In-center hemodialysis (HD) is the dominant kidney replacement therapy modality in many middle-income settings, including Türkiye, where 71% of patients on kidney replacement therapy receive in-center HD. HD is resource intensive: each 4-hour session requires sizable volumes of treated water, continuous electricity, and generates substantial single-use medical waste. As dialysis demand rises with aging and noncommunicable disease prevalence, quantifying and reducing these direct resource demands has become important for environmental sustainability and health-system stewardship.
This study presents a national, scenario-based assessment of point-of-care resource consumption for maintenance in-center HD in Türkiye. It uses published per-session parameters scaled to the 2023 national session count (9,273,992 sessions) and empirical calibration from measured utility and waste data collected at two outpatient HD centers in Ankara. The analysis deliberately focuses on direct, controllable inputs—water treatment at the dialysis unit, electricity used during treatment, and waste handling at point of care—rather than on upstream life-cycle stages such as equipment manufacture or transport.
This modelling study combined literature-derived per-session parameters with measured facility-level foreground data. National scaling used the 2023 Turkish Nephrology Registry session total. Upstream processes and manufacturing were excluded; only point-of-care resource inputs were modelled. Because only aggregate publicly available data and facility billing records were used, ethical approval was not required.
Three electricity scenarios captured variation in facility configuration: a low-demand scenario (6.2 kWh/session) representing efficient on-demand RO units, a moderate scenario (12.0 kWh/session) reflecting central RO with intermittent operation, and a high-demand scenario (19.6 kWh/session) representing continuously running older systems. Water baseline used 0.493 m³ (493 L) per 4‑hour session at 500 mL/min dialysate flow and ~42% RO recovery, while higher RO recovery rates (70–90%) were discussed as improvement potential. Waste mass was modelled using an 8.0 kg/session upper bound, with two handling scenarios: good segregation (1.5 kg hazardous + 6.5 kg municipal) and poor segregation (all 8.0 kg treated as hazardous).
Greenhouse gas (GHG) emissions were calculated using a Türkiye-specific electricity emission factor of 0.494 kg CO2e per kWh and international conversion factors for waste incineration and other processes where applicable.
To ground the literature-based model in local practice, resource consumption data were collected from two private outpatient HD centers in Ankara for one billing cycle (February–March 2026). Center A performed about 1,500 sessions per month with dedicated on-demand RO units; Center B about 2,500 sessions per month with a central water system.
Billing records provided electricity consumption from commercial invoices, water consumption from utility bills, and hazardous waste quantities from licensed medical waste disposal invoices. Combined measured values from these centers (approximately 4,000 sessions/month) were compared with literature parameters and used to generate an empirically calibrated national estimate alongside the literature-based scenarios.
Water: Baseline literature parameter was 493 L/session (including product water, priming, rinsing, and RO reject), indicative of ~42% RO recovery. The study retained 493 L as baseline but noted that modern high-efficiency RO configurations can achieve 70–75% recovery (reducing consumption to ~260–285 L/session) and specialist systems can approach 90% recovery (~228 L/session).
Electricity: Three scenarios were modelled—low-demand (6.2 kWh/session), moderate (12.0 kWh/session), and high-demand (19.6 kWh/session)—reflecting differences in RO configuration, centralized versus per-machine systems, and equipment age/operation patterns.
Waste: A literature-based upper bound of 8.0 kg/session was used, with two handling scenarios: good segregation (1.5 kg hazardous material incinerated; rest municipal) and poor segregation (all 8.0 kg treated and incinerated as hazardous).
Measured Turkish foreground data provided alternative per-session values: water ~390 L/session, electricity mean ~10.8 kWh/session, and hazardous waste ~1.19 kg/session under good segregation practices.
Scaling literature parameters to 9.27 million sessions in 2023 produced annual national estimates: approximately 4.57 million cubic meters of water consumption; 57.5–181.8 GWh electricity across the low-to-high scenarios; and 13.9–74.2 kilotonnes of solid waste depending on segregation assumptions. Total greenhouse gas emissions across scenarios were estimated to range from 45.3 to 171.1 kilotonnes CO2e per year when using the Türkiye-specific electricity emission factor.
When calibrated to measured Turkish facility data, per-session water consumption was lower (~390 L vs. 493 L, a 21% relative reduction). Electricity measured at roughly 10.8 kWh/session aligned with the literature’s moderate scenario. Measured hazardous waste per session under good segregation (1.19 kg) was lower than the 1.5 kg assumed in the conservative literature good-segregation scenario.
Across scenarios, electricity constituted the dominant share of GHG emissions (approximately 63–72%), with waste incineration contribution increasing markedly when segregation was poor (up to 47% in some modelled situations).
A one-at-a-time sensitivity analysis varied electricity consumption, water volume (incorporating the 390–493 L measured range), and waste segregation practice. Electricity consumption had the largest influence on per-session GHG emissions given the national electricity emission factor and the range modelled. Waste handling practices meaningfully affected emissions when a larger fraction of total waste was treated as hazardous and incinerated.
The analysis identifies actionable, high-impact interventions for dialysis units and policymakers. Key levers include improved waste segregation to reduce the mass of material sent for incineration, adoption of higher-efficiency RO systems or reuse of RO reject water for nonclinical purposes, and strategies to decarbonize electricity supply (on-site renewables, procurement from lower-emission sources, or grid decarbonization).
Measured Turkish facility data demonstrate that local practice can yield lower water consumption than some international benchmarks, supporting the value of locally measured parameters when estimating environmental footprints. The study emphasizes point-of-care measures that dialysis providers can control immediately rather than broader life-cycle interventions.
This work is not a full life-cycle assessment and excludes upstream processes such as equipment manufacture, packaging, and transport. National estimates were linearly scaled from per-session parameters and from measured data in two private outpatient centers in Ankara; results may not capture heterogeneity across all facility types or regions. Details on some modelling choices and conversion factors are reported in the study; where data were not available in the source, those specifics were not reported.
In-center HD in Türkiye imposes a substantial but modifiable environmental burden. Empirical data from two Turkish centers indicate lower per-session water use and hazardous waste generation under good segregation than conservative literature estimates, while electricity use remains a principal driver of greenhouse gas emissions. Practical mitigation priorities are improved waste segregation, higher-efficiency water treatment and reuse, and electricity decarbonization—measures that dialysis units and health systems can implement to reduce the sector’s environmental impact without altering clinical care.