Cost of Illness in Subarachnoid Hemorrhage
A German Longitudinal Study
Background and Purpose—Aneurysmal subarachnoid hemorrhage (SAH) is a cerebrovascular disease with a high mortality rate and severe disability. Longitudinal studies investigating health-economic costs in SAH are scare and only one of them analyzed cost-driving factors. The objective was to evaluate first-year costs in German patients with aneurysmal SAH and to identify independent determinants of costs.
Methods—One hundred thirteen incident cases of aneurysmal SAH treated in the Department of Neurosurgery and Neuroradiology at the University of Bonn (catchment area of 500 000 people) between January 2004 and December 2005 were eligible for the study. Cost data were collected using health-economic questionnaires applied at baseline and 6- and 12-month follow-up time. All costs are expressed in (year 2009 values). Clinical assessments were performed using Hunt and Hess scale, Barthel Index, and Rankin Scale. Independent cost-driving factors were determined using multiple regression analysis.
Results—The total first-year costs were 38 300 (95% CI, 34 490 to 43 100) per patient. Direct costs accounted for 58.7% of total costs and were mainly paid by the health insurance (92.0%). Inpatient costs were the main cost component of direct and total costs (42.8% of total costs). The major cost-driving factors of total costs were younger age and worse functional outcome at 12-month follow-up (Barthel Index).
Conclusions—Aneurysmal SAH is a cerebrovascular disease with considerable health-economic burden. Healthcare programs aimed at reducing the burden of SAH on society and individuals should consider cost-driving factors of SAH. Further health-economic studies investigating cost-driving factors of SAH in different countries are needed.
Aneurysmal subarachnoid hemorrhage (SAH) is a cerebrovascular disease that is characterized by a high mortality rate and severe long-term disability. It accounts for approximately 5% of strokes and imposes a significant economic burden on society, individual patients, and their families.1–3 In Western populations, the incidence of SAH is 4 to 19 per 100 000/year.4 More than 11 400 people per year are diagnosed with SAH in Germany.5
An increasing number of health-economic evaluations in cerebrovascular diseases was published in recent years. The majority calculated costs of ischemic stroke, intracerebral hemorrhage, and transient ischemic attacks.2,3,6–9 To date, only a few health-economic evaluations were performed in aneurysmal SAH. They focused mainly on costs of hospitalization and other inpatient costs related to SAH.7,10,11 Two studies used a longitudinal approach and evaluated costs in the first 12 to 24 months after SAH.1,12 The study by Roos et al calculated direct costs of aneurysmal SAH,12 whereas direct and indirect costs were evaluated in the post hoc analysis of the International Subarachnoid Aneurysm Trial (ISAT) published recently.1 To the best of our knowledge, other studies investigating direct and indirect costs of aneurysmal SAH are not currently available. Only one of the studies performed multivariate analyses to identify independent factors in aneurysmal SAH.1
The objective of this longitudinal study was to evaluate first-year costs in German patients after aneurysmal SAH and to identify independent cost-driving factors.
Patients and Methods
Study Design and Inclusion Criteria
The study was performed in the Department of Neurosurgery and Neuroradiology at the University of Bonn, which is a large German neurosurgical center with a well-defined catchment area of approximately 500 000 people. The overall amount of 143 incident cases with ruptured and unruptured cerebral aneurysms treated between January 2004 and December 2005 was screened for the participation in the study. All patients underwent either endovascular coiling or surgical clipping in accordance with the results of the ISAT study.1 The inclusion criteria were: (1) definite diagnosis of SAH proven by cranial CT, cranial MRI, or lumbar puncture; (2) aneurysmal cause of SAH as demonstrated by intra-arterial or by cranial CT angiography; and (3) patients who could give informed consent for participation in the study. Patients with incidental aneurysms were excluded. According to the inclusion criteria, 113 patients with aneurysmal SAH were eligible for the study. Of these, 12 were excluded because their clinical or health-economic data were incomplete. There were no significant differences in age, gender, or disease severity between dropouts and remaining 101 study completers. The study protocol was approved by the local ethics committee of the University of Bonn. Informed consent was obtained from all study participants or their legal caregivers.
A complete medical and neurological examination was performed by a specialist in cerebrovascular diseases (study investigator) in each patient at admission at the time of discharge and at 6- and 12-month follow-up time points. Clinical status at admission was evaluated using the World Federation of Neurological Surgeons scale13 and the Hunt and Hess scale (H&H).14 The following clinical parameters were evaluated at discharge as well as at 6- and 12-month follow-up time points: (1) functional status and disability (modified Rankin Scale and Barthel Index). The Barthel Index (BI) measures the disability on a scale from 0 (totally dependent and bedridden state) to 100 (fully independent).15 The modified Rankin Scale (mRS) is a measure of functional status ranging from 0 (no symptoms) to 5 (severe impairment). Scores from 0 to 2 on the mRS were considered as independency in activities of daily living and scores from 3 to 5 were classified as moderate to severe disability16; (2) Depression (Beck Depression Inventory)17 and cognitive impairment (Mini-Mental State Examination)18; and (3) Health-related quality of life. The EuroQol consists of a questionnaire describing 5 dimensions of the health-related quality of life (“mobility,” “self-care,” “usual activities,” “pain/discomfort,” “anxiety/depression”) and a visual analog scale scoring from 0 to 100 with 0 representing the worst and 100 the best imaginable health state.19
Health-economic data were collected through a standardized questionnaire administered at 6- and 12-month follow-up. Each questionnaire covered the resource use of the preceding 6 months, respectively. The health-economic questionnaire used in this study was developed and applied in previous cost of illness studies in neurological diseases, including cerebrovascular diseases.2,20–22
Costs were calculated for the 12-month period from a societal perspective and adjusted to 2009 Euros () using the Consumer Price Index.5 Only SAH-related costs were included in the cost calculation. The following cost categories were examined in the analysis.
The expenditures of the health insurance company and the out-of-pocket cost of patients and their families were categorized as direct costs.
Health Insurance Costs
Costs for inpatient care (hospitalization and rehabilitation) were calculated using the official reimbursement catalogue of the German Diagnosis-Related Groups (2005).23 Costs for outpatient care were obtained by multiplying the number of outpatient visits by the provider-specific charge according to the official German doctor’s fee scale (“Einheitlicher Bewertungsmaßstab”).24 The number of outpatient diagnostic procedures and nonmedical ancillary therapies (physiotherapy, speech therapy, and occupational therapy) was multiplied by the German charge (“Einheitlicher Bewertungsmaßstab”) to calculate costs for diagnostics and ancillary treatment. Drug costs were calculated using the official German price list of drugs (“Rote Liste”).25 Only SAH-related drug costs were included in the analysis (eg, antihypertensive drugs, antiepileptic drugs for treatment of epileptic seizures after SAH).
Copayments by patients were subtracted from the official outpatient tariffs and drug prices and added to the out-of-pocket costs.
Formal care or professional home care was calculated according to the regulations for financial support in case of disability defined by the Federal Ministry for Health and Social Security. According to these regulations, financial support provided by health insurance for professional nursing care depends on the disability and amounts to 384 (first level), 921 (second level), and 1432 (third level) per month.26 Patients and their caregivers reported the level of financial support related to SAH.
Out-of-pocket costs consisted of costs for informal care and copayments for treatments and drugs. A replacement approach was used to calculate the costs for informal care or home care provided by family and friends. According to this approach, costs were calculated similarly to costs for formal care considering the three levels of financial support described previously.
Copayments by patients and families for treatments and drugs were calculated based on patient statements. Income losses caused by SAH-related early retirement or temporary sickness were not calculated separately to avoid double counting with the indirect costs. As such, they were treated as productivity losses (see subsequently).
Indirect costs, or productivity losses, due to premature retirement were calculated using the human capital approach. The mean gross income given by Federal Statistics5 was divided by 365 calendar days and then multiplied by the calendar days in the remaining study period before the 65th birthday (official retirement age in Germany). For productivity losses due to sick leave, the mean gross income was divided by 365 calendar days and then multiplied by the days of sick leave. In case of early mortality, productivity losses were calculated similar to premature retirement. The mean gross income was divided by 365 calender days and then multiplied by number of days in the remaining study period to official retirement age.
Statistical analysis was performed using STATA 9.2 (StataCorp, Stata Statistical Software, Release 9.2 ; College Station, Texas). Cost variables are in most cases right-skewed and, therefore, traditional statistical methods are inappropriate for analyzing differences in mean costs.27 Potential skewness in the cost data can be accommodated by using the 95% CI calculated by a nonparametric bootstrap algorithm.28 In our analysis, a bias-corrected and accelerated bootstrapping method with B=2000 bootstrap replications was used to calculate 95% CIs.28 Differences in mean costs between groups were tested using a bootstrap t test.27 Statistical inference was based on a probability value <0.05. Multiple regression analysis with SEs estimated using the bootstrapping technique was performed to identify cost-driving factors. Clinical variables used as potential cost-driving factors in the multivariate analyses were specified based on the results of the univariate analysis and data of previous studies.1,7,10,12,29 The R2 method was applied to calculate the fraction of variability explained.30
Demographics and Clinical Features
Among 101 patients with aneurysmal SAH who participated in the study, 29.7% (n=29) were male and 70.3% (n=72) were female. The mean age of study participants was 51.7±13.0 years. Demographic, social, and clinical parameters stratified by age groups are shown in Table 1.
A total of 19 patients died during the observation period. At the time of discharge, 42.7% (n=35) of survivals had moderate or severe disability (BI ≤70). The functional status of patients measured using BI improved at 6- and 12-month follow-up (21.9%, n=18 and 17.0%, n=14 of patients with BI ≤70, P<0.01). The proportion of patients with depressive symptoms (Beck Depression Inventory >9) increased from 24.8% (n=25) at discharge to 57.4% (n=58) at the 12-month follow-up point (P<0.001).
The total costs were calculated at 38 300 (95% CI, 34 490 to 43 100) per patient per year. Table 2 shows the components of total costs. Total costs stratified by H&H at admission are depicted in the Figure. They were increased in patients with H&H 3 to 5 by factor 1.4 as compared with patients with H&H 1 to 2 (P=0.04).
Direct costs accounted for 58.7% of total costs and were mainly paid by the health insurance company (92.0%). Inpatient costs were the main cost component of direct and total cost in the first year after aneurysmal SAH (42.8% of total costs). All patients with SAH were hospitalized and 46.5% (n=47) of them were discharged to the rehabilitation clinics. The mean length of stay in the hospital was 14.9±19.1 days. Outpatient care received 91.1% (n=92) of patients. In 78.3% of patients (n=72), follow-up with neuroimaging was performed in an outpatient setting. In 44.4% (n=32) of these patients, neuroimaging was performed more than once. Angiography was the most expensive imaging procedure and accounted for 59.3% of imaging costs.
Drug costs made up to 4.2% of direct costs (Table 2). Patients with SAH received informal care more often than formal care and costs of informal care exceeded costs of formal care by a factor of 1.6. Indirect costs caused by SAH-related premature retirement or sick leave amounted to 15 840 (95% CI, 13 150 to 18 820) and accounted for 41.4% of total costs.
The results of bivariate analyses of determinants of total costs are shown in Table 3. The following clinical variables were associated with increased total costs: younger age, increased disease severity at admission (H&H 3 to 5), and worse functional outcome after 12 months (BI, mRS). These variables were used in the multivariate analysis (Table 4). mRS was excluded from the multivariate analysis due to collinearity with the BI. The independent determinants of increasing costs identified in the multiple regression analysis were younger age and worse functional outcome at 12-month follow-up (BI). Together, these were able to explain 44.0% (adjusted R2) of the variance in total costs.
In this longitudinal study, the socioeconomic burden of SAH was evaluated in a cohort of German patients treated in a large neurosurgical/neuroradiological center with a well-defined catchment area of approximately half a million people. Cost data were calculated on the level of individual patients. This approach, also known as the “bottom-up” approach, allows obtaining more precise data than a “top-down” approach, which calculates costs by splitting down highly aggregated statistical data such as reports of national statistics.
The total costs in the first year after aneurysmal SAH were 38 300 per patient and consisted largely of direct costs (58.7% or 22 470). Roos et al calculated similar first-year direct costs of aneurysmal SAH in The Netherlands (28 830, extrapolated to the 2009 price level).12 According to data of Rivero-Arias et al, direct costs of aneurysmal SAH in the United Kingdom were 33 290 per patient (2009 price level) during the first 24 months after SAH.1 However, authors did not include primary care and informal care in the cost calculation and underestimated direct costs. To the best of our knowledge, there are no published studies investigating first-year costs after aneurysmal SAH in other countries. Considering the incidence of SAH in Germany and annual costs per patient calculated in our study, the total annual costs of incident SAH in Germany can be estimated at approximately 437 million.
A few studies focused on inpatient costs of aneurysmal SAH. The highest inpatient costs were calculated in the United States (US $41 905 per patient, or 28 900, 2009 price level).10 Inpatient costs of aneurysmal SAH in our study were lower (16 390) and were comparable to costs calculated by Venketasubramanian et al in Singapore (16 030, 2009 price level).7 In Australia, inpatient costs were US $25 880 (16 590, 2009 price level).11 Generally, inpatient costs were one of the main cost components. Depending on the study, they accounted for 75% to 92% of first-year direct costs after SAH.1,12 Similarly, inpatient costs accounted for 73% of first-year direct costs in our study.
Corresponding to the results of other studies of acute cerebrovascular diseases, first-year outpatient costs of SAH in our study were much lower than inpatient costs.2,6,12 Kolominski-Rabas et al showed that the ratio outpatient cost/inpatient costs is increasing in the subsequent years after an acute cerebrovascular event.6 This pattern can be expected for the long-term costs of SAH; however, they were beyond the scope of our study. Neuroimaging was one of the main components of outpatient costs in the first year after SAH. Among other imaging procedures, angiography was the most costly (59% of imaging costs). This corresponds well with the results of the study by Roos et al.12
In contrast to chronic neurological diseases, costs of medication in acute cerebrovascular diseases are usually low.2,8,12 Costs of antihypertensive medication and drugs used in the treatment of complications of SAH in our study were relatively low in comparison to other cost components. Similar to the results of another European study, they accounted for 4% of direct costs in SAH.12
Only one previous study by Rivero-Arias et al used multivariate analysis to identify cost-driving factors in SAH.1 In contrast to the results of our study, increasing age was associated with higher total costs. As mentioned previously, authors did not include primary care and informal care in cost calculation and this interfered with the results of the multivariate analysis. Costs of informal care in our study were higher in younger patients, because they were preferably cared for by their family members. This could explain the difference in the results of the multivariate analysis between the study by Rivero-Arias et al and our study.1 Besides younger age, a worse functional outcome at 12-month follow-up was identified as a cost-driving factor in our analysis. Unfortunately, clinical outcome was not included in the multivariate analysis of Rivero-Arias et al.1 These authors identified disease severity at admission measured in World Federation of Neurological Surgeons grades as another cost-driving factor in SAH. In our study, clinical outcome measured by the BI was a stronger predictor of increasing costs than disease severity at admission.
The cost-driving factors identified in our study should be considered in healthcare programs in SAH. The patients of working age should be provided with advanced rehabilitation programs aimed on return to working life. The healthcare programs should provide adequate investments in the acute phase of SAH treatment to improve functional outcome. Our previous analysis showed that cost-intensive specialized treatment approaches in acute stroke care such as stroke unit treatment are associated with better functional outcome and reduce postacute costs.8 Increased investments in the early disease pay off in reduced costs of rehabilitation and reduced indirect costs.
Despite a careful study design, our study carries a few limitations: (1) it is a single-center study and, therefore, extrapolation of data on the national level is limited. However, SAH treatment in Germany is carried out in specialized neurosurgical centers, using similar standardized procedures, and costs calculated in our study center should provide an adequate approximation of per-patient costs in Germany. Future multicenter studies will provide more precise estimations; (2) the “bottom-up” approach relies on data collected directly from individuals. We cannot exclude that some details were not reported by patients resulting in an underestimation or bias toward costs; (3) informed consent could not be obtained from patients with aphasia or impaired consciousness. The exclusion of these patients with severe SAHs may lead to an underestimation of costs; and (4) finally, residual confounding by unmeasured variables in the multivariate analyses cannot be excluded.
In conclusion, aneurysmal SAH is a cerebrovascular disease, which affects predominantly young people and results in a considerable economic burden for the society, individual patients, and their families. Unfortunately, independent determinants of SAH-related costs were evaluated only in one previous study. Our study identified younger age and worse long-term functional outcome as independent factors increasing socioeconomic costs of aneurysmal SAH. Further health-economic studies investigating cost-drivers of SAH in different countries are needed to promote the development of healthcare programs aimed at reduction the economic burden of aneurysmal SAH on society and patients. Meta-analysis of these studies will show how different healthcare systems influence the components of cost in SAH and which patterns of health care in SAH should be integrated in other countries to optimize costs.
We thank Jennifer Manne, MD, PhD, for language editing.
↵*R.D. and Y.W. contributed equally to this article.
- Received April 7, 2010.
- Revision received May 14, 2010.
- Accepted July 29, 2010.
Rivero-Arias O, Wolstenholme J, Gray A, Molyneux AJ, Kerr RS, Yarnold JA, Sneade M. The costs and prognostic characteristics of ischaemic neurological deficit due to subarachnoid haemorrhage in the United Kingdom. Evidence from the MRC International Subarachnoid Aneurysm trial. J Neurol. 2009; 256: 364–373.
de Rooij NK, Linn FH, van der Plas JA, Algra A, Rinkel GJ. Incidence of subarachnoid haemorrhage: A systematic review with emphasis on region, age, gender and time trends. J Neurol Neurosurg Psychiatry. 2007; 78: 1365–1372.
The Federal Statistical Office Germany. Epidemiology of cerebrovascular diseases in Germany. Available at: www.gbe-und.de/gbe10/trecherche.prc_them_rech2?tk=8500&tk2=8700&ut_string=Zerebrovaskuläre_Krankheiten/Schlaganfall&ber=1&tab=1&gra=1&def=1&link=1&son=1&anz_ber=2&anz_tab=23&anz_gra=4&anz_def=4&anz_link=0&anz_son=4&p_sprache=D&x=&p_news=&button=1&p_uid=gastd&p_aid=56496601&cb_wk=dummy&p_th_id=8900&p_bread_fid=&p_item_news=&next_tr=1&erg_art=ALL#ALL. Accessed January, 10, 2010.
Kolominsky-Rabas PL, Heuschmann PU, Marschall D, Emmert M, Bnurs NB, Neundoerfer B, Schoeffski O, Krobot KJ. Lifetime cost of ischemic stroke in Germany: results and national projections from a population-based stroke registry: the Erlangen Stroke Project. Stroke. 2006; 37: 1179–1183.
Roos YB, Dijkgraaf MG, Albrecht KW, Beenen LF, Groen RJ, de Haan RJ, Vermeulen M. Direct costs of modern treatment of aneurysmal subarachnoid hemorrhage in the first year after diagnosis. Stroke. 2002; 33: 1595–1599.
Teasdale GM, Drake CG, Hunt W, Kassell N, Sano K, Pertuiset B, De Villiers JC. A universal subarachnoid hemorrhage scale: report of a committee of the World Federation of Neurosurgical Societies. J Neurol Neurosurg Psychiatry. 1988; 51: 1457.
Beck A, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Geriatr Psychiatry. 1961; 4: 561–567.
Schepelmann K, Winter Y, Spottke AE, Claus D, Grothe C, Schroder R, Heuss D, Vielhaber S, Mylius V, Kiefer R, Schrank B, Oertel WH, Dodel R. Socioeconomic burden of amyotrophic lateral sclerosis, myasthenia gravis and facioscapulohumeral muscular dystrophy. J Neurol. 2010; 257: 15–23.
The German Diagnosis Related Groups (G-DRG). The official reimbursement catalogue of the German Diagnosis-Related Groups in 2005. Available at: www.g-drg.de/cms/index.php/inek_site_de/layout/set/standard/Archiv/Systemjahr_2005_bzw._Datenjahr_2003#sm2. Accessed December, 15, 2009.
Deutscher Arzteverlag. Einheitlicher Bewertungsmasstab. Köln: Deutscher Arzteverlag; 2006.
The Federation of German Pharmaceutical Industry. Rote liste. Weinheim: Rote Liste Verlag; 2006.
The Federal Ministry for Health and Social Secure. Available at: www.bmg.bund.de. Accessed January 5, 2010.
Efron B, Tibshirani RJ. An Introduction to the Bootstrap. London: Chapman and Hall; 1993.
Bardach NS, Olson SJ, Elkins JS, Smith WS, Lawton MT, Johnston SC. Regionalization of treatment for subarachnoid hemorrhage: a cost-utility analysis. Circulation. 2004; 109: 2207–2212.
Harrell FE. Regression Modeling Strategies. New York: Springer; 2001.