The economic burden of systemic lupus erythematosus in Mexico

Abstract

Aims

Our cost of illness study aimed to provide an estimate of the burden related to systemic lupus erythematosus (SLE) in the Mexican context.

Methods

Our model was used to simulate the resource utilization and economic consequences over a period of 5 years for patients with SLE in Mexico. The model simulated four health states—three phenotypes of SLE, including mild, moderate, and severe states, and death. Clinical parameters were retrieved from the literature. Resource utilization in our model represents the most common practice in the Mexican healthcare system. These include disease management, transient events (e.g. infections, flares, and complications due to SLE-related organ damage), and indirect costs. Direct non-medical costs were not considered. One-way sensitivity analysis was performed.

Results

The number of targeted Mexican SLE patients was 57,754. The numbers of SLE patients diagnosed with mild, moderate, and severe phenotypes were 8,230, 44,291, and 5,233, respectively. Disease management costs, including the treatment of each phenotype and disease follow-up, were MXN 4 billion ($ 415 million); the costs of transient events (infections, flares, and consequences of SLE-related organ damage) were MXN 5 billion ($ 478 million). Productivity loss costs among adult employed Mexican patients with SLE were estimated at MXN 17 billion ($ 1.6 billion). The total SLE cost in Mexico over 5 years from the payer and societal perspectives is estimated at MXN 9 billion ($ 893 million) and 26 billion ($ 2.5 billion), respectively. Over 5 years, the costs per patient per year from the payer and societal perspectives were MXN 32,131($ 3,095) and MXN 91,661($ 8,830), respectively.

Conclusion

The findings pointed out the substantial economic burden associated with SLE, including the costs of disease progression and SLE transient events, such as flare-ups, infections, and organ damage, in addition to productivity loss due to work capacity impairment.

Keywords:

• SLE
• costs
• Mexico
• economic burden
• complications

JEL CLASSIFICATION CODES:

• C51
• C5
• C
• G28
• G2
• G
• H51
• H5
• H

View correction statement:

Correction

Introduction

Systemic lupus erythematosus (SLE) is a chronic systemic autoimmune disease characterized by complex and heterogeneous disease severity ranging from mild cutaneous disease to catastrophic organ failure¹. SLE imposes a substantial burden on patients and healthcare systems². In Mexico, the healthcare expenditure represents 5.5% of gross domestic product. Health care in Mexico is delivered via public insurance (population coverage around 72%) and private insurance. There was an increase in out‐of‐pocket (41% of health expenditure) and catastrophic health expenses between 2018 and 2022. Mexico needs to improve access to health services in the public sector³.

Epidemiological data will continue to expand our understanding of SLE. Thus, the SLE-related burden continues to evolve. Globally, the incidence of SLE is estimated as 5.14 per 100,000 person-years, and the annual number of newly diagnosed patients is estimated to be 0.4 million patients⁴. Notably, this disease is more prevalent in women⁵. Epidemiological studies of SLE in Mexico are lacking. However, Peláez-Ballestas et al. reported that the adjusted point prevalence of SLE in Mexico was 0.06%⁶.

In Mexico, the overall mortality risk for SLE patients is higher than that for the general population⁷. A study utilizing a nationwide health registry reported recent updates on all-cause mortality among patients in Mexico⁷. The study pointed out that the age-specific mortality rate (ASMR) grew from 0.49 per 100,000 persons in 1998 to 0.89/100,000 persons in 2017, representing a relative increase of 81.6%⁷. During the same period, ASMR decreased by 8.6% in the general population⁷.

SLE is a highly complex disease with challenging diagnosis/treatment². Therefore, it appears to affect patients’ quality of life (QoL) owing to concerns about adverse events (AEs) or the negative impact of the disease on the work/economic status of the patient⁸. A cross-sectional study indicated that Mexican patients with active SLE had the worst health-related quality of life (HRQoL)⁹.

Patients with SLE frequently experience flares and persistent symptoms. It was estimated that only 6.5% of patients with SLE experienced remission for at least 1 year¹⁰. Notably, flares are associated with significant financial/clinical burden¹¹. A retrospective cohort study utilizing claims data from the United States (US) reported that the cost per flare increased with the degree of severity¹². The mean cost was $ 1,606 for mild flare and $ 2,587 and $ 14,829 for moderate and severe flares, respectively¹².

Patients with SLE are at high risk of developing infections¹³. The relative risk of infection among SLE-patients shows a 2- to 6-time increase than that among the general population¹³. Furthermore, infections have been reported to contribute to an increased mortality risk in patients with SLE¹⁴. A systematic literature review of studies containing adult/pediatric patients with SLE showed that infections were the main cause of death among SLE patients¹⁵. Approximately 15.1% of deaths in high-income countries and 37.5% of deaths in low- or middle-income countries are due to infections¹⁵. Conversely, cardiovascular diseases (CVDs) account for 11.3% of deaths in high-income countries and 10.6% in low- or middle-income countries¹⁵.

Owing to its natural course, SLE affects multiple organs². In particular, it has a significant effect on the renal system. Lupus nephritis (LN) is one of the most severe clinical manifestations of SLE. Notably, 40–70% of patients develop LN¹⁶. Moreover, LN remains a major cause of morbidity and mortality despite the availability of SLE treatments¹⁷. Therefore, LN has been reported to contribute to an increasing SLE-related burden¹⁸. The mean annual direct cost of SLE patients with LN was 7.5 folds higher per patient than that of SLE patients without LN¹⁸.

Several studies have highlighted the substantial burden of SLE. Using real-world data from Canada, the average cost estimate for SLE patients per year was two-fold higher than that for non-SLE patients¹⁹. A systematic review of 51 articles concluded that the mean emergency department visits were 0.3–3.5 per year, and mean hospitalizations were 0.1–2.4 per year (mean length of stay 0.4–13.0 days). Physician visits were 10–26 visits/year. Mean annual direct total costs were $ 17,258–$ 63,022 per patient and were greater for patients with moderate or severe SLE ($ 19,099–$ 82,391) compared with mild SLE ($ 12,242–$ 29,233).²⁰. Between 2009 and 2014 in Germany, the average annual costs of mild, moderate, and severe SLE were from € 1,890–3,010, € 4,867–5,876, and € 8,396–10,001, respectively²¹.

Despite advances in SLE treatment, patients experience poor outcomes²². For example, the risk of disease-related mortality remains high⁷. Moreover, patients with SLE suffer from several unmet needs, including persistent disease activity and frequent flares, dependency and toxicity of glucocorticoids (GC), high disease comorbidities, poor HRQoL, limited access to high-quality care, and long-term survival¹⁷.

Overall, recent cost of illness (COI) studies continue to enrich our perception of the substantial SLE-related burden on patients and healthcare systems, provide the required knowledge to understand disease outcomes, and inform policymakers about the necessity of new biologics that can help allocate resources. Our model sought to provide an estimate of the considerable burden of SLE in the Mexican context, utilizing direct and indirect costs over a period of 5 years among SLE patients.

Methodology

The COI model was used to inform stakeholders about the substantial burden of SLE after diagnosis in Mexico. We used a Markov model, an analytical framework that is frequently used in decision analysis, and use disease states to represent all possible consequences of SLE patient’s cohort starting from year 2023. During disease diagnosis, patients were categorized as having mild, moderate, and severe SLE, considering the severity of clinical manifestations according to British Isles Lupus Assessment Group (BILAG) and physician interpretation²³. Our state-transition model estimated the cost of SLE according to disease progression. The model structure is considered the most standardized local clinical practice in Mexico and was validated by the local Delphi panel.

The prevalence-based approach, the most commonly used method in COI, was used to quantify the monetary effects of SLE in our model. The prevalence of SLE among Mexican populations (0.06%) was retrieved from a cross-sectional study conducted in Mexico to describe predictive variables linked with rheumatic diseases and evaluate the prevalence of musculoskeletal (MSK) disorders⁶. The simulated population in our model comprised Mexican individuals aged ≥ 15 years and diagnosed with SLE. The target population in our model was retrieved from the most recent published population data by the World Bank in 2022 (population aged 15–64 and ≥ 65)^24,25.

Our model simulated four health states: i) the three phenotypes of SLE, including mild, moderate, and severe SLE, and ii) death. Moreover, our model captured transient events throughout the course of the disease, such as infections, flares, and complications due to SLE-related organ damage. Owing to a lack of data, the percentages of patients at different SLE phenotypes (mild and moderate states) after diagnosis were extracted from a study that used data of the SLE population (21,993 individuals) between 2014 and 2017 out of 1,375,863 medical records from the Colombian contributive health system²⁶. At any point in time after diagnosis, patients may transit and enter a health status of higher disease severity (for instance, from mild to moderate/severe SLE or from moderate to severe SLE). The progression into a health status of higher severity was retrieved from a study that assessed 462 patients with SLE over a median follow-up period of 36 months to simulate the transition in disease severity²⁷. Figure 1 shows the structure of the model used in the COI study.

Figure 1. The model structure for SLE patients. Abbreviation. SLE, Systemic lupus erythematosus.

In our model, the main cause of mortality is disease specific based on our Delphi panel. The transition probabilities from different SLE phenotypes/states (alive health state) to an absorbing health state (death) were extracted from a case-control study performed on 72 Chinese SLE patients (15 had hematological malignancies, while 57 had no hematological malignancies) between January 2013 and December 2020²⁸. We extrapolated the overall survival (OS) Kaplan-Meier’s (KM) curves of 57 patients with SLE without hematological malignancies to estimate lifetime patient-level data based on parametric models²⁸. An exponential parametric function provided a plausible fit of the OS curves²⁸.

This manuscript was written according to the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) statement²⁹. The inputs were retrieved from a literature review of all the clinical trials and validated by a local Delphi panel. This Delphi panel composed of five experts; three rheumatologists affiliated to Centro Médico Nacional and Hospital General de México, and two health economists. We collected insights from experts through three rounds virtual meetings by using the quasi-Delphi panel approach. The Delphi consensus includes the most standardized treatment strategies for the Mexican healthcare system.

The transition to a health status with a higher degree of severity was assumed to occur in a four-week cycle period. The probability to go from mild to moderate in 36 months is 29.1% while it is 20.7% to directly go from mild to severe based on Nikolopulos et al.²⁷. We transferred this probability to 4 weeks transition probability by using this equation: tp1 = 1-(1-tpT)^(1/t) where tp1 is 4 weeks transition probability and tpT is overall probability over time t and then extrapolate it to 5 years. According to the International Society for Health Economics and Outcomes Research model of good research practices, a half-cycle correction was applied to adjust the cost distribution through the model³⁰.

We simulated the payer and societal perspectives, which included the costs of disease management and transient events; the societal perspective calculated productivity costs as well. Costs were discounted at 5% in the base case, as recommended by other economic evaluations conducted in Mexico³¹.

Clinical inputs

Our COI model simulated Mexican patients with SLE at diagnosis over a period of 5 years. The simulated population comprised Mexican SLE patients aged ≥ 15 years with an average weight of 70 kg. The Mexican population aged ≥ 15 years (96,256,568) was retrieved from the most recently published population data by the World Bank in 2022 (population aged 15–64 and ≥ 65)^24,25. It was multiplied by the adjusted point prevalence of SLE in Mexico⁶. Therefore, the target population for the COI model included 57,754 patients.

Our model simulates different SLE phenotypes. Upon diagnosis, the patients were categorized into one of three SLE phenotypes: mild, moderate, or severe. Table 1 lists all input parameters of the model.

Table 1. The model input parameters.

(A) Eligible patients:	Base case	Low value	High value	References
Number of populations ≥ 15 years of age	96,256,568	77,005,254	115,507,882	[24,25]
Prevalence of SLE in Mexico (%)	0.06%	0.05%	0.07%	6
Mild SLE (%)	14%	11%	17%	26
Moderate SLE (%)	77%	61%	92%
(B) Baseline patient characteristics:
Weight (kg)	70	56	84	Delphi panel
Cardiovascular manifestations (%)	55%	44%	66%	32
Cutaneous manifestations (%)	87%	69%	100%
Neuropsychiatric manifestations (%)	37%	30%	44%
Renal manifestations (%)	44%	35%	53%	33
Renal patients progressed into ESRD (5 years) (%)	14%	11%	16%	34
Respiratory manifestations (%)	5%	4%	6%	35
ESRD managed by renal transplantation (%)	25%	20%	30%	Assumption
ESRD managed by dialysis (%)	75%	60%	90%
(C) Clinical parameters: (1) Transition to severe state:
Mild to moderate SLE (%) on 4 weeks	0.009	0.007	0.010	27
Mild to severe SLE (%) on 4 weeks	0.006	0.005	0.007
Moderate to severe SLE (%) on 4 weeks	0.009	0.007	0.011
(2) OS
OS: exponential λ	−0.0036407	−0.0029	−0.0044	31
(3) Risk of flares
SLE patients experience flares annually (%)	95%	85%	100%	36
Mean annual mild flare rate per patient	1.1	0.9	1.3
Mean annual moderate flare rate per patient	2.2	1.8	2.6
Mean annual severe flare rate per patient	0.2	0.2	0.2
(4) Risk of infection
Risk of serious infection	14.5%	12%	17%	37
(D) Discount rate:
Discount rate	5%	4%	6%	30
(E) Indirect cost parameters:
Labor force among population ≥ 15 years (%)	62%	50%	70%	38
Work productivity impairment among mild-SLE (%)	15%	10%	20%	39
Work productivity impairment among moderate-SLE (%)	38%	30%	50%
Work productivity impairment among severe-SLE (%)	50%	40%	60%
Mean monthly sick days per patient	2.3	1.8	2.8	40
Mexican GDP per capita (MXN)	143,956	115,165	172,747	41

Abbreviations. SLE, Systemic lupus erythematosus; ESRD, end stage renal disease; OS, overall survival; GDP, gross domestic product; MXN, Mexican peso.

According to a Delphi panel and literature review, most Mexican patients with SLE develop mucocutaneous manifestations, CVD, LN, neuropsychiatric lupus (NPSLE), and respiratory complications (pulmonary hemorrhage; PH/diffuse alveolar hemorrhage; DAH). The proportion of SLE patients who experience mucocutaneous manifestation (86.5%), CVD (55.4%), and neurologic manifestations (37%) was obtained from a study that included 1,480 Latin American SLE patients. The study aimed to assess the characteristics of late-onset SLE patients within the GLADEL (Grupo Latino Americano de Estudio del Lupus) cohort of SLE-patients (mean of SLE-patients’ data aged <50 and ≥ 50)³². Further, a multicenter, retrospective study demonstrated the differences in SLE severity and its clinical manifestations between two groups—3,305 SLE European Caucasian patients and 185 SLE Latin American mestizos patients. The study pointed out that 44% of Latin American mestizo patients developed LN³³. Among patients with LN, the percentage of patients whose disease progressed to end-stage renal disease (ESRD) within 5 years (average of 16% and 11%) was obtained from a systematic review and Bayesian meta-analysis that included data from 18,309 patients from 187 articles to demonstrate the risk of ESRD³⁴. We assumed that 75% of the patients with ESRD were managed with dialysis; the remaining patients were managed with renal transplantation. SLE-associated PH was found in 5.4% of a cohort of 630 SLE patients in a study that aimed to assess PH among Mexican patients with SLE³⁵.

Our model simulates different types of flares in patients with SLE. Between 2005 and 2014, a retrospective cohort study that appraised 2,227 SLE-patients in the US demonstrated that the annual percentage of patients with SLE developing flares was 95.1%. The mean annual mild, moderate, and severe flares per patient were 1.1, 2.2, and 0.2, respectively³⁶. A time-to-event analysis that assessed 346 patients with SLE over a mean period of 6.6 years pointed out that the risk of serious infection among patients with SLE was 14.5%³⁷.

Costs

The COI model considered the following costs: disease management costs (i.e. treatment strategies for each SLE phenotype and required disease follow-up); transient event costs (i.e. management of SLE-related organ damage, infections, and flares); indirect costs (productivity loss costs among living and dead employed Mexican adults with SLE). Additionally, transient event costs were calculated in accordance with the average resources used and the length of hospital stay, if required, to manage a single event.

To calculate total healthcare costs, resource use was multiplied by the unit costs of each component. All healthcare resource utilization (HCRU) was captured from the most standardized local practices within the Mexican healthcare system, as informed by the Delphi panel. All unit costs measured in Mexican peso (MXN) were obtained from the Mexican hospital system based on the financial year 2023. Local currency conversions to international dollar were performed using the purchasing power parity rate, sourced from World Bank. Table 2 shows all the unit costs and treatment doses used in our model.

Table 2. The unit costs and treatment doses used in the model.

Cost parameters
(1) Disease management	Base case	Low value	High value	Treatment doses and regimens	References
Chloroquine 150 mg tablet	MXN 0.4	0.5	0.6	300 mg/day	Mexican healthcare system
Methotrexate 10 mg tablet	MXN 2	1	3	A single dose not to exceed 20 mg/week orally
Folic acid 1 mg tablet	MXN 0.01	0.01	0.01	1 mg oral everyday with methotrexate with methotrexate to decrease the risk for side effects
Cyclosporine 100 mg capsule	MXN 9	7	10	2.5 mg/kg/day (175 mg daily)
Cyclosporine 25 mg capsule	MXN 0.3	0.2	0.3
Azathioprine 50 mg oral tablet	MXN 6	5	7	2 mg/kg/day (140 mg daily)
Azathioprine 100 mg oral tablet	MXN 12	10	14
MMF 500 mg oral tablet	MXN 12	10	14	2 gm daily
Rituximab 500 mg vial	MXN 7,340	5,872	8,808	1000 mg infusion and repeat after 2 week (2 infusions separated by 2 weeks is 1 course) and to be repeated every 24 weeks
Cyclophosphamide 500 mg vial	MXN 331	265	397	500 mg IV every 2 weeks for 6 doses (3 mos)
QFT	MXN 433	346	520	Should be performed before the initiation of biologic treatment
Vaccination against seasonal influenza	MXN 5,200	4,160	6,240	Once annually
Vaccination against pneumococcal infection	MXN 184	147	221	Once every five years
Outpatient visit	MXN 2,082	1,666	2,498	When required based on severity level
(2) Disease follow-up	Base case	Low value	High value	Treatment doses and regimens
KFT cost	MXN 101	81	121	When required based on severity level
LFT cost	MXN 101	81	121
CBC cost	MXN 101	81	121
C3/C4 cost	MXN 1,100	880	1320
ESR cost	MXN 101	81	121
CRP cost	MXN 101	81	121
Physician visit cost	MXN 1,692	1354	2030
Anti-dsDNA antibodies	MXN 1,230	984	1476
(3) Flares management	Base case	Low value	High value	Treatment doses and regimens
Prednisone 5 mg tab.	MXN 0.2	0	0	0.2–0.4 mg/kg (Mild flares) 0.4–0.7 mg/kg (Moderate flares) 0.8 mg/kg (Severe flares)
Prednisone 10 mg tab.	MXN 0.5	0	1
Prednisone 20 mg tab.	MXN 2	2	2
Methylprednisolone 1 g iv.	MXN 161	129	194	1 g daily
Ward day	MXN 11,919	9535	14303	7 days
Densitometry	MXN 29	23	35	Once annually
HbA1c	MXN 1,758	1407	2110	Once every three months
(4) Infection management	Base case	Low value	High value	Treatment doses and regimens
Levofloxacin 5 00 mg tablets	MXN 2	1	2	500 mg orally once daily for 14 days
(5) Organ complications	Base case	Low value	High value	Treatment doses and regimens
Topical retinoids	MXN 9	8	11	When required
Renal biopsy	MXN 38,830	31,064	46,596
Hemodialysis session	MXN 3,925	3,140	4,710
Renal transplantation	MXN 62,705	50,164	75,246
Enalapril 10 mg tab.	MXN 0.08	0.1	0.1	10 mg once daily
Atorvastatin 10 mg tab.	MXN 4	4	5	10 mg once daily
Gabapentin 300 mg tab.	MXN 1	1	1	300 mg orally three times daily
Plasma exchange	MXN 3,925	3,140	4,710	When required
ECG	MXN 230	184	276
ECHO	MXN 571	457	685

Abbreviations. MMF, Mycophenolate mofetil; QFT, QuantiFERON; HbA1c, hemoglobin A1C; ECG, electrocardiogram; ECHO, echocardiogram; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; C4, complement component 4; C3, complement component 3; anti-dsDNA, anti-double stranded DNA; LFT, liver function test; KFT, kidney function test; CBC, complete blood count.

The indirect costs in our COI model represented productivity losses among living and dead Mexican patients with SLE. The average labor force within the Mexican population aged ≥ 15 years (62%) was published by the World Bank in 2022³⁸. The percentage of mild, moderate, and severe SLE patients with work impairment was extracted from a study that appraised a large cohort of SLE patients (N = 689 SLE patients) to determine the SLE burden on work productivity and employment in the US³⁹. According to a study that aimed to evaluate the work disability and productivity loss among 344 SLE patients and 322 matched controls, the mean monthly sick days per patient was 2.3 days⁴⁰. The average wage/hour for Mexican people was estimated using the most recently published gross domestic product (GDP) published by the World Bank for 2022⁴¹.

Treatment regimens and follow up

Within the Mexican population, the cornerstone for managing SLE, despite its disease activity, is an antimalarial (chloroquine). For SLE patients with moderate and severe phenotypes, immunosuppressants were added to ensure more case controls. In accordance with the Delphi panel, the selection of immunosuppressants was dependent on the disease manifestation/affected organs. Therefore, we used the mean costs of the most commonly used immunosuppressants (methotrexate, azathioprine, mycophenolate mofetil, MMF, cyclosporine, and cyclophosphamide) in our model. Patients with severe SLE require further biological treatments (rituximab). Belimumab is not used among the standard clinical practice in the public sector in Mexico based on our Delphi panel.

Patients with SLE are regularly followed up. Owing to the administration of chloroquine, the patients were assessed annually by an ophthalmologist. SLE monitoring was conducted for all patients, regardless of disease severity or administered medications. The follow-up tests were erythrocyte sedimentation rate (ESR), anti-double-stranded deoxyribonucleic acid antibodies (anti-dsDNA), C-reactive protein (CRP), complement component 3 (C3), complement component 4 (C4), complete blood count (CBC), kidney function test (KFT), and liver function test (LFT) performed once every 3–6 months.

According to our Delphi panel, the most common type of infection among Mexican SLE patients was pneumonia, which required management with oral antibiotics (levofloxacin) without the need for hospitalization.

In the case of flares, SLE patients were managed according to the severity using prednisone, which was tapered subsequently to the least allowed daily dose to avoid the development of GC-related AEs^42,43. The initial prednisone dose was continued for two weeks and then tapered once every two weeks (5 mg decrements)⁴⁴. Patients with SLE and severe flares required hospitalization for one week. Owing to the administration of GC, patients were assessed by densitometry once annually and hemoglobin A1C (HbA1c) once every three months.

Owing to the natural complexity of the disease, the time interval between symptom onset and accurate disease diagnosis is long (mean, 47 months)⁴⁵. Therefore, most SLE patients develop signs of SLE-related organ complications at diagnosis. Consequently, SLE-related organ damage was used in our model from the first cycle. According to the Delphi panel, SLE-related mild cutaneous manifestations required the administration of topical retinoids (assuming one tube for each cycle) with a dermatologist follow-up once every six months. In terms of CVD, patients required regular assessment (once every six months) by a cardiologist using electrocardiography, ECG, echocardiography, and ECHO with the administration of atorvastatin. Owing to the presence of LN, patients were initially assessed by renal biopsy. Notably, patients with LN required follow-up by a nephrologist every four months. For LN patients without ESRD, an angiotensin-converting enzyme inhibitor (ACEI; enalapril) was administered. Patients with NPSLE require anticonvulsants (gabapentin). For SLE-related DAH, plasma exchange was included in the treatment plan, along with the administration of immunosuppressants and GCs. It was assumed that patients performed only four cycles of plasma exchange, started in the first cycle of our model^46,47.

Sensitivity analysis

A one-way sensitivity analysis was performed to ensure robustness of the results. The various parameters were varied by a plausible range, 10–20% above or below their base-case values as performed in many health economic studies. The parameters tested were epidemiological data, clinical parameters, and unit cost data for all treatments.

Results

The number of targeted Mexican SLE patients in our COI was 57,754. The numbers of SLE patients diagnosed with mild, moderate, and severe phenotypes were 8,230, 44,291, and 5,233, respectively. Over a period of 5 years, the disease management costs, including the treatment of each phenotype and disease follow-up, were MXN 4 billion ($ 415 million); the costs of transient events (infections, flares, and consequences of SLE-related organ damage) were MXN 5 billion ($ 478 million). Productivity loss costs among adult employed Mexican patients with SLE were estimated at MXN 17 billion ($ 1.6 billion). The total SLE cost in Mexico over 5 years from the payer and societal perspectives is estimated at MXN 9 billion ($ 893 million) and 26 billion ($ 2.5 billion), respectively (Table 3). The results demonstrated the huge costs associated with productivity loss resulting from SLE-related morbidity and mortality (Figure 2). The total disease management costs were MXN 164 million ($15.8 million) for mild SLE and MXN 1.88 billion ($182 million) and MXN 1.80 billion ($173 million) for moderate and severe SLE, respectively. Over a period of 5 years, the costs per patient per year from the payer and societal perspectives were MXN 32,131 ($ 3,095) and MXN 91,661 ($ 8,830), respectively.

Figure 2. The cost components of SLE burden from societal perspective. Abbreviation. SLE, Systemic lupus erythematosus.

Table 3. The model results of SLE cost over 5 years.

The payer perspective Total disease cost (MXN)	MXN 9,278,598,119 ($ 893,891,919)
Cost components	Disease management cost	Transient events cost
	MXN 4,315,017,810 ($ 415,704,991)	MXN 4,963,580,308 ($ 478,186,928)
The societal perspective Total disease cost (MXN)	MXN 26,468,975,837 ($ 2,549,997,672)
Cost components	Disease management cost	Transient events cost	Indirect cost
	MXN 4,315,017,810 ($ 415,704,991)	MXN 4,963,580,308 ($ 478,186,928)	MXN 17,190,377,719 ($ 1,656,105,753)

Abbreviations. MXN, Mexican Peso, the local currency was converted to international dollar using the 2022 purchasing power parity conversion factor (10.38), sourced from the World Bank.

Extrapolating the time horizon of the analysis to10 years did not alter the conclusions reached on the basis of other analyses.

Sensitivity analysis

One-way sensitivity analysis was performed to test the uncertainty of the results. Tornado diagrams showed that the number of adult patients with SLE in Mexico was the most sensitive parameter from the payer’s perspective, and the number of employed persons in our cohort was the most sensitive parameter from a societal perspective (Figures 3 and 4).

Figure 3. One-way sensitivity analyses for SLE cost from payer perspective over 5 years. The light green colour indicates the low value of the results, while the dark blue colour indicates the high value of the results. Abbreviations. OS, overall survival; ESRD, end stage renal disease; SLE, systemic lupus erythematosus.

Figure 4. One-way sensitivity analyses for SLE cost from societal perspective over 5 years. The red colour indicates the high value of the results, while the light green colour indicates the low value of the results. Abbreviations. ESRD, end stage renal disease; SLE, systemic lupus erythematosus.

Discussion

This study was the first in Mexico to report the HCRU to demonstrate the financial burden of SLE on the healthcare system and patients. The findings of our model showed the high costs of SLE, as it affects patients during their prime working years, leading to high indirect and direct costs. Therefore, our study highlights the need for novel treatment strategies to reduce the massive burden on patients with SLE in Mexico.

Patients with SLE have a high risk of organ damage and greater utilization of resources originating from long-term disease activity. In the US, a systematic review that included 51 studies indicated that SLE is a significant cost driver.²⁰ The findings demonstrated that the total direct costs associated with SLE were $ 17,258–$ 63,022 per patient per year.²⁰ However, SLE patients with higher disease activity incurred higher costs than patients with mild SLE did.²⁰ For instance, the cost for patients with moderate/severe SLE was $ 19,099–$ 82,391, whereas that for patients with mild disease was $ 12,242–$ 29,233.²⁰

Despite advances in SLE treatment, SLE patients remain prone to experience one or more disease flares/exacerbations, which negatively impacts long-term outcomes and contributes to increasing the burden associated with SLE. SLE patients with flares incur higher costs—both direct and indirect—than patients without flares do⁴⁸. According to a retrospective COI study conducted on 306 Chinese patients with SLE, the total average costs for patients with flares were two times higher than those for patients without flares: $ 22,580 and $ 10,870 per patient-year, respectively⁴⁸. Additionally, Bell et al. demonstrated the flare-associated burden in 11,663 patients with SLE in the US¹². Their findings showed that the mean cost per flare increased with the degree of flare severity¹². For instance, the mean cost per moderate and severe flare was $ 2,587 and $ 14,829, respectively, and that per mild flare was $ 1,606¹².

SLE has a substantial effect on the QoL of patients and caregivers. For instance, employed caregivers reported missing 12.8% of their paid work time because of caregiver responsibility⁴⁹. Moreover, half of the caregivers (49.4%) reported that caregiving responsibilities affected their ability to socialize with friends⁴⁹. Almost all caregivers (97.6%) reported an increase in anxiety and stress due to their role⁴⁹. Multiple studies have identified the burden of SLE in patients^50,51. By assessing 1,029 SLE patients with SLE in Sweden, Jönsen et al. reported that the average annual sick leave was 123–148 days per patient⁵⁰. Therefore, 70% of the total SLE costs are indirect, while the remaining (30%) were direct⁵⁰. Furthermore, Macejova et al. have demonstrated that SLE has a significant negative influence on many aspects of patients’ lives⁵¹. For instance, 56.9% of patients with SLE reported that the impact of SLE on life was generally restrictive, while 23.1% reported a considerably restrictive impact⁵¹.

Owing to the nature of the disease, SLE-related organ damage is a major complication that has a significant negative impact on patients’ HRQoL and healthcare system costs. For instance, 77% of patients with SLE experience accrual organ damage⁵². Moreover, the development of SLE-related organ damage increases mortality⁵². The findings of a systematic review and meta-analysis by Murimi-Worstell et al. pointed out that each 1-unit increase in the damage index score led to mortality increase (pooled hazard ratio [HR] = 1.34; 95% confidence interval [CI] = 1.24:1.44)⁵³. Additionally, Barber et al. highlighted the economic burden and substantial costs associated with organ damage in SLE⁵⁴. Notably, the study concluded that SLE patients with more organ damage had higher healthcare cost⁵⁴. For instance, the annual cost of SLE patients with high organ damage index scores (SDI score ≥5) was 12-fold higher than that of patients with the lowest organ damage index scores (SDI score = 0)⁵⁴. The 10-years cumulative healthcare costs of SLE patients with the highest organ damage index score were 9-fold higher than those of patients with the least organ damage⁵⁴.

Our prevalence-based COI study aimed to inform policymakers in Mexico about the economic impact on patients with SLE after diagnosis. Our study has several strengths and limitations. The strengths are that most of our clinical parameters in the model were extracted from studies that represented the Latin American population and were validated by a Delphi panel. Additionally, resource utilization in our model represents the most common local practice within the Mexican healthcare system. Nevertheless, this study has some limitations. There is a lack of local data in the literature (e.g. the percentage of SLE patients who experienced work impairment and mean monthly sick days per patient). Instead, we captured data from other studies, and all these parameters were validated using our Delphi panel. Another limitation, we relied on retrospective study conducted on 72 SLE patients to capture the probability of death from different SLE phenotypes as there was no other reliable studies to inform this parameter, but we validated it from our Delphi panel. Furthermore, direct non-medical costs and informal costs were not considered due to the paucity of local data.

Conclusion

Our model provides reliable insights into SLE costs in Mexico. The findings pointed out the substantial economic burden associated with SLE, including the costs of disease progression and SLE transient events, such as flare-ups, infections, and organ damage, in addition to productivity loss due to work capacity impairment. Our model highlights the need for novel treatments to reduce the economic consequences of SLE in Mexico.

Transparency

Author contributions

GE, LO, MP, GR involved in conception and design, LO, MP, GR involved in analysis and interpretation of the data; GE draft the paper, GE, LO, MP, GR revised it critically for intellectual content; and the final approval of the version to be published; and that all authors agree to be accountable for all aspects of the work.

Reviewer disclosures

Peer reviewers on this manuscript have received an honorarium from JME for their review work but have no other relevant financial relationships to disclose.

Supplement statement

This article is part of a supplement sponsored by AstraZeneca. All articles within this supplement have been rigorously peer reviewed by experts in the field, as per Journal of Medical Economic’s peer review policy. Any conflicts of interest are stated in the “Declaration of financial/other relationships” section.

Acknowledgements

The authors gratefully acknowledge Mariam Elattar for the writing assistance utilized in the production of this manuscript and Neveen Kandil for her efforts in organizing the Delphi panel.

Declaration of funding

This study was funded by AstraZeneca, they were not involved in the study design, analysis, interpretation of results or manuscript writing.

The funding received was used to pay for the submission and the open access publication fees.

Declaration of financial/other relationships

GE was employed by HTA Office, LLC. GE is a speaker for Janssen, Merck, Novartis, AstraZeneca, Roche, Eva pharma and Pfizer. The authors have no other financial relationships to disclose.

The experts did not receive any compensation for their participation in the Delphi panel.

Correction Statement

This article was originally published with errors, which have now been corrected in the online version. Please see Correction (http://dx.doi.org/10.1080/13696998.2024.2341210)