initiatives exhibit specific cultures and capabilities that enable them to approach change proactively. These organizations typically serve as thought leaders within their broader network and can influence the adoption decisions of other organizations (27). As our findings indicate, U.S. health systems that already demonstrate a commitment to educating and training clinicians, represented by their teaching intensity, are more likely to voluntarily reduce greenhouse gas emissions as set forth by the Climate Pledge. This could suggest that academic medical centers, for example, may perceive an advantage of being an early adopter of the Climate Pledge because it may align with their existing values and practices to ensure efficient and high-quality care. However, we caution against overinterpreting this result and recommend that future studies further explore this relationship. Furthermore, we found an association between hospital net revenue of the system and signatory status, which means that having potentially more resources and capacity to absorb the changes necessary to support decarbonization efforts could potentially influence system signatory status. However, based on our findings, larger and more geographically dispersed health systems may face greater barriers to pledge adoption than individual hospitals such that larger systems, despite having more resources, may require greater coordination in planning and implementing initiatives to achieve net-zero emissions, which could represent a barrier for these health systems to sign the pledge.

Health systems operate within a broader socio-political context such that political will, stakeholder engagement, and regulations can impact their ability to commit to the Health Sector Climate Pledge emission reduction goals (28). If climate change is not recognized as a high priority in the broader context in which the health system operates, it may be difficult to allocate resources where the need is highest and, thus, comply with long-term sustainability efforts. Alternatively, based on our findings, health systems headquartered in states in which the population predominantly identifies as liberal may suggest a broader readiness by local and state-level stakeholders to support health system decarbonization commitments. Additionally, the limited adoption of the Health Sector Climate Pledge by health systems may also suggest a broader unreadiness by leaders to invest time and scarce resources to develop and implement plans to reduce emissions or modernize their infrastructure for climate resiliency. Nonetheless, achieving emissions reduction goals is a complex and long-term endeavor, but more rapid progress is achievable when the goals and priorities of the socio-political and health system environments are aligned.

Future research and practical implications

Additional research would be helpful in several ways. For example, research on the perspectives of healthcare leaders and managers of U.S. health systems would be helpful in gaining a deeper understanding of specific challenges faced by decision-makers. Research involving executive leaders from subdisciplines such as information systems, finance, and supply chain, could help with our understanding of how these disciplines interoperate in sustainability decision-making. Qualitative studies would also provide valuable information that could help develop guidance and actionable strategies for health systems seeking to advance decarbonization efforts and describe a broader range of Climate Pledge decision considerations and criteria (e.g., purchasing and supply chain optimization) to better understand why health systems sign or don’t sign onto the Climate Pledge.

Broader and more systematic outreach by leaders from early-adopter organizations, through educational programs and associations (29), could potentially accelerate the number of health systems that commit to climate resiliency and reduce the uncertainty surrounding a public commitment to achieving net-zero emissions (30). For example, peer networks such as Health Care Without Harm (31), an international coalition of health systems and civil society organizations, provides training and resources to help health care facilities reduce their emissions and implement climate mitigation strategies. By engaging with peers affiliated with Practice Greenhealth (32), for example, the early majority can tap into a network of more than 1,400 hospitals and health systems across North America that are actively piloting decarbonization projects and implementing innovative measures to reduce greenhouse gas emissions. These coalitions, as well as the platforms they use to promote climate resiliency, could help educate the early majority about effective implementation strategies that can sustain decarbonization efforts and offer guidance to leaders (e.g., Sustainability Officers) who will oversee implementation efforts. The engagement and collaboration of stakeholders across different public and private sector organizations is essential to the development of strategies that result in broader reductions of greenhouse gas emissions by health systems.

Limitations

Our study has several important limitations that may limit the generalizability of our results. First, our data represent a cross-section of U.S. health systems and Climate Pledge signatories. While we are able to draw inferences about potential relationships, we cannot say with certainty the factors identified in our analyses were causal factors in health systems’ decisions to sign the pledge. There may be confounding variables outside the scope of our measures that are associated with both organizational characteristics and pledge adoptions—for example, pre-existing leadership commitments as well as state and local policy incentives. Additionally, while we used Diffusion of Innovations to frame the landscape of Climate Pledge adoption in the U.S., our study did not empirically test the theory’s constructs. Future research is needed to more directly apply the Diffusion of Innovations constructs in the context of health systems adopting emission reduction goals.

Second, our measure of signatory status represents a proxy for health system engagement in emissions reduction efforts and does not represent the extent to which a health system is implementing decarbonization initiatives or their success in reducing greenhouse gas emissions. More granular data on the implementation of initiatives at the health system level could allow researchers and practitioners to develop more precise and qualitatively distinct typologies of health systems which would also help better distinguish between health systems that have signed the Climate Pledge or made other voluntary commitments to reduce greenhouse gas emissions. Similarly, since we did not have information about ongoing initiatives (e.g., reduction of anesthesia use or standardizing of waste removal), it is unclear whether existing activity may have reduced barriers for health systems that were already implementing these changes to sign the pledge. Thus, while the use of proxies for organizational and contextual factors enhanced the pragmatism of our research, they do not adequately capture the internal context of health systems, which can influence health systems’ decisions to commit to improving climate resiliency. For instance, collecting more granular data on how health systems monitor and assess water, sanitation, and waste management practices (33) could provide more direct evidence about health systems’ readiness, or lack thereof, to sign voluntary climate pledges.

Third, our study takes place in a single country, the U.S. Whether our findings translate to other international contexts that operate under different circumstances (e.g., single-payer healthcare systems like the United Kingdom National Health System) warrants investigation. For example, several countries have made commitments, in alignment with the Paris Agreement, to make health systems climate-resilient by submitting nationally determined contributions (NDCs). These NDCs (e.g., “The need to transition to renewable energy sources in hospitals and for sustainability of health care facilities”) (5) may signal to health systems about their country’s priorities related to delivering on the Paris Agreement, which could also influence if and how the health systems in these countries announce their intentions to commit to climate resiliency and take the necessary steps to develop and implement policies that support greenhouse gas mitigation and adaptation (5).

Fourth, given the small number of health systems headquartered in states where most of the population identified as liberal (i.e., approximately 4% of the study sample), estimates involving this category should be interpreted with caution due to potential instability. And, finally, given the voluntary nature of the pledge and our inability to prospectively ascribe signatory status to individual health systems, our findings may be influenced by selection bias and thus reduce their generalizability.

Conclusions

Our analysis underscores the critical role that both health system characteristics and context play in influencing the adoption of emission reduction goals. To accelerate wider adoption, early adopters should actively engage peers to effectively address uncertainty and reinforce broader commitments to climate resiliency. Future work should investigate additional decision-making considerations and criteria, such as purchasing practices and supply chain optimization, to deepen our understanding of why some health systems choose to commit to emission reduction goals.

Acknowledgments

Parts of the analysis contained in this manuscript have been presented elsewhere at the 2024 CleanMed Conference as a poster presentation.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jhmhp.amegroups.com/article/view/10.21037/jhmhp-25-27/rc

Peer Review File: Available at https://jhmhp.amegroups.com/article/view/10.21037/jhmhp-25-27/prf

Funding: This work was supported by a research seed grant from the College of Health Sciences at Rush University.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jhmhp.amegroups.com/article/view/10.21037/jhmhp-25-27/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The Institutional Review Board of Rush University Medical Center reviewed this study’s protocol and determined it did not constitute human subjects research.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. World Health Organization. WHO calls for urgent action to protect health from climate change – sign the call. Geneva: World Health Organization; 2015. Available online: https://www.who.int/news/item/06-10-2015-who-calls-for-urgent-action-to-protect-health-from-climate-change-sign-the-call
2. Rodríguez-Jiménez L, Romero-Martín M, Spruell T, et al. The carbon footprint of healthcare settings: A systematic review. J Adv Nurs 2023;79:2830-44. [Crossref] [PubMed]
3. Eckelman MJ, Sherman J. Environmental Impacts of the U.S. Health Care System and Effects on Public Health. PLoS One 2016;11:e0157014. [Crossref] [PubMed]
4. Eckelman MJ, Huang K, Lagasse R, et al. Health Care Pollution And Public Health Damage In The United States: An Update. Health Aff (Millwood) 2020;39:2071-9. [Crossref] [PubMed]
5. Blom IM, Beagley J, Quintana AV. The COP26 health commitments: a springboard towards environmentally sustainable and climate-resilient health care systems? J Clim Chang Health 2022;6:100136.
6. Executive Office of the President of the United States. Climate and health: 2024 year in review. Washington, DC: Office of Science and Technology Policy; 2025 Jan 13 [cited 2025 Mar 31]. Available online: https://bidenwhitehouse.archives.gov/ostp/news-updates/2025/01/13/climate-and-health-2024-year-in-review/
7. TechTarget. HHS relaunched health sector climate Pledge . Newton, MA: TechTarget, Inc.; 2023 Apr 04 [cited 2025 Aug 16]. Available online: https://www.techtarget.com/pharmalifesciences/news/366606445/HHS-Relaunched-Health-Sector-Climate-Pledge
8. Rogers EM. Diffusion of innovations. 4th ed. New York: Free Press; 1995.
9. Mortimer F. The sustainable physician. Clin Med (Lond) 2010;10:110-1. [Crossref] [PubMed]
10. Health Care Climate Council. Climate action playbook for hospitals. Boston: Health Care Without Harm; [cited 2025 Mar 31]. Available online: https://climatecouncil.noharm.org/
11. American Medical Association. Fight climate change and save money—go green at your practice. Chicago: AMA; 2022 Apr 19 [cited 2025 Mar 31]. Available online: https://www.ama-assn.org/delivering-care/public-health/fight-climate-change-and-save-money-go-green-your-practice
12. Crowley R, Mathew S, Hilden D, et al. Environmental Health: A Position Paper From the American College of Physicians. Ann Intern Med 2022;175:1591-3. [Crossref] [PubMed]
13. Chen C, Miller G, Setoguchi S. Climate change and excess length of stay: A call to action for health equity and environmental sustainability. J Hosp Med 2024;19:1059-65. [Crossref] [PubMed]
14. Nilsen P, Seing I, Ericsson C, et al. Characteristics of successful changes in health care organizations: an interview study with physicians, registered nurses and assistant nurses. BMC Health Serv Res 2020;20:147. [Crossref] [PubMed]
15. Lee VS, Gerwig K, Hough E, et al. Decarbonizing health care: engaging leaders in change. NEJM Catal Innov Care Deliv 2023;4.
16. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Epidemiology 2007;18:800-4. [Crossref] [PubMed]
17. Agency for Healthcare Research and Quality. Compendium of U.S. health systems, 2022. Rockville (MD): AHRQ; 2022. Available online: https://www.ahrq.gov/chsp/data-resources/compendium-2022.html
18. Health Care Without Harm. HHS health sector climate pledge. Reston (VA): Health Care Without Harm; [cited 2025 Mar 31]. Available online: https://us.noharm.org/climate-and-health/hhs-health-sector-climate-pledge
19. U.S. News & World Report. Best hospitals honor roll and overview. Washington (DC): U.S. News & World Report; Available online: https://health.usnews.com/health-care/best-hospitals/articles/best-hospitals-honor-roll-and-overview
20. National Center for Healthcare Leadership. US Cooperative for International Patient Programs (USCIPP). Chicago: NCHL; [cited 2025 Mar 31]. Available online: https://www.nchl.org/uscipp/
21. U.S. Energy Information Administration. State carbon dioxide emissions data. Washington (DC): U.S. Department of Energy; [cited 2025 Mar 31]. Available online: https://www.eia.gov/environment/emissions/state/
22. Pew Research Center. Religious landscape study. Washington (DC): Pew Research Center; [cited 2025 Mar 31]. Available online: https://www.pewresearch.org/religious-landscape-study/
23. Microsoft Corporation. Power Query documentation. Redmond (WA): Microsoft; [cited 2025 Mar 31]. Available online: https://learn.microsoft.com/en-us/power-query/
24. R Core Team. R: a language and environment for statistical computing. Vienna (Austria): R Foundation for Statistical Computing; 2024 [cited 2025 Mar 31]. Available online: https://www.r-project.org/
25. Greenhalgh T, Robert G, Macfarlane F, et al. Diffusion of innovations in service organizations: systematic review and recommendations. Milbank Q 2004;82:581-629. [Crossref] [PubMed]
26. Grol R, Wensing M, Eccles M, et al., eds. Improving Patient Care: The Implementation of Change in Health Care. 2nd ed. Chichester, UK: Wiley Blackwell; 2013.
27. Balas EA, Chapman WW. Road Map For Diffusion Of Innovation In Health Care. Health Aff (Millwood) 2018;37:198-204. [Crossref] [PubMed]
28. Lakatos K, Teherani A, Thottathil SE, et al. A race to net zero—early lessons from healthcare’s decarbonization marathon. Health Aff Sch 2023;1:qxad006. [Crossref] [PubMed]
29. Molthan-Hill P, Blaj-Ward L, Mbah MF, et al. Climate change education at universities: relevance and strategies for every discipline. In: Lackner M, Sajjadi B, Chen W, editors. Handbook of climate change mitigation and adaptation. Cham: Springer International Publishing; 2022. p. 3395-457.
30. Seervai S, Gustafsson L, Abrams MK. The impact of climate change on our health and health systems. New York: Commonwealth Fund; 2022 May 4 [cited 2025 Mar 31]. Available online: https://www.commonwealthfund.org/publications/explainer/2022/may/impact-climate-change-our-health-and-health-systems
31. Health Care Without Harm. Our strategies. Reston (VA): Health Care Without Harm; [cited 2025 Mar 31]. Available online: https://noharm.org/our-strategies
32. Practice Greenhealth. Hospitals and health systems. Reston (VA): Practice Greenhealth; [cited 2025 Mar 31]. Available online: https://practicegreenhealth.org/membership/hospitals-and-health-systems
33. Mosadeghrad AM, Isfahani P, Eslambolchi L, et al. Strategies to strengthen a climate-resilient health system: a scoping review. Global Health 2023;19:62. [Crossref] [PubMed]