The carbon footprint of emergency care for low back pain in Sydney, Australia, 2016–2021

Technical Summary: The Carbon Footprint of Emergency Care for Low Back Pain in Sydney, Australia, 2016–2021

Problem Statement
Low back pain (LBP) represents a significant burden on global health systems, accounting for approximately 4.4% of all emergency department (ED) presentations worldwide and ranking among the top reasons for ED attendance in Australia. Despite evidence indicating that only a small fraction (estimated at 2.1%) of these presentations involve serious spinal pathology requiring urgent intervention, the majority of patients receive care within the ED. Current clinical guidelines recommend conservative management outside the hospital for non-specific LBP, yet rates of "low-value" care—such as routine diagnostic imaging and opioid prescription—remain high. While the healthcare sector contributes significantly to global greenhouse gas emissions (estimated at 4.4% globally), the specific carbon footprint associated with the high-volume pathway of ED care for LBP has not been quantified. This study addresses the gap in understanding the environmental impact of this common clinical pathway, specifically examining the emissions generated by patient travel, diagnostic imaging, pathology testing, and medication.

Methodology
The authors conducted a retrospective environmental footprint analysis using routinely collected Electronic Health Record (EHR) data from three large hospitals in New South Wales, Australia (Royal Prince Alfred Hospital, Concord Repatriation General Hospital, and Canterbury Hospital) covering the period from 2016 to 2021.

• Cohort: The study included 15,736 adult presentations (aged ≥18) for LBP, identified via SNOMED CT–AU discharge diagnostic codes. Patients were categorized into three clinical subgroups: non-specific LBP, LBP with neurological signs/symptoms, and LBP originating from serious spinal pathology (e.g., fracture, malignancy).
• Carbon Calculation: The study employed a process-based Life Cycle Assessment (LCA) approach to estimate carbon emissions (kg CO₂e) for four key components:
  1. Patient Transport: Emissions were calculated based on mode of arrival (private car, ambulance, public transport, walking) and distance traveled. Distances were estimated using the centroid of the patient's residential postcode and the ED location. Emission factors were sourced from the ecoinvent inventory, assuming a medium-sized petrol car for private transport.
  2. Diagnostic Imaging: Emissions for MRI, CT, and X-ray were applied using standardized LCA values (17.5, 9.2, and 0.8 kg CO₂e per test, respectively).
  3. Pathology Testing: Emissions for blood tests (e.g., Full Blood Count, Coagulation profile, CRP) were derived from established LCA values.
  4. Medication: Emissions for prescribed analgesics (opioids, NSAIDs, paracetamol) were estimated using data from the YewMaker carbon classifier, with proxy assumptions applied for medications lacking specific LCA data.
• Analysis: Outcomes were reported as mean per-patient emissions and mean annual totals. Statistical differences between discharge destinations (home vs. admitted) and LBP categories were assessed using t-tests and one-way ANOVA.

Key Results

• Overall Footprint: The mean carbon footprint per LBP ED presentation was 5.8 kg CO₂e (SD 15.1). The mean annual total emissions across the three hospitals were 15,174.1 kg CO₂e (SD 1,485.6). This annual total is equivalent to driving a medium-sized petrol car approximately 45,000 kilometers, or just over the Earth's circumference.
• Emission Drivers:
  • Patient Travel: Accounted for the largest proportion of emissions (57%, 3.3 kg CO₂e per patient).
  • Imaging: Contributed 39% (2.3 kg CO₂e per patient).
  • Pathology and Medication: Contributed 3% and <1% respectively.
• Discharge Destination: Patients admitted to hospital (short-stay or inpatient) had a significantly higher per-patient carbon footprint (10.4 kg CO₂e) compared to those discharged home (4.8 kg CO₂e). This difference was primarily driven by higher imaging utilization in admitted patients.
• Clinical Subgroups: Patients with serious spinal pathology had the highest per-patient emissions. However, due to the high volume of presentations, non-specific LBP accounted for the largest share of total annual emissions (12,474.5 kg CO₂e).
• Imaging Specifics: Among discharged patients, imaging for non-specific LBP generated an estimated 2,572.3 kg CO₂e annually, a volume the authors note is equivalent to driving one-fifth of the Earth's circumference.

Significance and Claims
The paper claims to be the first study to quantify the carbon footprint of ED care specifically for low back pain. The authors assert that their findings demonstrate that emergency presentations for LBP contribute substantially to healthcare-related carbon emissions, with the majority driven by patient travel and diagnostic imaging.

The study posits that a significant portion of these emissions is attributable to potentially avoidable care, including unnecessary ED presentations for non-specific LBP, low-value imaging, and hospital admissions. The authors suggest that aligning care with evidence-based clinical standards—specifically by reducing unnecessary investigations, promoting community-based management, and utilizing digital models of care (such as the "Back@Home" virtual hospital)—could yield meaningful carbon savings without compromising patient outcomes. They conclude that integrating environmental considerations into clinical decision-making, alongside cost and patient outcomes, is a necessary step toward developing more sustainable healthcare systems.

Limitations Noted by Authors
The authors acknowledge that their estimates represent a partial carbon footprint, as they excluded downstream interventions (e.g., surgery), follow-up care, and facility energy consumption. They also note potential measurement errors arising from the use of postcode centroids for distance calculation and the assumption of 0% electric vehicle usage for private transport. Furthermore, the study was limited to metropolitan hospitals, meaning emissions for rural and remote patients may be underestimated.