Non–technique-related healthcare utilization and costs among dialysis and kidney transplant recipients in Spain: a real-world analysis

Abstract

Background

Kidney failure requiring kidney replacement therapy (KRT) is associated with a substantial and growing economic burden. While the costs directly attributable to KRT modalities are well documented, the resource use and costs unrelated to KRT modalities remain unclear.

Objective

To quantify healthcare resource utilization (HCRU) and costs in patients receiving dialysis or kidney transplantation, excluding technique-related costs.

Methods

This retrospective observational study used BIG-PAC real-world data to analyze dialysis or kidney transplant patients as of January 1, 2018, with a two-year follow-up. Patients were identified based on related diagnostic or procedure codes recorded in electronic health records (EHRs) prior to the index date. Major adverse cardiovascular events (MACE), hospitalization for CKD and mortality were evaluated. HCRU, direct healthcare costs, and indirect productivity costs due to sick leave were assessed, excluding technique-related KRT costs.

Results

Of 70,973 individuals, 356 (0.5%) were on dialysis and 232 (0.3%) had kidney transplantation. Over two years, MACE per 1000 patient-years were 276.3 for dialysis and 85.9 for transplantation. CKD hospitalization rates were 245.7 and 183.0, with mortality rates of 253.8 and 35.6, respectively. First-year health costs averaged €25,410.4 (SD €17,369.0) for dialysis patients and €4,472.8 (SD €2,859.1) for transplant recipients. Most costs among dialysis and kidney transplant patients were driven by hospitalization, 72.4% and 39.8%, respectively. Indirect costs due to sick leave were also substantial.

Conclusion

Non–technique-related costs represent a substantial component of the overall economic burden in patients receiving KRT, largely driven by hospitalizations. These findings provide important context for healthcare planning, although they do not represent total costs of care and should be interpreted as descriptive estimates.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12882-026-05223-2.

Introduction

Chronic kidney disease (CKD) affects 10–15% of Spain’s population, with recent data suggesting this could be as high as 30% based on electronic health records (EHRs) [1, 2]. Prevalence is expected to rise due to an ageing population and increasing rates of hypertension and diabetes [2].

It has been estimated that in Spain, among CKD patients, kidney replacement therapy (KRT) accounts for roughly 40% of related healthcare costs [1]. Various studies in Spain have examined the expenses related to KRT, but these figures mostly reflect only the procedural costs, such as sessions, supplies, and access, rather than the patient’s overall expenses during follow-up care. In fact, variability in costs is largely due to variations in staff and consumable expenses. However, in most cases, additional non-technical costs, including hospital stays, primary or specialist visits, medication, and sick leave, are not thoroughly measured [3–7].

This issue is significant, as budget impact models are increasingly used to guide strategic decisions on the organization of KRT, drawing on technique-specific direct costs. Nevertheless, these models may require accurate inputs reflecting the total cost of care, which should include additional expenditures not directly associated with treatment techniques [3, 7, 8]. An understanding of these non–technique-related costs of care is important to provide an accurate assessment of the true economic burden of patients on KRT. This knowledge is important to help healthcare decision-making, optimize resource use, and guide policy decisions within healthcare systems.

In previous studies we showed that, in real-world settings, declining kidney function and increasing albuminuria are associated with higher risks of cardiovascular and renal complications, along with increased costs and healthcare resource utilization (HCRU) [9, 10]. The primary objective of this study was to quantify non–technique-related healthcare resource utilization and associated direct and indirect costs in patients receiving dialysis or living with a kidney transplant in routine clinical practice in Spain. In addition, clinical outcomes were described as contextual variables to better characterize the overall burden of this population. This approach addresses an important knowledge gap, as most previous economic evaluations have focused primarily on procedure-related costs rather than the broader cost of patient care.

Methods

We conducted a retrospective observational study using real-world data from the BIG-PAC database, which includes fully anonymized records from 1.8 million patients at primary care centers and referral hospitals across seven Autonomous Communities in Spain. The database is representative of the Spanish population [9–11]. The study was approved by the Investigation Ethics Committee of Consorci Sanitari from Terrassa (Barcelona, Spain). The committee approved the exemption of informed consent, as this was a secondary data study and data were fully anonymized and dissociated from patients. Furthermore, consent for publication was not required as the data utilized were anonymized secondary sources. The study was carried out in accordance with the ethical principles of the Declaration of Helsinki and the regulations in force.

This analysis is part of a broader study spanning all KDIGO stages of CKD [12], with results for G1 A1 through G5 non-dialysis A3 previously published [2, 9]. In the overall cohort, adult patients were included if they were aged 18 years or older and had at least one measurement of both estimated glomerular filtration rate (eGFR) and urinary albumin-to-creatinine ratio (UACR) performed by local laboratories within six months of 1 January 2018 (index date). In contrast, patients receiving dialysis and kidney transplant recipients were identified exclusively based on related diagnostic and/or procedure codes recorded in the electronic health records (EHRs) at the index date, regardless of the availability of eGFR or UACR measurements. Therefore, the present study focuses on patients receiving dialysis or living with a kidney transplant identified by codes in EHRs at the index date, regardless of laboratory availability. This approach reflects real-world identification strategies commonly used in large healthcare databases but may introduce some degree of misclassification. All participants were followed for a period of two years to analyze the occurrence of cardiovascular outcomes, hospitalization for CKD, HCRU and costs.

Baseline data were collected at the index date and included demographic information, existing health conditions, and current medications. Demographics covered age, sex, body mass index, and blood pressure. All known comorbidities, such as cardiovascular disease, coronary artery disease, heart failure, stroke, atrial fibrillation, peripheral artery disease, and diabetes, were identified using both ICD-9 and ICD-10 codes from records before the index date [13]. Medication data were obtained from prescription registries based on the Anatomical Therapeutic Chemical Classification System, reflecting standard prescribing practices [14]. Treatments included renin angiotensin system inhibitors, sodium-glucose cotransporter 2 (SGLT2) inhibitors (for people with or without diabetes), beta blockers, diuretics, calcium channel blockers, statins, and diabetes medications, including oral and injectable drugs and also immunosuppressive therapies in the case of kidney transplant recipients. Laboratory results nearest to the index date were used, including eGFR, creatinine, UACR, potassium, HbA1c, lipid profile, and hemoglobin.

Cardiovascular events and hospitalization for CKD were defined as the primary diagnosis recorded during a hospital visit or admission within two years following the index date. Cardiovascular outcomes encompassed myocardial infarction, stroke, heart failure, peripheral artery disease, and major adverse cardiovascular events (MACE), which included any combination of stroke, myocardial infarction, or all-cause mortality. These outcomes were assessed in patients receiving dialysis and kidney transplant recipients.

HCRU, excluding technique-related, was assessed over a 2-year follow-up period and included primary care, specialist and emergency room visits, lab and radiology tests, hospitalizations (number and length), sick leave (frequency and duration), and drug count.

Costs, excluding technique-related expenses, were estimated using 2019 unit costs from the Spanish National Healthcare System, obtained from the eSalud Consulting database (Supplementary Table 1) [15]. The analysis included costs related to hospitalizations, medications, primary care visits, specialist visits, emergency department visits, diagnostic tests, and productivity losses due to sick leave. Cost related to dialysis sessions, transplant surgery, and access procedures were excluded. Therefore, this study assessed only non–technique-related costs and did not provide a comprehensive estimate of total KRT expenditures. Hospitalization costs were calculated as length of stay multiplied by the unit cost per hospital day; diagnosis-related groups were not used [15]. Medication costs were based on recorded prescriptions and dispensing costs, including immunosuppressive therapies when recorded for kidney transplant recipients [16]. Costs related to primary care, specialist and emergency department visits, and diagnostic tests were estimated using the corresponding unit costs. Productivity losses due to sick leave were estimated only among working-age patients, defined as those younger than 65 years, using a human capital approach. Specifically, sickness-related absence days were multiplied by the mean daily salary of a working person in Spain, as reported in national statistics [17]. The friction cost approach was not applied. All unit costs were applied uniformly across study groups to ensure comparability of cost estimates. Costs were calculated annually per patient.

Statistical analysis

This study was designed as a descriptive analysis of real-world data. No hypothesis testing or adjusted analyses were performed. Therefore, differences observed between groups reflect underlying differences in baseline characteristics and should not be interpreted as causal or comparative effects [18]. Categorical variables are shown as counts and proportions, while continuous variables are summarized by mean (or median) with standard deviation. Incidence rates are reported as events (N) and per 1000 patient-years. Follow-up ended at observation completion or death unless an event occurred. Health care costs were analyzed from the index date over a two-year period and are presented as mean costs per patient. SPSS v25.0 (SPSS Inc., Chicago, IL, USA) was used for analysis, and R (v4.0.2; R Foundation for Statistical Computing, Vienna, Austria) generated all visualizations.

Results

At the index date, the total population included 70,973 individuals. Of these, 356 subjects (0.5%) were receiving dialysis, and 232 (0.3%) had a kidney transplant. Results corresponding to the remaining CKD stages within the study population have been reported elsewhere [2, 9].

Baseline demographics and clinical characteristics showed marked differences between the two groups (Table 1). For patients receiving dialysis, the mean (SD) age was 66.75 (14.81) years, with 40.7% female representation, 32.0% diagnosed with type 2 diabetes, 17.1% with coronary artery disease, and 21.3% with heart failure. Among individuals who had undergone kidney transplantation at the index date, the mean (SD) age was 60.08 (13.32) years; 34.9% were female, 26.3% had type 2 diabetes, and both coronary artery disease and heart failure were present in 9.9% of this group.

Table 1.

CKD population baseline characteristics in patients on dialysis or with kidney transplant

	G5 Dialysis (N = 356; 0.5%)	Kidney transplant (N = 232; 0.3%)
Biodemographic data
Age (years), mean (SD)	66.75 (14.8)	60.08 (13.3)
Gender (female), n (%)	145 (40.7)	81 (34.9)
BMI, kg/m2, mean (SD)	25.05 (7.8)	24.02 (7.0)
Comorbidities
CVD, n (%)	174 (48.9)	63 (27.2)
Coronary ischemic disease, n (%)	61 (17.1)	23 (9.9)
Heart failure, n (%)	76 (21.3)	23 (9.9)
Stroke, n (%)	17 (4.8)	3 (1.3)
Peripheral artery disease, n (%)	66 (18.5)	17 (7.3)
Type 2 diabetes, n (%)	114 (32.0)	61 (26.3)
CV/renal drugs
RAASi, n (%)	130 (36.5)	131 (56.5)
ACEi, n (%)	43 (12.1)	25 (10.8)
ARBs, n (%)	87 (24.4)	107 (46.1)
MRAs, n (%)	7 (2.0)	8 (3.4)
ARNI, n (%)	0 (0.0)	0 (0.0)
SGLT2i (in non-T2DM patients), n (%)	0 (0.0)	0 (0.0)
SGLT2i (in T2DM patients), n (%)	8 (2.2)	2 (0.9)

ACEi: angiotensin-converting enzyme inhibitors; ARB: angiotensin II receptor blockers; ARNI: angiotensin receptor-neprilysin inhibitor; BMI: body mass index; CVD: cardiovascular disease; DPP4i: dipeptidyl peptidase-4 inhibitors; eGFR: estimated glomerular filtration rate; GLP-1 RA: glucagonlike peptide-1 receptor agonists; KDIGO: Kidney Disease: Improving Global Outcomes; MRAs: mineralocorticoid receptor antagonists; RAASi: renin–angiotensin–aldosterone system inhibitors; SGLT2 i: sodium-glucose cotransporter-2 inhibitors; SU: sulfonylureas; T2DM: type 2 diabetes mellitus; UACR: urine albumin–creatinine ratio

For cardiovascular treatments, renin–angiotensin system inhibitors were prescribed in 36.5% of those receiving dialysis, and 56.5% of kidney transplant recipients. Corresponding rates of statin use were 46.9%, and 59.1%, respectively. SGLT2 inhibitors were prescribed in 2.2% on dialysis, and 0.9% with kidney transplants among those with diabetes, while their use among patients without diabetes was almost nonexistent (Table 1; supplementary Table 2).

Over two years, combined MACE incidence rates per 1000 patient-years were 276.3 for dialysis, and 85.9 for transplant groups. Heart failure was the most frequent cardiovascular outcome in both groups. Hospitalization rates for CKD were 245.7, and 183.0 per 1000 patient-years, respectively. All-cause death rates were 253.8, and 35.6 per 1000 patient-years (Table 2).

Table 2.

Incidence rates of complications in patients on dialysis or with kidney transplant during 2 years of follow-up

		G5 Dialysis (N = 356; 0.5%)		Kidney transplant (N = 232; 0.3%)
		N	Rates+	N	Rates+
Mortality	All-cause death	145	253.8	16	35.6
CV outcomes	Myocardial infarction	39	71.1	6	13.6
	Stroke	39	70.5	4	9.0
	Heart failure	93	174.8	19	43.45
	PAD	43	78.3	12	27.2
	MACE mod*	129	276.3	36	85.9
Renal outcomes	Hospitalization for CKD	125	245.7	70	183.0

⁺Per 1000 patient-year; ^*MACE mod: stroke, myocardial infarction or all-cause death;. CKD: chronic kidney disease; CV: cardiovascular; eGFR: estimated glomerular filtration rate; KDIGO: Kidney Disease: Improving Global Outcomes; MACE mod: major adverse cardiovascular events modified; MARE: major adverse renal events; NA: not applicable; PAD: peripheral artery disease

Regarding HCRU during the first year of follow-up, 68.3% of those on dialysis and 38.8% of kidney transplant patients were hospitalized. The high hospitalization burden among dialysis patients was largely driven by recurrent admissions in a subgroup of highly complex patients. The average number of primary care visits (mean [SD]) was 11.9 (2.4) for dialysis, and 5.5 (1.2) for the transplant group; specialized care visits averaged 7.5 (1.8), and 4.3 (0.6), respectively. Among subjects younger than 65 years, the proportion with a sick leave was 48.9% among dialysis patients, and 26.3% in the transplant group. The mean (SD) number of prescribed drugs was 10.4 (4.4) for dialysis, and 7.7 (4.2) for transplant recipients. In the second year, HCRU measures were generally lower than in the first year (Table 3, supplementary Table 3, Fig. 1, and supplementary Figs. 1 and 2).

Table 3.

Annual Health resources utilization in patients on dialysis or with kidney transplant

First year
	G5 Dialysis (N = 356; 0.5%)	Kidney transplant (N = 232; 0.3%)
Primary care visits, mean (SD)	11.9 (2.4)	5.5 (1.2)
Specialized care visits, mean (SD)	7.5 (1.8)	4.3 (0.6)
Emergency rooms visits, mean (SD)	7.2 (1.3)	4.1 (0.5)
Laboratory requests, mean (SD)	20.45 (2.3)	9.1 (0.9)
Radiology and other tests, mean (SD)	1.2 (0.6)	0.6 (0.5)
Hospitalization
N (%)   Days, mean (SD)	243 (68.3) 38.3 (31.2)	90 (38.8) 3.7 (5.1)
Sick leave
Patients with sick leave among patients < 65 years, N (%)   Days of sick leave, mean (SD)	69 (48.9) 288.9 (47.9)	35 (26.3) 146.4 (22.8)
Number of drugs, mean (SD)	10.4 (4.4)	7.7 (4.2)

Second year
	G5 Dialysis (N = 293)	Kidney transplant (N = 225)
Primary care visits, mean (SD)	10.2 (3.8)	5.4 (1.3)
Specialized care visits, mean (SD)	6.5 (2.4)	4.2 (0.7)
Emergency rooms visits, mean (SD)	6.3 (2.1)	4.1 (0.6)
Laboratory requests, mean (SD)	17.56 (5.8)	8.9 (1.2)
Radiology and other tests, mean (SD)	0.92(0.6)	0.5 (0.5)
Hospitalization
N (%)   Days, mean (SD)	205 (70.0) 33.3 (30.89)	89 (39.6) 3.7 (5.1)
Sick leave
Patients with sick leave among patients < 65 years, N (%)   Days of sick leave, mean (SD)	63 (44.7) 139.6 (31.1)	35 (26.3) 73.5 (11.3)
Number of drugs, mean (SD)	10.2 (4.6)	7.8 (4.3)

Fig. 1.

Primary, specialized, emergency visits and hospitalizations during the first year of follow-up in patients on dialysis (G5_D) or with kidney transplant (G5_Tx)

During the first year, average total health-related costs (excluding direct technique-related KRT costs) were €25,410.4 (SD €17,369.0) for dialysis patients, and €4,472.8 (SD €2,859.1) for transplant recipients. The main contributors to healthcare costs in patients receiving dialysis were hospitalizations (72.4%) and medications (17.2%). For transplant recipients, hospitalizations made up 39.8% of expenses and medications 29.5%. Indirect costs due to sick leave were €5,667.5 (SD €11,767.8), and €2,235.8 (SD €5,388.7) for each group, respectively. In the second year, health-related, indirect, and overall costs were generally lower than in the first year in both groups (Table 4, supplementary Table 4, Fig. 2, and supplementary Figs. 3–8).

Table 4.

Annual cost in 2018–2019 in patients on dialysis or with kidney transplant, excluding renal replacement therapy technique-related expenses (population with index date 2018)

	G5 Dialysis (N = 356; 0.5%)	Kidney transplant (N = 232; 0.3%)
First year
Cost of primary care visits, mean (SD)	286.8 (56.9)	132.5 (29.4)
Cost of laboratory requests, mean (SD)	661.4 (74.7)	292.2 (27.9)
Cost of radiology and other tests, mean (SD)	115.9 (56.8)	59.0 (48.4)
Cost of specialized visits, mean (SD)	709.6 (169.4)	403.3 (60.7)
Cost of emergency rooms visits, mean (SD)	856.1 (149.4)	487.3 (62.9)
Cost of Hospitalization, mean (SD)	18,398.5 (15,016.3)	1,780.6 (2431.7)
Cost of all medications, mean (SD)	4,382.1 (3430.0)	1,318.0 (1218.0)
Healthcare cost, mean (SD)	25,410.4 (17,369.0)	4,472.8 (2859.1)
Indirect cost/Sick Leave, mean (SD)	5,667.5 (11,767.8)	2,235.8 (5388.7)
Total cost (health care + indirect costs), mean (SD)	31,077.9 (21,343.7)	6,708.6 (6569.3)
Second year
Cost of primary care visits, mean (SD)	246.5 (91.1)	131.5 (30.3)
Cost of laboratory requests, mean (SD)	567.4 (188.6)	287.8 (38.6)
Cost of radiology and other tests, mean (SD)	90.4 (60.9)	47.1 (49.3)
Cost of specialized visits, mean (SD)	617.3 (228.4)	400.7 (67.1)
Cost of emergency rooms visits, mean (SD)	743.8 (250.7)	483.5 (69.3)
Cost of Hospitalization, mean (SD)	15,999.4 (14,837.2)	1,799.6 (2431.6)
Cost of all medications, mean (SD)	4,132.1 (3565.7)	1,314.7 (1226.5)
Healthcare cost, mean (SD)	22,396.9 (18,863.2)	4,465.0 (2867.3)
Indirect cost/Sick Leave, mean (SD)	3,038.7 (5994.0)	1,156.9 (2738.2)
Total cost (health care + indirect costs), mean (SD)	25,435.6 (20,346.1)	5,621.9 (4296.9)

Fig. 2.

Healthcare cost, excluding renal replacement therapy technique-related expenses, and proportional breakdown of healthcare cost components during the first year of follow-up in patients on dialysis (G5_D) or with kidney transplant (G5_Tx)

Discussion

Our findings showed that patients receiving KRT, including dialysis and kidney transplantation, had substantial HCRU and costs beyond those directly related to treatment techniques. In particular, hospitalizations represented the main driver of non–technique-related costs, especially among dialysis patients.

In previous studies we have assessed cardiovascular and renal outcomes, along with HCRU and costs, for Spanish patients ranging from G1 A1 to G5 non-dialysis A3. The findings indicated that as renal function declined, both outcome risks and costs increased [2, 9, 19]. However, these studies did not include individuals on dialysis or those who had received a kidney transplant. In Spain, the Inside CKD Project estimates that by 2027, among diagnosed CKD patients, 3.9% will need KRT, a 15% increase from 2022 [1]. In our study population, of nearly 71,000 subjects, 356 (0.5%) were on dialysis and 232 (0.3%) had received a kidney transplant at the index date. The clinical profile of the study population was consistent with findings from earlier research on both dialysis and kidney transplant groups [20–25], suggesting that our study accurately reflects patients receiving KRT.

With respect to cardiovascular therapies, slightly more than one-third of patients undergoing dialysis and over half of kidney transplant recipients were prescribed renin-angiotensin system inhibitors. Renin-angiotensin system inhibitors play an essential role in slowing CKD progression; however, their application in dialysis patients presents complexities, with primary advantages observed in the management of heart failure and hypertension [26]. In kidney transplant recipients, certain studies indicate potential benefits of renin-angiotensin system inhibitors within this population [27]. In our cohort, the use of SGLT2 inhibitors was marginal. More studies are required to establish the role of SGLT2 inhibitors in individuals undergoing dialysis or in kidney transplant recipients [28–30].

Previous studies have largely examined events, HCRU, and costs directly tied to KRT techniques [3–6, 8]. In contrast, we specifically analyzed non–technique-related factors, providing novel evidence that may inform healthcare policy decisions. In this descriptive analysis, cardiovascular events, CKD-related hospitalization, and mortality were frequent among patients receiving dialysis and, to a lesser extent, among kidney transplant recipients. Dialysis patients have a high burden of vascular disease and are at substantial risk of cardiovascular events, which drive most complications and mortality in this group. This is due to many factors, such as fluid overload, inflammation, mineral bone disorders, anemia, or uremic toxins, as well as the presence of traditional risk factors such as hypertension and diabetes. This underscores the need for better management of these interrelated risks [31–33]. Kidney transplant recipients also showed a considerable burden of cardiovascular events, although observed event rates were numerically lower than those reported among patients receiving dialysis. This ongoing vulnerability is attributed to both persistent traditional factors, such as hypertension, diabetes and dyslipidemia, and transplant-specific issues, including immunosuppressive therapy and CKD progression. Although outcomes following transplantation have significantly improved in recent years, cardiovascular disease persists as one of the leading causes of mortality within this population, underscoring the necessity for a comprehensive management of these conditions [34].

Patients receiving dialysis and kidney transplant recipients showed substantial healthcare resource utilization, including frequent primary care, specialist, and emergency department visits, laboratory testing, hospitalizations, sick leave, and medication use. Hospitalization rates were especially high for dialysis patients (68%) and kidney transplant recipients (39%) during the first year of follow-up. Although hospitalization categories would have been informative, a reliable post hoc classification of all-cause hospitalizations into cardiovascular, infectious, access-related, transplant-related, and other causes was not available for this analysis. Dialysis patients are frequently hospitalized due to comorbidities like heart failure and dialysis-related complications [35, 36]. Transplant recipients also have elevated hospitalization rates because of immunosuppression-derived complications such as infections, graft rejection and existing comorbidities. While initial hospitalizations often relate to the surgery itself, later admissions are driven by infections, graft dysfunction or cardiovascular disease [37]. Importantly, it should be noted that frailty, multimorbidity, and care fragmentation are likely to contribute significantly to high HCRU beyond KRT. Sick leave was also common in patients under 65 who were on dialysis (49%) or had a kidney transplant (26%). Despite the ability of many CKD patients to remain employed, the disease often restricts employment opportunities, emphasizing the importance of flexible work arrangements and legal protections [38].

We assessed direct and indirect healthcare costs during follow-up in patients receiving dialysis and kidney transplant recipients, focusing the analysis beyond procedure- or surgery-related expenses and providing novel insights beyond previous research [3, 8, 39–42]. Our study found that non–technique-related healthcare costs were substantial among patients receiving dialysis and kidney transplant recipients.

Hospitalizations represented the main driver of non–technique-related costs. However, hospitalization categories were not available in sufficient detail to determine the relative contribution of cardiovascular, infectious, access-related, transplant-related, or other causes. Indirect costs due to sick leave were also substantial, although this analysis captured only productivity losses related to sick leave and did not include caregiver costs, premature mortality, presenteeism, or other societal costs. Therefore, these estimates should be interpreted as a partial assessment of the broader societal burden.

Taken together, these findings suggest that non–technique-related HCRU and costs may provide relevant additional context alongside technique-related or postprocedural costs in healthcare planning and future economic evaluations. They are also consistent with the broader rationale for early kidney care and prevention or optimal management of conditions such as diabetes and hypertension, which may help reduce progression to advanced kidney disease and downstream healthcare burden. However, because the present study did not evaluate preventive, organizational, or therapeutic interventions, no conclusions can be drawn regarding which specific strategies may reduce healthcare utilization or costs in this population [43–46].

This study has several limitations. First, its retrospective observational design relies on information recorded in electronic health records, which may be subject to coding inaccuracies, missing data, or underreporting of comorbidities and healthcare utilization. Second, dialysis and kidney transplant patients were identified using diagnostic and procedure codes without laboratory confirmation. This is the standard in large real-world databases, since laboratory measures like eGFR or UACR are not reliable for this purpose. In dialysis patients, serum creatinine, and thus eGFR, is influenced more by dialysis efficiency, muscle mass, and nutrition than by kidney function, making GFR estimates inaccurate and misleading. In transplant recipients, eGFR varies widely with graft function and time since surgery and does not indicate transplant status. Therefore, code-based identification is the most appropriate and methodologically sound approach for these populations. Third, the analysis was purely descriptive and did not include adjustment for relevant confounders such as age, sex, comorbidities, or disease severity. Moreover, some healthcare costs had skewed distributions and were presented as mean costs with standard deviation. Therefore, differences observed between groups reflect underlying baseline differences and should not be interpreted as causal or comparative effects. Fourth, since the BIG-PAC database is a primary care and specialist outpatient database, rather than a nephrology-specific registry, detailed clinical records specific to the dialysis unit (modality, vascular access type) and transplant follow-up data (time since transplantation, graft function, immunosuppressive regimen) were not systematically captured in the database, limiting the interpretation of drivers of healthcare utilization. Finally, the exclusion of technique-related costs prevents estimation of the total economic burden of these patients, and indirect costs were assessed only in individuals under 65 years of age, likely underestimating the full societal impact. Therefore, these findings should be interpreted within the context of a descriptive real-world analysis and are intended to complement, rather than replace existing estimates of total kidney replacement therapy costs.

In summary, patients receiving dialysis or living with a kidney transplant showed high observed rates of cardiovascular and renal complications, mortality, healthcare resource utilization, and non–technique-related costs during follow-up. Hospitalizations were the main driver of non–technique-related healthcare costs, particularly among patients receiving dialysis. These estimates exclude technique-related KRT costs and should not be interpreted as total costs of care or as incremental costs attributable to KRT status. The findings provide descriptive real-world evidence that may inform future health economic evaluations and healthcare planning in Spain.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

CKD

Chronic kidney disease

eGFR

Estimated glomerular filtration rate

EHRs

Electronic health records

HCRU

Healthcare resource utilization

KDIGO

Kidney Disease: Improving Global Outcomes

KRT

Kidney replacement therapy

MACE

Major adverse cardiovascular events

SD

Standard deviation

SGLT2

Sodium-glucose cotransporter 2

UACR

Urinary albumin-to-creatinine ratio

Author contributions

RS and CE wrothe the main manuscript text and prepared the figures. UA, BP, MC, IH, AC, MG and RG have participated in the analysis, supervision and review of the manuscript. All authors reviewed and approved the manuscript.

Funding

This study was sponsored and financed by AstraZeneca.

Data availability

Data supporting this study is available from the BIG-PAC database upon reasonable requests.

Declarations

Ethics approval and consent to participate

The study was approved by the Investigation Ethics Committee of Consorci Sanitari from Terrassa (Barcelona, Spain). The committee approved the exemption of informed consent, as this was a secondary data study and data were fully anonymized and dissociated from patients. The study was carried out in accordance with the ethical principles of the Declaration of Helsinki and the regulations in force.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.