Transportation-Related Carbon Footprint of Coronary Heart Disease Ambulatory Care in the United States

As the leading cause of death in the United States (US), managing coronary heart disease (CHD) is not only a critical objective of the healthcare system but a source of resource utilization. With the US emitting a quarter of the world’s healthcare-related carbon footprint, the impact of healthcare access on travel-related carbon emissions must also be examined.^1,2 Thus, we investigated the transport-related carbon emissions for routine ambulatory care for CHD in the US.

We obtained data on the crude prevalence (%) of CHD (patients ≥ 18 years) from the Centers for Disease Control (CDC) PLACES database – 2022. National Provider Identifier records from the National Plan and Provider Enumeration System were obtained in July 2023. We identified cardiologists and their practice business addresses (geocoded to longitude and latitude). We calculated the Haversine distance between each census tract (CT) centroid (2010 Census) and the nearest cardiologist. We obtained the estimated CO2 emissions from a typical passenger vehicle from the Environmental Protection Agency (EPA): tailpipe CO₂ emission of 248.55 grams (g) per km, based on a fuel consumption of 1 gallon per 35.73 km.³ CT were considered urban if they had at least 2,000 housing units or had a population of at least 5,000 (Census Bureau). We obtained the 2018 Social Vulnerability Index (SVI) from the Agency for Toxic Substances and Disease Registry. SVI is a measure that provides a relative indication of the vulnerability of CTs based on 16 social factors, with greater values indicating increased vulnerability.

We calculated the total CT tailgate CO2 emissions (in metric tons, MTCO2) for a round trip to the nearest cardiologist; referred to CHD ambulatory care transport-related Carbon Footprint (ACT-CF) in this paper, per year, calculated as:

$$\left[\frac{\left(CTdistanceinKmtonearestcardiologist\right)\mathrm{*}248.55gCO2\mathrm{*}\left(numberofindividualswithCHDintheCT\right)\mathrm{*}2}{\mathrm{1,000,000}}\right]$$

The estimated number of individuals in a CT with CHD was obtained by multiplying the crude CHD prevalence (%) by the adult CT population. We reported the national estimate of the ACT-CF as the median and interquartile range (IQR), weighted by the number of individuals with CHD in the CT. We reported results separately for rural and urban CT and stratified our analysis by SVI quartiles. As a comparator, we estimated the total and national median emissions if the follow-up was provided by primary care providers (PCP, defined as internal medicine or family medicine specialty). We used R4.3.0 for statistical analyses and maps. Due to the use of publicly available data, ethical approval was not required.

The ACT-CF was calculated and mapped for 13,381,715 CHD patients in 69,328 CTs linked with 49,587 cardiologists in the contiguous US (Figure 1A), totaling 63,005 MTCO2. Patients residing in urban CTs contributed 27,892 MTCO2, and rural CTs 35,113 MTCO2. The national median ACT-CF was 0.42 MTCO2 (IQR: 0.17, 1.31). For urban CTs (10,309,198 CHD patients), median emissions were 0.29 MTCO2 (IQR: 0.13, 0.61), while for rural residents (3,072,517 CHD patients), it was 2.13 MTCO2 (IQR: 1.26, 3.59). Figure 1B shows the median ACT-CF by SVI quartile according to urban and rural settings. In comparison, emissions would total 21,747 MTCO2, with a national median of 0.16 MTCO2 (IQR: 0.07, 0.46) if patients had similar number of visits with the nearest PCP.

Figure 1:

A) Map of the census tract-level tailpipe CO₂ emissions in metric tons (MTCO2) associated with patients with CHD traveling to the nearest cardiologists (CHD ambulatory care transport-related carbon footprint). B) CHD ambulatory care transport-related carbon footprint stratified by setting and according to social vulnerability index (SVI) quartiles.

We identified substantial heterogeneity between urban and rural emissions, with rural residents emitting approximately 7.3x more tailpipe emissions than urban residents, likely due to longer commutes. The total ACT-CF for patients with CHD in the US is equivalent to powering over 12 thousand homes annually. To offset this carbon, over 75,000 acres of forest would be needed.³ As a result, possible measures to offset these carbon emissions must be envisioned. One strategy involves improving preventive care to reduce specialty healthcare visits. Additionally, communities with high levels of social vulnerability suffer from the paradox of poor access and yet needing higher acuity care. Recent advancements in tele-medicine with tele-visits and home monitoring devices (e.g. blood pressure), and even remote electrocardiography and ultrasonography, may provide an avenue to improve healthcare access while reducing environmental footprint of healthcare. While telemedicine usage may increase electricity consumption, this carbon footprint is much lower than that of transportation-related emissions.²

Typically, individuals in affluent areas with higher socioeconomic status emit more CO₂, due to higher consumption.⁴ However, our results demonstrate that patients in vulnerable and rural areas, given the longer distances to cardiovascular care, inevitably produce greater vehicle emissions owing to reduced geographic proximity to cardiologists. Additionally, rural socioeconomically disadvantaged populations may have higher CHD prevalence and morbidity compared to their urban counterparts, potentially leading to more frequent follow-up visits.⁵

In conclusion, ambulatory care of CHD in the US is a considerable source of traffic-related greenhouse gas emissions particularly in rural areas with higher social vulnerability. Exploring responsible mitigation strategies to reduce these emissions, while ensuring adequate healthcare, must be a priority.

Glossary of Terms

CT

Census Tracts

US

United States

Km

Kilometers

IQR

Interquartile range

MTCO2

Carbon dioxide metric tons

ACT-CF

Coronary Heart Disease ambulatory care transport-related Carbon Footprint

GHG

Greenhouse gas

SVI

Social Vulnerability Index