Introduction: Healthcare costs and societal impact of myasthenia gravis (MG), a potentially life-threatening rare, chronic neuromuscular disease, are sparsely studied. We assessed healthcare resource utilization (HCRU) and associated costs among patients with newly diagnosed (ND) and preexisting (PE) MG in Sweden. Methods: This observational, retrospective cohort study used data from four linkable Swedish nationwide population-based registries. Adult MG patients receiving pharmacological treatment for MG and having ≥24-month follow-up during the period January 1, 2010, to December 31, 2017, were included. Results: A total of 1,275 patients were included in the analysis, of which 554 patients were categorized into the ND MG group and 721 into the PE MG group. Mean (±SD) age was 61.3 (±17.4) years, and 52.3% were female. In the first year post-diagnosis, ND patients had significantly higher utilization of acetylcholinesterase inhibitors (96.0% vs. 83.9%), corticosteroids (59.6% vs. 45.8%), thymectomy (12.1% vs. 0.7%), and plasma exchange (3.8% vs. 0.6%); had higher all-cause (70.9% vs. 35.8%) and MG-related (62.5% vs. 18.4%) hospitalization rates with 11 more hospitalization days (all p < 0.01) and an increased risk of hospitalization (odds ratio [95% CI] = 4.4 [3.43, 5.64]) than PE MG. In year 1 post-diagnosis, ND MG patients incurred EUR 7,302 (p < 0.01) higher total all-cause costs than PE MG, of which 84% were estimated to be MG-related and the majority (86%) were related to inpatient care. These results remained significant also after controlling for baseline demographics and comorbidities (p < 0.01). In year 2 post-diagnosis, the all-cause medical costs decreased by ∼55% for ND MG from year 1 and were comparable with PE MG. Conclusion: In this population-based study, MG patients required significantly more healthcare resources in year 1 post-diagnosis than PE MG primarily due to more pharmacological treatments, thymectomies, and associated hospitalizations. These findings highlight the need to better understand potential factors including disease characteristics associated with increased health resource use and costs and need for more efficacious treatments early in the disease course.

Myasthenia gravis (MG) is a rare, chronic autoimmune disorder of the neuromuscular junction, characterized by muscle weakness that worsens with activity and improves with rest [1]. Most patients initially present as ocular MG with diplopia and ptosis. Approximately 85% of these patients will develop generalized MG (gMG), with bulbar symptoms such as dyspnea, dysarthria, and difficulties chewing and swallowing, as well as symptoms from the extremities [2, 3]. Patients experience periods of calmer disease state alternating with exacerbations, which may result in potentially life-threatening myasthenic crisis with respiratory failure [1, 4]. Though the prevalence of MG varies geographically [5], both incidence and prevalence have been increasing worldwide over the past decades [5‒7]. Based on recent studies, the prevalence is estimated to be 12.4–24.8 per 100,000 population worldwide [7‒10], with a higher incidence among people aged 50 years or more [7, 8].

With no curative treatment available, clinical practice aims at symptom management and prevention of new relapses while minimizing treatment-associated side effects [11, 12]. If adequate symptom control is not achieved through symptomatic treatment with anticholinesterase inhibitors (AChEis), addition of corticosteroids (CSs) alone or in combination with other conventional immunosuppressive therapies (IS), such as azathioprine, cyclosporin, and tacrolimus, is considered. International consensus-based guidance recommends plasma exchange (PLEX) and intravenous immunoglobulin as rescue therapies for the treatment of severe gMG or myasthenic crisis. Further, the use of monoclonal antibodies such as rituximab and eculizumab is recommended for patients with treatment refractory gMG [12]. Based on the recently published retrospective analyses, currently available treatments are often inadequate and approximately 10–20% of patients experience residual symptoms and considerable tolerability issues [13‒17]. However, the treatment paradigm is rapidly evolving with the availability of newly approved targeted biologics such as efgartigimod, ravulizumab, and rozanolixizumab [17, 18].

There is limited evidence on healthcare resource utilization (HCRU) and economic burden among patients with MG. A study using the US claims data from the pharmacy point-of-sale system estimated the mean annual direct all-cause costs for newly diagnosed (ND) gMG patient to USD 26,419 (∼EUR 22,425) in 2018; the annual costs increased by an additional USD 18,000 (∼EUR 15,278) for those with MG exacerbation [19]. Studies from the UK and the USA report 2–3.5 times higher hospitalization rates in treatment refractory compared to non-refractory MG patients [13, 20]. However, population-level real-world evidence in MG within the Nordic region is limited. This study aimed to assess treatment use, associated all-cause and MG-related HCRU, and costs among patients with ND versus preexisting (PE) MG in Sweden. A subgroup analysis of employed MG patients who experienced sickness absence and associated indirect costs was also conducted.

Data Sources

Data were collected from four national registries provided by Statistics Sweden and National Board of Health and Welfare in Sweden. Information on diagnoses, procedures, hospitalizations, and outpatient specialist services were collected from the National Patient Register [21]. Inpatient data were available from 1987 and data on outpatient specialist care from 2001. The Swedish version of the International Classification of Diseases Version 10 (ICD-10-SE) was used to define MG diagnosis, and Swedish Classification of Health Interventions codes were used to define the procedures or surgeries (e.g., intubation/ventilation, PLEX, and thymectomy). Prescribed Drug Register (PDR), including information on dispensed items, amount, date of filled prescriptions from the outpatient care, was available from July 2005 [22]. The Anatomical Therapeutic Chemical classification system was used to code dispensed prescriptions. The Cause of Death Register, including date and cause of death, comprises data on all deaths of people registered in Sweden [23]. Socioeconomic characteristics, available on a yearly basis, including education, income, employment status, were obtained from the Longitudinal Integrated Database for Health Insurance and Labor Market Studies [24]. Data from the four national registers were linked through the unique personal identity number, resulting in full coverage of the total Swedish population.

Study Patients

There are no standard identification criteria to select MG patients using secondary database [8, 19, 25]. In our study, patients were required to have ≥2 primary diagnosis for MG (ICD-10-SE: G70.0) filed at least 12 months apart within a 24-month period, with ≥1 primary MG diagnosis recorded by a neurologist between January 1, 2010, and December 31, 2017 (Fig. 1). Date of the first primary MG diagnosis was designated as the index date. Patients were to be aged ≥18 years as of index date and have ≥24-month follow-up period. To increase the specificity of patient identification criteria and ensure patients were treated, patients were also required to have ≥1 record of AChEis, CS, nonsteroidal immunosuppressive treatment (NSIST), immunoglobulin, or PLEX on or after the index date within study period. Patients were further categorized into the ND or PE group, based on whether they had an MG diagnosis before the index date (back to the earliest date of available data, i.e., January 1, 2001).

Fig. 1.

Patient inclusion flowchart. *Diagnosis as per the Swedish version of the International Classification of Diseases Version 10 (ICD-10-SE: G70.0). At least one diagnosis must have been a primary MG diagnosis filed by neurologist specialists in an inpatient or outpatient specialist visit from January 1, 2010, to December 31, 2017. Index date: date of the first primary MG diagnosis between January 1, 2010, and December 31, 2017. To ensure confirmed diagnosis of MG, patients were also required to have ≥1 record of AChEis, NSIST, CSs, rituximab, eculizumab, intravenous immunoglobulin, subcutaneous immunoglobulin, PLEX. AChEis, acetylcholinesterase inhibitors; MG, myasthenia gravis; NSIST, nonsteroidal immunosuppressive therapy.

Fig. 1.

Patient inclusion flowchart. *Diagnosis as per the Swedish version of the International Classification of Diseases Version 10 (ICD-10-SE: G70.0). At least one diagnosis must have been a primary MG diagnosis filed by neurologist specialists in an inpatient or outpatient specialist visit from January 1, 2010, to December 31, 2017. Index date: date of the first primary MG diagnosis between January 1, 2010, and December 31, 2017. To ensure confirmed diagnosis of MG, patients were also required to have ≥1 record of AChEis, NSIST, CSs, rituximab, eculizumab, intravenous immunoglobulin, subcutaneous immunoglobulin, PLEX. AChEis, acetylcholinesterase inhibitors; MG, myasthenia gravis; NSIST, nonsteroidal immunosuppressive therapy.

Close modal

Study Measures

Use of pharmacological treatments and procedures were evaluated. All-cause HCRU, defined as resource use incurred from any medical or pharmacy services, and MG-related HCRU, defined as resource use and costs associated with MG primary diagnosis or MG-related pharmacological treatment, were assessed. All-cause and MG-related direct costs were calculated as the sum of medical costs and prescription costs incurred. Medical costs were estimated by multiplying the diagnosis-related group (DRG) weight with cost per DRG weight for each recorded inpatient admission and outpatient specialist visit. Pharmacy costs were estimated by summing patient-paid costs and the reimbursement paid by the county council for recorded prescription drugs. Among patients with ND MG, the primary outcome measures were compared between the first year (year 1) and the second year (year 2) post-index period.

In a subgroup of patients with valid employment records at index and subsequent 2 years, sickness absence and associated indirect costs were calculated by multiplying days of sickness absence with the average gross daily wage (adding payroll taxes) in the Swedish population. All costs were adjusted to the 2019 Swedish Krona and subsequently converted to Euro using 2019 average conversion rate from Sweden Central Bank.

Statistical Analyses

Descriptive analyses were performed for all study variables. The primary outcomes were compared between the ND and PE groups. The difference in patient characteristics and outcome measures between ND and PE MG groups was compared using χ2 tests or exact Fisher tests (categorical variables), t tests (continuous variables) and Wilcoxon Mann-Whitney tests (non-normally distributed variables). Multivariable logistic regression models were used to assess the association between annual hospitalization rates in ND versus PE MG, whereas generalized linear models with gamma distribution and log link function were used to estimate the incremental annual costs associated with ND versus PE MG after adjusting for baseline age, gender, employment status during the year prior to index year, Charlson Comorbidity Index (CCI) at baseline as well as hypertension, stroke, and hypothyroidism at baseline. A p value of <0.05 was considered statistically significant. SAS 9.4 (SAS Institute, Cary, NC, USA) or R version 4.1.2 was used for data analysis.

Sociodemographic and Clinical Characteristics

A total of 1,275 patients were eligible for inclusion in the study, of which 43.5% (n = 554) were patients with ND MG (Table 1). The mean [SD] age was similar between ND MG and PE MG groups (62.0 [18.2] vs. 60.8 [16.7], p = 0.21); patients in the ND group had a higher mean [SD] CCI score (3.22 [1.48] vs. 2.90 [1.73], p < 0.01). A total of 370 employed patients were included in subgroup analyses (ND: n = 158; PE: n = 212). Mean (SD) age was 48.1 (12.3) years, and mean (SD) CCI score was 2.63 (1.45).

Table 1.

Patient characteristics of ND versus PE MG in Sweden

All patients with MGND MG patientsPE MG patientsp value
Patients, N 1,275 554 721 
Age, years, mean (SD) 61.3 (17.4) 62.0 (18.2) 60.8 (16.7) 0.211 
 Median (Q1–Q3) 65 (49–75) 66 (52–76) 64 (48–74) 
Age >65 years, n (%) 615 (48.2) 289 (52.2) 326 (45.2) 0.016 
Female, n (%) 667 (52.3) 253 (45.7) 414 (57.4) <0.001 
Education, n (%)    0.895 
 0–9 years 374 (29.3) 166 (30.0) 208 (28.8)  
 10–13 (high school) 495 (38.8) 207 (37.4) 288 (39.9)  
 13+ (post-secondary education) 363 (28.5) 154 (27.8) 209 (29.0)  
 Doctoral education 12 (0.9) 5 (0.9) 7 (1.0)  
 Missing 31 (2.4) 22 (4.0) 9 (1.2)  
Employment status, n (%)    0.004 
 In employment 467 (36.6) 211 (38.1) 256 (35.5)  
 Not in employment 242 (19.0) 81 (14.6) 161 (22.3)  
 Not working due to older age 554 (43.5) 251 (45.3) 303 (42.0)  
 Missing 12 (0.9) 11 (2.0) 1 (0.1)  
CCI score*, mean (SD) 3.04 (1.64) 3.22 (1.48) 2.90 (1.73) <0.001 
 Median (Q1–Q3) 3 (3–3) 3 (3–4) 3 (3–3) 
Comorbid conditions, n (%)* 473 (37.1) 208 (37.5) 265 (36.8) 0.224 
 Hypertension 183 (14.4) 98 (17.7) 85 (11.8) 0.004 
 Diabetes 89 (7.0) 39 (7.0) 50 (6.9) 1.000 
 Ischemic heart disease 49 (3.8) 20 (3.6) 29 (4.0) 0.816 
 Hypothyroidism 38 (3.0) 23 (4.2) 15 (2.1) 0.047 
 Hyperlipidemia 34 (2.7) 18 (3.2) 16 (2.2) 0.339 
 Stroke 21 (1.6) 16 (2.9) 5 (0.7) 0.005 
All patients with MGND MG patientsPE MG patientsp value
Patients, N 1,275 554 721 
Age, years, mean (SD) 61.3 (17.4) 62.0 (18.2) 60.8 (16.7) 0.211 
 Median (Q1–Q3) 65 (49–75) 66 (52–76) 64 (48–74) 
Age >65 years, n (%) 615 (48.2) 289 (52.2) 326 (45.2) 0.016 
Female, n (%) 667 (52.3) 253 (45.7) 414 (57.4) <0.001 
Education, n (%)    0.895 
 0–9 years 374 (29.3) 166 (30.0) 208 (28.8)  
 10–13 (high school) 495 (38.8) 207 (37.4) 288 (39.9)  
 13+ (post-secondary education) 363 (28.5) 154 (27.8) 209 (29.0)  
 Doctoral education 12 (0.9) 5 (0.9) 7 (1.0)  
 Missing 31 (2.4) 22 (4.0) 9 (1.2)  
Employment status, n (%)    0.004 
 In employment 467 (36.6) 211 (38.1) 256 (35.5)  
 Not in employment 242 (19.0) 81 (14.6) 161 (22.3)  
 Not working due to older age 554 (43.5) 251 (45.3) 303 (42.0)  
 Missing 12 (0.9) 11 (2.0) 1 (0.1)  
CCI score*, mean (SD) 3.04 (1.64) 3.22 (1.48) 2.90 (1.73) <0.001 
 Median (Q1–Q3) 3 (3–3) 3 (3–4) 3 (3–3) 
Comorbid conditions, n (%)* 473 (37.1) 208 (37.5) 265 (36.8) 0.224 
 Hypertension 183 (14.4) 98 (17.7) 85 (11.8) 0.004 
 Diabetes 89 (7.0) 39 (7.0) 50 (6.9) 1.000 
 Ischemic heart disease 49 (3.8) 20 (3.6) 29 (4.0) 0.816 
 Hypothyroidism 38 (3.0) 23 (4.2) 15 (2.1) 0.047 
 Hyperlipidemia 34 (2.7) 18 (3.2) 16 (2.2) 0.339 
 Stroke 21 (1.6) 16 (2.9) 5 (0.7) 0.005 

CCI, Charlson Comorbidity Index; MG, myasthenia gravis; ND, newly diagnosed MG; PE, preexisting MG; Q, quartile; SD, standard deviation.

*During 12-month pre-index period.

Treatment Utilization after Initial MG Diagnosis

In year 1 post-index, ND MG patients had significantly higher utilization of AChEis (96% vs. 83.9%, p < 0.01), CSs (59.6% vs. 45.8%, p < 0.01), PLEX (3.8% vs. 0.6%, p < 0.01), and thymectomy (12.1% vs. 0.7%, p < 0.01) (Fig. 2). In both groups, treatment utilization reduced in year 2 post-index but remained slightly higher in ND MG patients as compared to the PE MG patients. Most frequently used treatments were pyridostigmine (77.1%), azathioprine (39.0%), and prednisolone (34.5%) (online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000538640).

Fig. 2.

MG-related treatment utilization for ND versus PE MG patients. AChE, acetylcholinesterase; MG, myasthenia gravis; ND, newly diagnosed; PE, preexisting; NSIST, nonsteroidal immunosuppressive therapy.

Fig. 2.

MG-related treatment utilization for ND versus PE MG patients. AChE, acetylcholinesterase; MG, myasthenia gravis; ND, newly diagnosed; PE, preexisting; NSIST, nonsteroidal immunosuppressive therapy.

Close modal

Healthcare Resource Utilization and Costs

The ND MG group had significantly higher all-cause (70.9% vs. 35.8%, p < 0.01) and MG-related (62.5% vs. 18.4%, p < 0.01) hospitalization rates than the PE MG group in year 1 (Fig. 3), with 11 additional hospitalization days (p < 0.01), resulting in an adjusted OR = 4.4, 95% CI (3.43, 5.64) for the risk of hospitalization versus PE MG. Further, all-cause and MG-related hospitalization for ND MG group decreased in year 2 (Fig. 3).

Fig. 3.

Hospitalization for ND versus PE MG patients. MG, myasthenia gravis; ND, newly diagnosed MG; PE, preexisting MG.

Fig. 3.

Hospitalization for ND versus PE MG patients. MG, myasthenia gravis; ND, newly diagnosed MG; PE, preexisting MG.

Close modal

During year 1, the all-cause costs were EUR 7,302 higher in the ND MG group compared to the PE MG group (p < 0.01) with an incremental difference of EUR 6,275 derived from inpatient costs. ND MG group incurred EUR 6,188 higher total MG-related costs (p < 0.01), of which 84% was attributable to inpatient costs (Fig. 4). Age-stratified analyses showed that, among patients aged >65 years (n = 615), the all-cause costs were EUR 14,496 and EUR 8,159 for the ND MG and PE MG group during year 1 (p < 0.01); while the all-cause costs for patients aged 18–65 years were EUR 15,122 and EUR 6,695, respectively, for the ND MG and PE MG group (p < 0.01). After adjusting age at index date, gender, employment status during the year prior to index year, Charlson Comorbidity Index (CCI), and comorbidities (hypertension, stroke, and hypothyroidism) between the ND and PE MG groups, the all-cause costs in the ND MG group estimated as EUR 7,438 higher than in the PE MG group (p < 0.01), and the difference in MG-related costs was estimated as EUR 6,373 (p < 0.01). Further, patients with ND MG incurred EUR 7,294 (p < 0.01) higher all-cause costs and EUR 6,629 (p < 0.01) higher MG-related costs in year 1 than year 2. Comparing the ND and PE group in year 2, no significant differences in either all-cause or MG-related costs were observed (Fig. 4).

Fig. 4.

Healthcare costs for ND versus PE MG patients. Medical cost: inpatient cost + outpatient cost. MG, myasthenia gravis; ND, newly diagnosed MG; PE, preexisting MG; Rx, prescription drug.

Fig. 4.

Healthcare costs for ND versus PE MG patients. Medical cost: inpatient cost + outpatient cost. MG, myasthenia gravis; ND, newly diagnosed MG; PE, preexisting MG; Rx, prescription drug.

Close modal

Sickness Absence and Associated Indirect Costs among Employed MG Patients

During year 1 post-index period, ND MG group had a higher all-cause inpatient admission rate (64.6% vs. 34.9%, p < 0.01), with 6.5 more hospitalization days (p < 0.01) than PE MG. A higher percentage of patients with ND MG reported ≥1 sick leave episode (p < 0.01), with average 31 more days of sickness absence (53.2% vs. 32.5%, p < 0.01). The mean total annual costs for ND MG patients in year 1 were EUR 19,900, with indirect costs of EUR 7,499. In the PE MG group, the mean total annual costs in year 1 were EUR 9,951, with indirect costs of EUR 3,986 (Fig. 5). The mean total annual costs incurred by ND MG patients in year 2 remained high (EUR 11,422), of which 50% was attributable to indirect costs (EUR 5,757).

Fig. 5.

Direct and indirect costs for ND versus PE MG among employed MG patients (n = 370). MG, myasthenia gravis; ND, newly diagnosed; PE, preexisting; Rx, prescription drug.

Fig. 5.

Direct and indirect costs for ND versus PE MG among employed MG patients (n = 370). MG, myasthenia gravis; ND, newly diagnosed; PE, preexisting; Rx, prescription drug.

Close modal

MG is a rare, chronic autoimmune disease of the neuromuscular junction with a highly variable clinical course depending on the age of onset and gender [7, 8, 10, 26, 27]. However, the majority of patients experience the most severe level of disease activity during the first years of disease [28‒30]. The long duration to maximal therapeutic effect of most of the immunomodulatory treatments is thought to affect burden of disease during the first years after diagnosis. Moreover, the stepwise approach to clinical treatment in international guidelines may further amplify high disease and economic burden during the initial years of the disease. The burden of MG has significant negative impact on the social and emotional well-being of the patients [31, 32].

Using four Swedish linkable population-level registers, our study aimed to explore differences in the demographic and clinical characteristics of patients with ND versus PE MG in Sweden. In our study, significantly higher proportion of patients with ND MG were men, aged 65 years or older, and with higher mean CCI score than patients with PE MG. Corroborating our findings, studies have reported an increase in the incidence of late-onset MG, with a higher incidence observed among men over the past 2 decades [33‒37]. Reports of diagnostic delays in older patients due to overlapping symptoms such as muscle weakness, fatigue, and other comorbidities associated with aging that mimic symptoms of MG have also been published [33, 35, 38‒40].

With adjustment of covariates including age at index date, gender, employment status during the year prior to index year, Charlson Comorbidity Index (CCI) and comorbidities (hypertension, stroke, and hypothyroidism), results from multivariable analyses showed that treatment use, HCRU, and associated costs were significantly higher for ND MG compared with PE MG. Specifically, in the first year after the initial MG diagnosis, all-cause and MG-related costs and HCRU were higher for patients with ND MG and decreasing in the subsequent year. One of the strengths of the Swedish registers is the long-term nationwide coverage with universal access to medical services, allowing for reliable and accurate variables and providing population-based estimations without the need for extrapolation.

The observed use of treatments in our study is consistent with previous studies, with AChEis, CSs, and NSISTs as the most frequently used treatments. Nearly two-thirds of ND MG patients received CSs in year 1, and usage remained high in year 2 (48%). As chronic CS use is associated with complications including hypertension, osteoporosis, cataract, and gastrointestinal conditions [41], alternative treatment options with sustained efficacy and superior safety and tolerability profiles are warranted.

Consistent with previous studies conducted in the USA [19, 42, 43], the present study showed that ND patients displayed high HCRU and associated costs. In a US provider-based open claims study, Phillips et al. [19] observed all-cause and MG-related HCRU within the 12-month post-index was higher for patients with ND MG compared with previously diagnosed patients. However, as there are no true enrollment start and end dates with open claims, patient eligibility was not available, and data capture may underestimate actual services used. Utilizing the MGFA Patient Registry, Harris et al. [42] found that the predicted probability of emergency room visits and hospitalization decreased over time among gMG patients. In our study, HCRU for patients with ND MG significantly decreased from year 1 to year 2, yielding significantly decreased costs, primarily attributable to lower inpatient costs. Our findings suggest that more treatments and surgeries are performed in the first year after diagnosis, presumably reflecting that the disease often is more active early in the disease course. Early diagnosis and effective treatments may be important factors in reducing disease burden and optimizing MG management.

The impact of MG on employment status and absence from work has been indicated previously [44]. Frost et al. [45] used Danish national registers and found that, compared to the general population, patients with MG are 8.6 times more likely to take long-term leaves from work in the initial 2 years of disease; in addition, female and those treated with both AChEis and immunosuppressant were more likely to be unemployed. In our study, nearly 40% of the total all-cause costs among working MG patients were accounted by indirect costs, especially in the first year after diagnosis. These findings are consistent with a systematic review which reported high estimated indirect cost of up to USD 3,550 (∼EUR 3,013) associated with MG [46].

As with other retrospective studies based on administrative healthcare data, classification of patients was based on administrative diagnosis and treatment codes; however, patient misclassification is possible because diagnoses were not confirmed. In an earlier Swedish nationwide registry-based study, MG was defined as individuals receiving specialist or inpatient care with a primary or secondary diagnosis of MG [8]. Our study required patients to have ≥2 primary diagnosis for MG filed at least 12 months apart within a 24-month period, with ≥1 primary MG diagnosis recorded by a neurologist. These stricter patient identification criteria ensured that only those patients undergoing active MG treatment are included in the analysis. As primary healthcare data are still reported on a national level, there is a lack of information on diagnoses filed by primary care physicians, potentially limiting the understanding of the full clinical burden. Only treatments recorded in the PDR were used to identify the study cohorts because drug information for intravenous treatment (such as intravenous immunoglobulin and biologics) is not completely recorded in the Swedish National Registers; thus, number of patients with such treatments are underestimated. Our study used data from patients treated in hospitals or outpatient specialist visits in Sweden; therefore, the generalizability of the results outside of the Swedish population is limited, and future studies are warranted to confirm these results. Moreover, the administrative databases used for this study did not capture disease onset date and date of confirmed diagnoses, which are the confounding factors that may affect the study outcomes and results interpretation. In our study, there is a significant difference in HCRU between ND and PE groups in year 1 and no significant difference observed at year 2, which may suggest time from diagnosis as potential driver for the observed differences. This exploratory finding may be investigated in future studies accounting for additional potential inherent clinical and demographic differences, such as age of onset, that were unavailable in this dataset. As the disease is more likely to be severe in patients with late-onset MG and demographic characteristics differ markedly between early-onset MG and late-onset MG [10, 26], further studies by linking nationwide registries with clinical MG registry are warranted to better understand the incremental HCRU associated with the confirmed ND MG. Furthermore, results for some of the clinical measures need to be interpreted cautiously due to the small sample size. For example, stroke seems to be more common in ND than PE patients and likely to impact the HCRU; however, the number of patients with stroke at baseline in our study was very small (n = 21). Future studies with larger sample size are warranted to evaluate the impact of clinically relevant comorbidities on the HCRU.

Findings from our study support the current evidence of severe disease course after initial diagnosis [28]. Ting et al. [47] used US claims database to evaluate the factors affecting HCRU in patients with MG and found a decrease in HCRU in year 2 compared with year 1 among patients using second-line treatments for MG. This suggests that more efficacious treatments early in the disease course may result in improvements in the long-term disease control and reduced HCRU.

Results of this nationwide, population-based study demonstrated that MG patients incurred a substantially higher economic burden in the first year after diagnosis compared with later years. We report a higher rate of hospitalization and absenteeism, as well as higher associated costs, among patients with ND MG in Sweden. Our findings highlight a considerable disease burden at initial diagnosis and a need for efficacious therapies early in the disease course for optimal treatment. Future studies may investigate potential factors including disease characteristics associated with increased health resource use and costs at initial diagnosis.

We thank Uma Kundu, MPharm, CMPP™ (SIRO Clinpharm Pvt. Ltd., India) for medical writing assistance and Robert Achenbach (Janssen Global Services, LLC) for additional editorial support.

The study has been granted an exemption from requiring written informed consent, and the ethical approval for the data access was granted by Etikprövningsmyndigheten (EPN: Swedish Ethical Review Authority) 2018/2032-31 in October 2018.

Qian Cai, Alberto E. Batista, Qiaoyi Zhang, Peter Kunovszki, Kavita Gandhi, and Kristin Heerlein are employees of Janssen Pharmaceuticals and may hold stock or stock options in Johnson & Johnson. Jakob Börsum was employed by SDS Life Science AB, Uppsala, Sweden. Gabriel Isheden owns 100% of the shares of Intelligent Decision Analytics AB and was contracted with SDS Life Science AB to carry out work for this manuscript. Susanna Brauner received grants from UCB Pharma outside the submitted work.

This work was supported by Janssen Global Services, LLC.

Qian Cai: study concept and methodology, data interpretation, manuscript development, review and revision, and approval of the final manuscript. Alberto E. Batista: study concept and methodology, data interpretation, manuscript review and revision, and approval of the final manuscript. Jakob Börsum and Gabriel Isheden: data analysis, data interpretation, manuscript review and revision, and approval of the final manuscript. Qiaoyi Zhang, Peter Kunovszki, and Susanna Brauner: study concept and methodology, data interpretation, manuscript review and revision, and approval of the final manuscript. Kavita Gandhi: study funding, review and interpretation of results, review and revisions of draft manuscript version, and approval of the final manuscript. Kristin Heerlein: study concept and methodology, data interpretation, manuscript review and revision, and approval of the final manuscript.

The data used in this research are held by Statistics Sweden and National Board of Health and Welfare in Sweden. Data access can be granted by those agencies after approval by Etikprövningsmyndigheten (EPN: Swedish Ethical Review Authority).

1.
Gilhus
NE
,
Tzartos
S
,
Evoli
A
,
Palace
J
,
Burns
TM
,
Verschuuren
JJGM
.
Myasthenia gravis
.
Nat Rev Dis Primers
.
2019
;
5
(
1
):
30
.
2.
Jaretzki
A
,
Kaminski
HJ
,
Keesey
JC
,
Penn
AS
,
Barohn
RJ
,
Ernstoff
RM
, et al
.
Myasthenia gravis: recommendations for clinical research standards. Task force of the medical scientific advisory board of the myasthenia gravis foundation of America
.
Neurology
.
2000
;
55
(
1
):
16
23
.
3.
Beloor Suresh
A
,
Asuncion
RMD
. Myasthenia Gravis. StatPearls. Treasure Island (FL) relationships with ineligible companies. Disclosure: Ria Monica Asuncion declares no relevant financial relationships with ineligible companies: StatPearls Publishing Copyright © 2023, StatPearls Publishing LLC., 2023.
4.
Gilhus
NE
.
Myasthenia gravis
.
N Engl J Med
.
2016
;
375
(
26
):
2570
81
.
5.
Dresser
L
,
Wlodarski
R
,
Rezania
K
,
Soliven
B
.
Myasthenia gravis: epidemiology, pathophysiology and clinical manifestations
.
J Clin Med
.
2021
;
10
(
11
):
2235
.
6.
McGrogan
A
,
Sneddon
S
,
de Vries
CS
.
The incidence of myasthenia gravis: a systematic literature review
.
Neuroepidemiology
.
2010
;
34
(
3
):
171
83
.
7.
Wartmann
H
,
Hoffmann
S
,
Ruck
T
,
Nelke
C
,
Deiters
B
,
Volmer
T
.
Incidence, prevalence, hospitalization rates, and treatment patterns in myasthenia gravis: a 10-year real-world data analysis of German claims data
.
Neuroepidemiology
.
2023
;
57
(
2
):
121
8
.
8.
Fang
F
,
Sveinsson
O
,
Thormar
G
,
Granqvist
M
,
Askling
J
,
Lundberg
IE
, et al
.
The autoimmune spectrum of myasthenia gravis: a Swedish population-based study
.
J Intern Med
.
2015
;
277
(
5
):
594
604
.
9.
Salari
N
,
Fatahi
B
,
Bartina
Y
,
Kazeminia
M
,
Fatahian
R
,
Mohammadi
P
, et al
.
Global prevalence of myasthenia gravis and the effectiveness of common drugs in its treatment: a systematic review and meta-analysis
.
J Transl Med
.
2021
;
19
(
1
):
516
.
10.
Westerberg
E
,
Punga
AR
.
Epidemiology of myasthenia gravis in Sweden 2006–2016
.
Brain Behav
.
2020
;
10
(
11
):
e01819
.
11.
Sanders
DB
,
Wolfe
GI
,
Benatar
M
,
Evoli
A
,
Gilhus
NE
,
Illa
I
, et al
.
International consensus guidance for management of myasthenia gravis: executive summary
.
Neurology
.
2016
;
87
(
4
):
419
25
.
12.
Narayanaswami
P
,
Sanders
DB
,
Wolfe
G
,
Benatar
M
,
Cea
G
,
Evoli
A
, et al
.
International consensus guidance for management of myasthenia gravis: 2020 update
.
Neurology
.
2021
;
96
(
3
):
114
22
.
13.
Engel-Nitz
NM
,
Boscoe
A
,
Wolbeck
R
.
Burden of illness in patients with treatment refractory myasthenia gravis
.
Muscle Nerve
.
2018
.
14.
Harris
L
,
Graham
S
,
MacLachlan
S
,
Exuzides
A
,
Jacob
S
.
A retrospective longitudinal cohort study of the clinical burden in myasthenia gravis
.
BMC Neurol
.
2022
;
22
(
1
):
172
.
15.
Schneider-Gold
C
,
Hagenacker
T
,
Melzer
N
,
Ruck
T
.
Understanding the burden of refractory myasthenia gravis
.
Ther Adv Neurol Disord
.
2019
;
12
:
1756286419832242
.
16.
Mantegazza
R
,
Antozzi
C
.
When myasthenia gravis is deemed refractory: clinical signposts and treatment strategies
.
Ther Adv Neurol Disord
.
2018
;
11
:
1756285617749134
.
17.
Alhaidar
MK
,
Abumurad
S
,
Soliven
B
,
Rezania
K
.
Current treatment of myasthenia gravis
.
J Clin Med
.
2022
;
11
(
6
):
1597
.
18.
Menon
D
,
Barnett
C
,
Bril
V
.
Novel treatments in myasthenia gravis
.
Front Neurol
.
2020
;
11
:
538
.
19.
Phillips
G
,
Abreu
C
,
Goyal
A
,
Li
Y
,
Whangbo
A
,
Gelinas
D
, et al
.
Real-world healthcare resource utilization and cost burden assessment for adults with generalized myasthenia gravis in the United States
.
Front Neurol
.
2021
;
12
:
809999
.
20.
Harris
L
,
Graham
S
,
MacLachlan
S
,
Exuzides
A
,
Jacob
S
.
Healthcare resource utilization by patients with treatment-refractory myasthenia gravis in England
.
J Med Econ
.
2019
;
22
(
7
):
691
7
.
21.
Ludvigsson
JF
,
Andersson
E
,
Ekbom
A
,
Feychting
M
,
Kim
JL
,
Reuterwall
C
, et al
.
External review and validation of the Swedish national inpatient register
.
BMC public health
.
2011
;
11
:
450
.
22.
Wettermark
B
,
Hammar
N
,
Fored
CM
,
Leimanis
A
,
Otterblad Olausson
P
,
Bergman
U
, et al
.
The new Swedish prescribed drug register--opportunities for pharmacoepidemiological research and experience from the first six months
.
Pharmacoepidemiol Drug Saf
.
2007
;
16
(
7
):
726
35
.
23.
Brooke
HL
,
Talbäck
M
,
Hörnblad
J
,
Johansson
LA
,
Ludvigsson
JF
,
Druid
H
, et al
.
The Swedish cause of death register
.
Eur J Epidemiol
.
2017
;
32
(
9
):
765
73
.
24.
Ludvigsson
JF
,
Svedberg
P
,
Olén
O
,
Bruze
G
,
Neovius
M
.
The longitudinal integrated database for health insurance and labour market studies (LISA) and its use in medical research
.
Eur J Epidemiol
.
2019
;
34
(
4
):
423
37
.
25.
Andersen
JB
,
Heldal
AT
,
Engeland
A
,
Gilhus
NE
.
Myasthenia gravis epidemiology in a national cohort; combining multiple disease registries
.
Acta Neurol Scand
.
2014
;
129
(
198
):
26
31
.
26.
Myllynen
C
,
Sarkkinen
J
,
Atula
S
,
Tienari
P
,
Kekäläinen
E
,
Laakso
SM
.
A skewed ratio of free light chains is more common in patients with late-onset than early-onset myasthenia gravis
.
Immunol Lett
.
2023
;
260
:
81
8
.
27.
Cortés-Vicente
E
,
Álvarez-Velasco
R
,
Segovia
S
,
Paradas
C
,
Casasnovas
C
,
Guerrero-Sola
A
, et al
.
Clinical and therapeutic features of myasthenia gravis in adults based on age at onset
.
Neurology
.
2020
;
94
(
11
):
e1171
80
.
28.
Grob
D
,
Brunner
N
,
Namba
T
,
Pagala
M
.
Lifetime course of myasthenia gravis
.
Muscle Nerve
.
2008
;
37
(
2
):
141
9
.
29.
Wang
L
,
Zhang
Y
,
He
M
.
Clinical predictors for the prognosis of myasthenia gravis
.
BMC Neurol
.
2017
;
17
(
1
):
77
.
30.
Martinez-Harms
R
,
Barnett
C
,
Bril
V
.
Time to achieve a patient acceptable symptom state in myasthenia gravis
.
Front Neurol
.
2023
;
14
:
1187189
.
31.
Dewilde
S
,
Phillips
G
,
Paci
S
,
De Ruyck
F
,
Tollenaar
NH
,
Janssen
MF
.
People diagnosed with myasthenia gravis have lower health-related quality of life and need more medical and caregiver help in comparison to the general population: analysis of two observational studies
.
Adv Ther
.
2023
;
40
(
10
):
4377
94
.
32.
Lehnerer
S
,
Jacobi
J
,
Schilling
R
,
Grittner
U
,
Marbin
D
,
Gerischer
L
, et al
.
Burden of disease in myasthenia gravis: taking the patient’s perspective
.
J Neurol
.
2022
;
269
(
6
):
3050
63
.
33.
Aarli
JA
.
Late-onset myasthenia gravis: a changing scene
.
Arch Neurol
.
1999
;
56
(
1
):
25
7
.
34.
Alkhawajah
NM
,
Oger
J
.
Late-onset myasthenia gravis: a review when incidence in older adults keeps increasing
.
Muscle Nerve
.
2013
;
48
(
5
):
705
10
.
35.
Fan
L
,
Ma
S
,
Yang
Y
,
Yan
Z
,
Li
J
,
Li
Z
.
Clinical differences of early and late-onset myasthenia gravis in 985 patients
.
Neurol Res
.
2019
;
41
(
1
):
45
51
.
36.
Pasqualin
F
,
Guidoni
SV
,
Ermani
M
,
Pegoraro
E
,
Bonifati
DM
.
Outcome measures and treatment effectiveness in late onset myasthenia gravis
.
Neurol Res Pract
.
2020
;
2
:
45
.
37.
Pakzad
Z
,
Aziz
T
,
Oger
J
.
Increasing incidence of myasthenia gravis among elderly in British Columbia, Canada
.
Neurology
.
2011
;
76
(
17
):
1526
8
.
38.
Vincent
A
,
Clover
L
,
Buckley
C
,
Grimley Evans
J
,
Rothwell
PM
;
UK Myasthenia Gravis Survey
.
Evidence of underdiagnosis of myasthenia gravis in older people
.
J Neurol Neurosurg Psychiatry
.
2003
;
74
(
8
):
1105
8
.
39.
Volkova
L
,
Toporkova
M
,
Galunova
A
,
Sokolova
L
.
Clinical features of myasthenia gravis in elderly and senile patients
.
BIO Web Conf
.
2020
;
22
:
01018
.
40.
Tachikawa
T
,
Takeshima
S
,
Kawate
N
.
Diagnosis and evaluation of elderly-onset myasthenia gravis: a case report
.
Medicine
.
2024
;
103
(
4
):
e36989
.
41.
Rice
JB
,
White
AG
,
Scarpati
LM
,
Wan
G
,
Nelson
WW
.
Long-term systemic corticosteroid exposure: a systematic literature review
.
Clin Ther
.
2017
;
39
(
11
):
2216
29
.
42.
Harris
L
,
Allman
PH
,
Sheffield
R
,
Cutter
G
.
Longitudinal analysis of disease burden in refractory and nonrefractory generalized myasthenia gravis in the United States
.
J Clin Neuromuscul Dis
.
2020
;
22
(
1
):
11
21
.
43.
Mahic
M
,
Bozorg
A
,
Rudnik
J
,
Zaremba
P
,
Scowcroft
A
.
Healthcare resource use in myasthenia gravis: a US health claims analysis
.
Ther Adv Neurol Disord
.
2023
;
16
:
17562864221150327
.
44.
Vuong
TD
,
Wei
F
,
Beverly
CJ
.
Absenteeism due to functional limitations caused by seven common chronic diseases in US workers
.
J Occup Environ Med
.
2015
;
57
(
7
):
779
84
.
45.
Frost
A
,
Svendsen
ML
,
Rahbek
J
,
Stapelfeldt
CM
,
Nielsen
CV
,
Lund
T
.
Labour market participation and sick leave among patients diagnosed with myasthenia gravis in Denmark 1997–2011: a Danish nationwide cohort study
.
BMC Neurol
.
2016
;
16
(
1
):
224
.
46.
Landfeldt
E
,
Pogoryelova
O
,
Sejersen
T
,
Zethraeus
N
,
Breiner
A
,
Lochmüller
H
.
Economic costs of myasthenia gravis: a systematic review
.
Pharmacoeconomics
.
2020
;
38
(
7
):
715
28
.
47.
Ting
A
,
Story
T
,
Lecomte
C
,
Estrin
A
,
Syed
S
,
Lee
E
.
A real-world analysis of factors associated with high healthcare resource utilization and costs in patients with myasthenia gravis receiving second-line treatment
.
J Neurol Sci
.
2023
;
445
:
120531
.