Costs of rheumatoid arthritis during the period
1990–2010: a register-based cost-of-illness study
in Sweden
Almina Kalkan, Eva Hallert, Lars Bernfort, Magnus Husberg and Per Carlsson
Linköping University Post Print
N.B.: When citing this work, cite the original article.
This is a pre-copyedited, author-produced PDF of an article accepted for publication in Rheumatology following peer review. The definitive publisher-authenticated version:
Almina Kalkan, Eva Hallert, Lars Bernfort, Magnus Husberg and Per Carlsson, Costs of rheumatoid arthritis during the period 1990–2010: a register-based cost-of-illness study in Sweden, 2013, Rheumatology.
is available online at:
http://dx.doi.org/10.1093/rheumatology/ket290
Copyright: Oxford University Press (OUP): Policy B - Oxford Open Option B
http://www.oxfordjournals.org/
Postprint available at: Linköping University Electronic Press
Costs of rheumatoid arthritis 1990–2010:
A register based cost-of-illness study in Sweden
Almina Kalkan1*, MSc; Eva Hallert1,2, PhD; Lars Bernfort1, PhD; Magnus Husberg1, BSc; Per Carlsson1, PhD
1. Center for Medical Technology Assessment, Department of Medical and Health Sciences, Linköping University, SE- 581 83 Linköping, Sweden
2. Department of Cardiovascular Diseases and Specialty Medicine, University Hospital, SE- 581 85 Linköping, Sweden
*Corresponding author Email adresses: AK: almina.kalkan@liu.se EH: eva.hallert@liu.se LB: lars.bernfort@liu.se MH: magnus.husberg@liu.se PC: per.carlsson@liu.se
2
Abstract
Objectives: To study the total socioeconomic impact of rheumatoid arthritis (RA) in Sweden
1990-2010 and to analyze possible changes in costs during this period. The period is deliberately chosen to cover 10 years before and 10 years after the introduction of biological drugs.
Methods: A prevalence based cost of illness study is conducted, based on data from national and
regional registries.
Results: There was a decrease in the utilisation of RA-related inpatient care, as well as sick leave and
disability pension during 1990-2010 in Sweden.Total costs for RA are presented in current prices as well as inflation-adjusted with the consumer price index (CPI) and a healthcare price index. The total fixed cost of RA was 454 million euro in 1990, adjusted to the price level of 2010 with CPI. This cost increased to 600 million euro in 2010 and the increase was mainly due to the substantially increasing costs for pharmaceuticals. Of the total costs, drug costs increased from representing 3% to 33% between 1990 and 2010. Consequently, the amount of total costs accounting for indirect costs for RA is lowered from 75% in 1990 to 58% in 2010.
Conclusions: Inflation-adjusting with the CPI, which is reasonable from a societal perspective, there
is a 32% increase in the total fixed cost of RA between 1990 and 2010. This suggests that decreased hospitalisation and indirect costs have not fallen enough to offset the increasing cost of drug treatment.
Funding: Östergötland County Council , Sweden
3
INTRODUCTION
Rheumatoid arthritis (RA) is connected with substantial economic consequences both for patients and society (1–7). The cost picture of RA has changed during the last decade, with increasing drug costs after the introduction of biological drugs. A number of clinical trials have shown that biological drugs reduce disease activity and improve quality of life (8–10). They are, however, 30–40 times more costly than traditional disease-modifying antirheumatic drugs, DMARDs (3). Based on results from clinical trials, there has been an expectation that higher costs for drug treatment will partly be offset by a reduction in other healthcare consumption and increased work ability (11–14). However, recent research indicates that the effect on RA progression in real life might not be as good as reported in clinical trials (15–16). Data on work ability for patients prescribed biological drugs are conflicting, with reports of decreasing sick leave but also an increasing number of disability pensions (DPs) (17–22).
Long-term observational studies are of particular interest in evaluating the impact of biological drugs on the total costs of RA. Internationally, few longitudinal cost studies for RA that cover a part of the 2000s have been identified (15, 23–24), indicating a need for more comprehensive studies with a longer time perspective. Most previous cost studies in Sweden have been cohort based,focusing on partial costs and evaluating the mean annual cost per patient (25–28). Studies investigating total societal costs of RA have also been based on cohort data or calculated for one year at a time (29–31). In Sweden, there are national registries that provide a unique opportunity to examine long-term healthcare consumption as well as work participation for the most relevant cost items.
4
The objectives of the present study, using comprehensive Swedish national databases, are to examine changes in the total costs for RA during 1990–2010, ten years prior to the introduction of biological drugs and ten years after, and to discuss potential reasons for changes in costs.
METHODS
Swedish Healthcare System
Sweden has a tax-financed health insurance system allowing all residents sick leave benefits when they are unable to work. After reaching a payment level for outpatient visits of 1100SEK (about €100) annually, healthcare utilisation is free of charge. Similarly, all prescribed drugs, including biological drugs, are free of charge after a payment level of 2200SEK (about €200). There were no formal restrictions on the prescription of biological drugs in Sweden 1999-2010 and usage was among the highest in Europe (32).
Cost assignment
Costs were calculated using a societal perspective, as recommended for health economic analyses in Sweden (33). Total annual costs, both direct and indirect, were calculated based on prevalence of patients with RA as the primary cause of treatment or work absence. Direct costs included costs for inpatient and outpatient visits, surgical procedures and pharmaceuticals. Prices from healthcare pricelists were used as reasonable proxies for opportunity costs. Indirect costs comprised productivity loss due to sick leave and DP.
An important issue in the present study is to show to what extent costs for society due to RA have changed over time. One option is to simply make cost comparisons in nominal terms
5
using current prices. This, however, does not take into account the price inflation—a euro today is worth less than a euro yesterday. A price index should be used to put nominal costs into fixed terms, providing an accurate sense of what a euro can buy today vs. yesterday, be it healthcare or something else.
In the present study, both current and fixed prices are presented. For current prices, healthcare consumption and days with sick leave or DP for each specific year were multiplied with average healthcare prices and average annual cost of labour of that year. For fixed prices, the sum was thereafter inflation-adjusted to equal the price-level of 2010. Different indexes can be used in inflation- adjusting prices of various resources (34). If the prices of the studied item are rising at the rate of the general inflation and we are interested in what the amount of euro spent for RA in 1990 could buy in any sector of the economy in 2010, then the consumer price index (CPI) provides an adequate way to correct for inflation. However, if the specific item does not follow the rate of the general inflation and we are interested in what the euro spent for RA in 1990 could buy in solely healthcare in 2010, a specific healthcare measure should be used.
In the present study, prices were inflation-adjusted in two ways. First with CPI, reflecting the average price changes in the economy, and secondly with a healthcare price index, reflecting the price changes in the healthcare sector.
Costs were converted to euro using the exchange rate of 2010 (1€=9.54 SEK). Registers did not cover social services and care for the elderly in the municipalities, or non-medical direct costs such as transportation, devices and home healthcare services. They were hence not included in the analysis. Costs for premature death as well as intangible costs, such as pain
6
and suffering, were neither included due to lack of data. Table 1 depicts the included cost domains, their respective data sources and years covered. All data were derived from reliable national and regional registries except for outpatient care and sick leave during the 90s when data are missing for many years and had to be estimated.
Direct costs
Inpatient care
Complete inpatient care data, including surgical interventions, were available in the inpatient care register for the whole study period. Data was selected on number of patients, hospital admissions and days in hospital each year with RA as primary diagnosis. In 1992, a reform was implemented in Sweden that shifted responsibility for long-term and elderly care from the county councils to the municipalities. In order to avoid double counting, data for 1992 were also used as an estimate for the two previous years, 1990 and 1991. The average daily cost of internal medicine departments was used as unit cost, since cost data for rheumatology departments were limited (35-38).
Outpatient care
National data in the outpatient care register were not complete before 2005. Instead, data were derived from regional databases in the Östergötland County Council, Västra Götaland Region and the Region of Skåne, where official data linked to the diagnostic codes were available from 2000 and onwards. The total population of the three areas is equivalent to approximately 1/3 of the Swedish population, and the prevalence of RA in the three areas is similar to the national prevalence (39), suggesting that these data are an appropriate estimate for total outpatient costs. Costs for RA-related outpatient care, including
non-7
physician service utilization and diagnostic and surgical procedures were extracted and then extrapolated to the national level by the regional populations. The number of outpatient visits per inhabitant was similar in all three areas, and furthermore, the number of visits remained stable around ninety thousand per year during the studied years.
Since outpatient data were missing for the 1990s, and outpatient visits remained unchanged during the whole decade prior to 2010, the number of visits to physicians during 1990–1999 was also assumed to be ninety thousand per year. The average cost for physician visits was used as unit cost in the 1990s (40).
Drugs
Costs for drugs prescribed to patients with RA (excluding biological drugs) were available for 1991, 1997 and 2001 (41). A linear trend was assumed between the years with data. Drug costs for RA have previously been shown to be at a rather constant level between 1987 and 1997 (42) and our estimation is in line with the previous data. There were no corresponding data available for the 2000s. We assumed that the cost of drugs, excluding biological drugs, remained constant at the 2001 level throughout the 2000s, since the drugs were used both as mono-treatment and in combination with biological drugs. The total sales of biological drugs in Sweden, for all diagnoses, were available from the year 2000 onwards (43). Based on a recent Swedish study, the percentage of TNF-alpha inhibitors prescribed to patients with RA was calculated to be 80% of total prescriptions (44). This percentage was also applied to all other biological drugs, except rituximab. The proportion of rituximab prescribed to RA patients was 35%, based on information from the marketing
8
pharmaceutical company as well as a regional register database covering 230 rituximab patients (45).
Indirect costs
The human capital method was applied, assuming that all sickness absence of people below age 65 years is associated with loss of productivity. This method implies the simplified assumption of full employment. Our data for sick leave and DP do not show whether a person was on full- or part-time sickness absence. We assumed that our population followed the average full- and part-time distribution of the general population. Accordingly, average annual income, including costs for vacation, for men and women was used. In addition, pension funding and social fees of 40%, a weighted average for blue-collar and white-collar workers, were added (Statistics Sweden; http://www.scb.se).
Sick leave
Data for sick leave days due to RA, for men and women respectively, were obtained for 1991, 2001 and 2005–2010 (46–48). The level of sick leave for RA was correlated to the level of sick leave for all diagnoses, excluding RA, in the general population, r=0.823, p=0.012. The proportion of sick leave days due to RA was, however, slightly higher in 1991 than in 2001 and 2005-2010. For years with missing data, the number of sick leave days was calculated by multiplying the number of sick leave days for all diagnoses with the share of total sick leave days attributed to RA during the surrounding years, with a higher share for the first half of the period.
9
Data for the number of men and women with on-going RA-related DPs were obtained for 1991, 1996, 2002 and 2003–2010 (49–51). For years when data was missing, the number of DPs for RA was calculated by multiplying all DPs with the share of DPs attributed to RA for the years with data. We assumed that all individuals with DP received pensions during the whole year. Since the vast majority of patients with DP due to RA were above 55 years of age, the average income for ages 55–65 years was used.
No ethical approval was applied for since only official statistics and public databases were used.
RESULTS
The following section describes the changes in RA-related consumption of healthcare, sick leave and DPs. Total costs are then presented in current prices, as well as in 2010 prices inflation-adjusted both with CPI and the healthcare price index.
Inpatient care
Throughout the period studied, utilisation of inpatient care due to RA decreased in terms of numbers of patients, admissions and hospital days (Table 2). The annual decline in the numbers of admissions and hospital days was more pronounced between 1994 and 2000 than after year 2000. The number of hospital days dropped by 90% over the period and the number of admissions and number of patients decreased by more than 70% and 60%, respectively. The proportions of RA inpatient care decreased, with RA accounting for 1% of all inpatient days in the early 1990s and 0.3% in 2009.
10
Outpatient care due to RA remained rather unchanged with approximately 90,000 visits and 45,000 patients each year throughout the 2000s.
Sick leave
The number of sick leave days due to RA was halved over the period (Table 3). The annual percentage change fluctuated, but there was a downturn by the end of the period. Similar fluctuations and decline of sick leave was simultaneously seen in the general population, but was slightly more pronounced for RA.
Disability pension
The total prevalence of individuals with DPs due to RA declined after the mid-1990s from approximately 10 000 to 7000 (Table 3). Newly granted DPs due to RA fluctuate over the period. They decreased in the period of 1994 to 1998 and from 2005 to 2010, closely following the pattern of all newly granted DPs regardless diagnose in the general population. The changes in newly granted DPs are affecting the number of DPs the following years, which can be noted in decreasing prevalence at the end of the period. However, the prevalence is also affected by mortality and demographic factors such as age. During this period the amount of people in the older age groups increased. In total, there are decreasing proportions of RA-related DPs, where RA made up 3% of all DPs in the early 1990s and 1.5% in 2010. The corresponding figures for newly granted DPs due to RA were 2% in 1990 and 1% in 2010.
11
Costs of RA 1990-2010
The total cost of RA in current prices was doubled between 1990 and 2010 (Figure 1A). Hence, the decreases in healthcare consumption, disability pensions and sick leave due to RA are largely outweighed by price increases (see Supplement 1 for unit prices).
When total fixed costs were calculated to the price level of 2010 using CPI, there was a 32% increase in total fixed costs for RA (Figure 1B). The proportion of indirect costs decreased over the years, but still constituted the major part of total costs, with DP as major cost driver. Simultaneously, there was a sharp increase in the amount of direct costs, mainly after the introduction of biological drugs. In the later years, costs for biological drugs made up for an increasing share of total costs. A levelling off of total costs was however observed at the end of the period.
When inflation adjustment was made with the healthcare price index, there was a noticeable downturn in all costs except drug costs from the mid-90s and onwards (Figure 1C). This resulted in similar levels of total costs in the beginning and in the end of the period. Hence the increased costs for biological drugs were compensated by decreases in other cost domains. The main cost decreases occurred in indirect costs.
Figure 2 shows the proportion of total costs for RA represented by each cost type during the two decades. Drugs increased from 3% to 33% of total costs between 1990 and 2010, while indirect costs decreased from 75% to 57% of total costs. Simultaneously, inpatient care decreased from representing 15% to 3% of total costs.
12
DISCUSSION
There are a number of previously published cost studies for RA (1–5, 23–27, 29–31, 52–54). To our knowledge, this is the first study that covers the total costs of RA during 20 years. The period 1990-2010 was deliberatively chosen to cover 10 years before and 10 years after the introduction of biological drugs. By using national register data, selection bias has been avoided since the entire RA population was covered. We cannot distinguish the causality behind the cost changes but we can present the trajectory of cost development during this extended period. The present study differs from previous studies, which have been based on cohorts and/or examined individual cost items and years.
The various structures of healthcare systems in different countries make plain comparisons between studies difficult. In addition, access to biological drugs differs between countries (32). The present data reflect the development in Sweden and demonstrate that there was a decrease in the utilisation of RA-related inpatient care, as well as sick leave and DP during the two recent decades. Our data confirms previous research (21, 28), which also exhibited lowered numbers of days on sick leave, with the increased use of biological drugs. The continuous decreases in DP since the 90s are striking, since previous research has found DP either increasing or being stable with the increased use of biological drugs (21, 28).
However, while the decrease of sick leave in our data accelerated in the mid-2000s, the decrease in DPs started already in the 90s. Likewise, the decrease of inpatient care started in the 1990s and was more pronounced in the 1990s than in the 2000s. It has previously been shown that the number of DPs and sick leave days is correlated with changes in the economic situation in a country, the labour market and the regulatory setting (55). In
13
Sweden there were also general reductions of DPs, sick leave and inpatient care during this period.
Additionally, changes have occurred in the delivery of rheumatology care, starting in the nineties, with a shift from inpatient care to outpatient care, thus unrelated to the use of biologics (56). Nonetheless, when compared with all other diagnoses, the reductions for RA are more pronounced. Since the decreases started before the introduction of biological drugs, reductions may be due to other factors, such as intensified treatment with traditional DMARDs, starting in the early 1990s.
When prices are applied to the healthcare utilisation and productivity loss, there is an increase of about 32% in total costs of RA between 1990 and 2010 estimated with CPI 2010. Our cost estimate for RA is slightly lower than the estimates of a Swedish study from 2008 where the total cost for RA in Sweden in 2006 was at €820 million, in 2010 prices adjusted with CPI (30). Our estimates of drug costs and indirect costs are at a comparable level. However, Lundkvist et al. 2008 also included nonmedical costs and costs for informal care in the direct cost calculations, which represented 14% of total costs. In addition, the previous study estimated costs based on three cohort studies that were extrapolated to the national level, in contrast to the present study directly using total national costs for RA.
Our cost estimates are also in accordance with another study based on the Swedish rheumatology quality register estimating a total cost of RA to €524 million in 2009 (31). They are also in line with a previous estimate of the total costs of RA at €367 million in 1997, equivalent to €433 million in 2010 prices (29).
14
The choice of inflation adjusting index largely affects the results. Inflation-adjusting total costs with the healthcare price index results in a downward trend for all costs, apart from costs for biological drugs. The difference between the inflation-adjusting indexes is due to a faster price increase in the healthcare sector compared to the rest of the economy. While general prices in Sweden increased by 50% between 1990 and 2010, the nominal wages as well as prices in the labour-intense healthcare sector doubled (+100%).
The healthcare price index gives a result that might be intuitively more appealing since it follows the development in real terms (e.g. inpatient days) to a greater extent than the CPI. Also, if the only alternative use of resources spent on RA is in the healthcare sector, this result is more relevant. On the other hand, applying a societal perspective, the interesting thing is what the resources spent for RA could buy in any sector of the economy. Then the inflation-adjustment with CPI is more relevant. In addition, adjusting the nominal wages with CPI allows for an increase in inflation-adjusted, or real, annual wages. According to official statistics, real wages have increased by 38% during this period in Sweden. Using the healthcare price index, the whole increase in nominal wages is treated as inflation. In total, this speaks in favor of CPI.
Performing a sensitivity analysis of our results, the total costs increased by 9% at most which indicates robust results (see Supplement 2 for details).
Our study has several limitations. Firstly, some relevant data are lacking. This is the case for care given by the municipalities, informal care and intangible costs. Intangible costs, such as pain and fatigue, are substantial for RA patients, but are difficult to quantify. As a consequence, cost estimates of this kind inevitably underestimate the total effect of the
15
disease on the individual and society. We cannot dismiss the possibility that patients’ quality of life may improve with biological drugs, but this has not been within the scope of this study. Secondly, data for outpatient care and sick leave were not complete for all years in the 1990s, and average values were used, leading to possible over- or underestimation of minor importance for the total costs. There is also a risk of overestimation of the role of indirect costs using the human capital method and denying the possible friction in the labour market. Finally, as is always the case with cost of illness studies, it is not possible to state reasons for cost changes or correlation between changes. For example, costs and effects may not be affected only by the introduction of biological drugs but also by the intensification of treatment with traditional DMARDs in the 1990s.
To conclude, there is a 32% increase in the total fixed cost of RA between 1990 and 2010, inflation-adjusted with the CPI. This suggests that decreased hospitalisation and indirect costs have not fallen enough to offset the increasing cost of drug treatment. The benefit of biological treatment should be weighed against the considerable costs of biological drugs compared to traditional DMARDs.
Key Messages:
• The total fixed cost of RA in Sweden increased by approximately a third between 1990 and 2010.
• The increase was mainly due to increased direct costs after the introduction of biological drugs.
•
16
The authors declare no conflict of interest.
Funding
17
REFERENCES
1. Kvien TK. Epidemiology and burden of illness of rheumatoid arthritis. PharmacoEconomics 2004;22(2):1–12.
2. Hallert E, Husberg M, Jonsson D, Skogh T. Rheumatoid arthritis is already expensive during the first year of the disease (the Swedish TIRA Project). Rheumatology 2004;43:1374–82.
3. Kavanaugh A. Economic consequences of established rheumatoid arthritis and its treatment. Best Pract Res Clin Rheumatol 2007;21:929–42.
4. Franke LC, Ament AJ, van de Laar MA, Boonen A, Severens JL. Cost-of-illness of rheumatoid arthritis and ankylosing spondylitis. Clin Exp Rheumatol 2009;27:118– 23.
5. Kobelt G, Eberhardt K, Jönsson L, Jönsson B. Economic consequences of the progression of rheumatoid arthritis in Sweden. Arthritis Rheum1999;42:347–56. 6. Merkesdal S, Ruof J, Schffski O, et al. Indirect medical costs in early rheumatoid
arthritis: composition of and changes in indirect costs within the first three years of disease. Arthritis Rheum 2001;44:528–34.
7. Sokka T. Work disability in early rheumatoid arthritis. Clin Exp Rheumatol 2003;21:S71–4.
8. Klareskog L, van der Heijde D, de Jager JP, et al. Therapeutic effect of the combination of etanercept and methotrexate compared with each treatment alone in patients with rheumatoid arthritis: double-blind randomised controlled trial. Lancet 2004;363:675–81.
9. Chen YF, Jobanputra P, Barton P, et al. A systematic review of the effectiveness of adalimumab, etanercept and infliximab for the treatment of rheumatoid arthritis in adults and an economic evaluation of their cost-effectiveness. Health Technol Assess 2006;10(42): iii–138.
10. Nam, JL, Winthrop KL, van Vollenhoven, et al. Current evidence for the management of rheumatoid arthritis with biological disease-modifying antirheumatic drugs: a systematic literature review informing the EULAR recommendations for the management of RA. Ann Rheum Dis 2010;69:976–86.
11. Yelin E, Trupin L, Katz P, Lubeck D, Rush S, Wanke L. Association between etanercept use and employment outcomes among patients with rheumatoid arthritis. Arthritis Rheum 2003;48:3046–54.
12. Smolen JS, Han C, van der Heijde D, et al. Infliximab treatment maintains employability in patients with early rheumatoid arthritis. Arthritis Rheum 2006;54:716–22.
13. Kimel M, Cifaldi M, Chen N, Revicki D. Adalimumab plus methotrexate improved SF-36 scores and reduced the effect of rheumatoid arthritis (RA) on work activity for patients with early RA. J Rheumatol 2008;35:206–15.
14. Kavanaugh A, Smolen JS, Emery P, et al. Effect of certolizumab pegol with methotrexate on home and work place productivity and social activities in patients with active rheumatoid arthritis. Arthritis Rheum 2009;61:1592–600.
15. Wolfe F, Michaud K. The loss of health status in rheumatoid arthritis and the effect of biologic therapy: a longitudinal observational study. Arthritis Res Ther 2010;12:R35.
18
16. van Vollenhoven RF, Geborek P, Forslind K, et al. Conventional combination treatment versus biological treatment in methotrexate-refractory early rheumatoid arthritis: 2 year follow-up of the randomised, non-blinded, parallel-group Swefot trial. Lancet 2012;379: 1712–20.
17. Laas K, Peltomaa R, Kautiainen H, Puolakka K, Leirisalo-Repo M. Pharmacoeconomic study of patients with chronic inflammatory joint disease before and during infliximab treatment. Ann Rheum Dis 2006;65:924–8.
18. Wolfe F, Allaire S, Michaud K. The prevalence and incidence of work disability in rheumatoid arthritis, and the effect of anti-tumor necrosis factor on work disability. J Rheumatol 2007;34:2211–7.
19. Allaire S, Wolfe F, Niu J, Zhang Y, Zhang B, LaValley M. Evaluation of the effect of anti-tumor necrosis factor agent use on rheumatoid arthritis work disability: the jury is still out. Arthritis Rheum 2008;59:1082–9.
20. Sokka T, Kautiainen H, Pincus T, et al. Work disability remains a major problem in rheumatoid arthritis in the 2000s: data from 32 countries in the QUEST-RA Study. Arthritis Res Ther 2010;12:R42.
21. Olofsson T, Englund M, Saxne T, et al. Decrease in sick leave among patients with rheumatoid arthritis in the first 12 months after start of treatment with tumour necrosis factor antagonists: a population-based controlled cohort study. Ann Rheum Dis 2010;69:2131–6.
22. Hallert E, Husberg M, Bernfort L. The incidence of permanent work disability in patients with rheumatoid arthritis in Sweden 1990–2010: before and after introduction of biologic agents. Rheumatology 2012;51:338–46.
23. Kirchhoff T, Ruof J, Mittendorf T, et al. Cost of illness in rheumatoid arthritis in Germany in 1997–98 and 2002: cost drivers and cost savings. Rheumatology 2011;50:756–61.
24. Birnbaum HG. Changes in utilization and costs for patients with rheumatoid arthritis, 1997 to 2006. PharmacoEconomics 2012;30(4):323–36.
25. Hallert E, Husberg M, Skogh T. Costs and course of disease and function in early rheumatoid arthritis: a 3-year follow-up (the Swedish TIRA project). Rheumatology 2006;45:325–31.
26. Jacobsson LTH, Lindroth Y, Marsal L, Juran E, Bergström U, Kobelt G. Rheumatoid arthritis: what does it cost and what factors are driving those costs? Results of a survey in a community-derived population in Malmo, Sweden. Scand J Rheumatol 2007;36:179–83.
27. Kobelt G, Lindgren P, Geborek P. Costs and outcomes for patients with rheumatoid arthritis treated with biological drugs in Sweden: a model based on registry data. Scand J Rheumatol. 2009;38:409–18.
28. Neovius M, Simard JF, Klareskog L, Askling J. Sick leave and disability pension before and after initiation of antirheumatic therapies in clinical practice. Ann Rheum Dis 2011;70:1407–14.
29. Jonsson D, Husberg M. Samhällsekonomiska kostnader för reumatoid artrit och fibromyalgi. CMT Rapport 2000:4.
30. Lundkvist J, Kastang F, Kobelt G. The burden of rheumatoid arthritis and access to treatment: health burden and costs. Eur J Health Econ 2008;8(2):49–60.
19
31. SRQ. (Swedish Rheumatology Quality Register, Swerre) Rapport 2010/11:8. http://www.swerre.se/documents/SRQ%20Årsrapport%202010-2011.pdf.
32. Jönsson B, Kobelt G, Smolen J. The burden of rheumatoid arthritis and access to treatment: uptake of new therapies. Eur J Health Econ 2008;8 Suppl 2:61–86.
33. TLV. Läkemedelsförmånsnämndens allmänna råd om ekonomiska utvärderingar. LFNAR 2003: 2.
34. Luce BR, Manning WG, Siegel JE, Lipscomb J. Estimating Costs in Cost-Effectiveness Analysis. In: Gold MR, Siegel JE, Russel LB, Weinstein MC, ed. Cost Effectiveness in Health and Medicine. New York: Oxford University Press 1996: 176-213.
35. Landstingsförbundet, Kostnader per intagen patient, vårddag, läkarbesök, mm 1991, 1993, 1996. Landstingsförbundet, 1993, 1995, 1999.
36. Landstingsförbundet. Kostnad per vårdtillfälle, vårddag och läkarbesök på sjukhus 2001. Landstingsförbundet, 2003.
37. Sveriges Kommuner och Landsting. Kostnader och produktivitet på sjukhuskliniker 2008. SKL 2009.
38. Sveriges Kommuner och Landsting. Jämförelse av kostnader och prestationer på sjukhuskliniker 2009, 2010. SKL 2010, 2011.
39. Neovius M. Nationwide prevalence of rheumatoid arthritis and penetration of disease-modifying drugs in Sweden. Ann Rheum Dis 2011;70:624–9.
40. Socialstyrelsen. Hälso- och sjukvårdsstatistisk årsbok, 1996, 2002, 2009.
41. Apoteket AB. The Diagnosis and Therapy Survey. Ordered data through personal communication, 2011.
42. Blomqvist P, Feltelius N, Ekbom A, Klareskog L. Rheumatoid arthritis in Sweden: drug prescriptions, costs, and adverse drug reactions. J Rheumatol. 2000;27(5):1171–7.
43. Apotekens Service. Ordered data through personal communication, 2011.
44. Neovius M, Sundström A, Simard JF, et al. Small-area variations in sales of TNF inhibitors in Sweden between 2000 and 2009. Scand J Rheumatol 2011;40:8–15. 45. Personal communication, Per Eriksson, Linköping, Sweden.
46. Riksförsäkringsverket. Sjukfrånvarons diagnoser och avslutningsanledningar 1990. RFV redovisar 1996:3. Stockholm, 1996.
47. Riksförsäkringsverket. Vad kostar olika sjukdomar? Sjukpenningkostnaderna fördelade efter sjukskrivningsdiagnos. RFV Redovisar 2002:2.
48. Försäkringskassan. Ordered data through personal communication, 2011.
49. Riksförsäkringsverket. Sjukdomar som orsakat pensionering bland förtidspensionärerna i december 1991. RFV Statistisk rapport IS-R 1995:2.
50. Riksförsäkringsverket. Sjukdomar som orsakat pensionering bland förtidspensionärerna i december 1996. RFV redovisar 1998:3.
51. Riksförsäkringsverket. Sjukdomar som orsakat pensionering. Förtidspensionärerna i december 2002 i jämförelse med december 1991. RFV Statistikinformation Is-I 2004:3, bilaga.
52. Huscher D, Merkesdal S, Thiele K, et al. Cost of illness in rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis and systemic lupus erythematosus in Germany. Ann Rheum Dis 2006;65(9):1175–83.
20
53. Rosery H, Bergemann R, Maxion-Bergemann S. International variation in resource utilisation and treatment costs for rheumatoid arthritis: A systematic literature review. PharmacoEconomics 2005;23(3):243–57.
54. Hagel S, Petersson IF, Bremander A, Lindqvist E, Bergknut C, Englund M. Trends in the first decade of 21st century healthcare utilisation in a rheumatoid arthritis cohort compared with the general population. Ann Rheum Dis published on Dec 8, 2012. doi: 10.1136/annrheumdis-2012-202571.
55. Hogstedt C, Bjurvald M, Marklund S, Palmer E, Theorell T. Den höga sjukfrånvaron- sanning och konsekvens. Statens Folkhälsoinstitut 2004.
56. Kvalvik AG, Larsen S, Aadland HA, Hoyeraal HM. Changing structure and resources in a rheumatology combined unit during 1977-1999. Scand J Rheumatol 2007; 36(2):125-35.
21
TABLES AND FIGURES
Table 1 The included cost domains, respective data sources and years covered. Cost domain Years with data Data source
Direct costs
Inpatient care 1990-2010 Inpatient care register,
The National Board of Health and Welfare
Outpatient care 2000-2010 Regional databases,
Östergötland County Council, Region Västra Götaland and Region Skåne
Drugs 1991, 1997, 2001
2000-2010 Apoteket AB for drugs excluding biological. Apotekens Service for biological drugs. Indirect costs
Sick leave 1991, 2001
2005-2010 Swedish National Social Insurance Board Swedish Social Insurance Agency
Disability pension Disability pension, newly granted 1991, 1996, 2002 2003-2010 1990-2010
Swedish National Social Insurance Board Swedish Social Insurance Agency
22
Table 2 Inpatient and outpatient care consumption due to RA 1990–2010. Number of patients,
admissions and hospital days and number of visits to physician. Annual change (%) in brackets
Year Inpatient care Outpatient care
Patients Admissions Hospital days Physician visits
1990 7972 12 261 322 207 90 000 1991 8058 (+1) 12 267 (0) 400 797 (+24) 90 000 (0) 1992 7703 (-4) 12 063 (-2) 197 925 (-51) 90 000 (0) 1993 7835 (2) 12 600 (+4) 172 697 (-13) 90 000 (0) 1994 7455 (-5) 11 257 (-11) 136 752 (-21) 90 000 (0) 1995 6768 (-9) 10 095 (-10) 114 876 (-16) 90 000 (0) 1996 6530 (-4) 9618 (-5) 102 783 (-11) 90 000 (0) 1997 5769 (-12) 8693 (-10) 88 336 (-14) 90 000 (0) 1998 5236 (-9) 7578 (-13) 75 548 (-14) 90 000 (0) 1999 4692 (-10) 6867 (-9) 65 309 (-14) 90 000 (0) 2000 4368 (-7) 6526 (-5) 57 064 (-13) 88 251 (-2) 2001 4181 (-4) 6153 (-6) 53 129 (-7) 88 139 (0) 2002 3999 (-4) 5809 (-6) 45 997 (-13) 92 430 (+5) 2003 3852 (-4) 5167 (-11) 47 578 (3) 92 672 (0) 2004 3710 (-4) 4980 (-4) 44 601 (-6) 87 096 (-6) 2005 3562 (-4) 4687 (-6) 38 440 (-14) 84 433 (-3) 2006 3432 (-4) 4559 (-3) 35 048 (-9) 83 893 (-1) 2007 3280 (-4) 4486 (-2) 32 497 (-7) 87 856 (+5) 2008 3118 (-5) 4273 (-5) 28 684 (-12) 89 036 (+1) 2009 2539 (-19) 3178 (-26) 20 866 (-27) 92 402 (+4) 2010 2786 (+10) 3424 (+8) 23 362 (+12) 88 091 (-5)
23
Table 3 Disability pensions and sick leave due to RA 1990–2010. Annual change (%) in brackets.
Year Disability pensions, DPs Sick leave
Prevalence Newly granted Days
1990 10 033 957 1 022 600 1991 10 033 (0) 945 (-1) 1 001 748 (-2) 1992 11 279 (+12) 1117 (+18) 1 065 351 (+6) 1993 11 619 (+3) 1348 (+21) 1 119 623 (+5) 1994 11 570 (0) 1093 (-19) 1 003 818 (-10) 1995 11 237 (-3) 863 (-21) 940 828 (-6) 1996 10 572 (-6) 864 (0) 842 220 (-10) 1997 10 545 (0) 814 (-6) 769 621 (-9) 1998 10 029 (-5) 558 (-31) 764 212 (-1) 1999 9621 (-4) 601 (+8) 907 049 (+19) 2000 9422 (-2) 750 (+25) 868 994 (-4) 2001 9308 (-1) 759 (+1) 984 560 (+13) 2002 9209 (-1) 827 (+9) 1 049 834 (+7) 2003 8840 (-4) 799 (-3) 1 033 445 (-2) 2004 8981 (+2) 923 (+16) 1 034 691 (0) 2005 9048 (+1) 729 (-21) 881 993 (-15) 2006 8917 (-1) 559 (-23) 742 542 (-16) 2007 8668 (-3) 569 (+2) 668 499 (-10) 2008 8297 (-4) 425 (-25) 635 619 (-5) 2009 7693 (-7) 233 (-45) 583 773 (-8) 2010 6945 (-10) 143 (-39) 527 106 (-10)
24
Figure 1
Total annual cost for RA, € million 1990–2010. Costs A) in current prices, B) adjusted to 2010 prices with CPI and C) adjusted to 2010 prices with the healthcare price index.
0 m€ 100 m€ 200 m€ 300 m€ 400 m€ 500 m€ 600 m€ 700 m€ 19 90 19 92 19 94 19 96 19 98 20 00 20 02 20 04 20 06 20 08 20 10 A) Current prices 19 90 19 92 19 94 19 96 19 98 20 00 20 02 20 04 20 06 20 08 20 10
B) Adjusted to 2010 prices, CPI
0 m€ 100 m€ 200 m€ 300 m€ 400 m€ 500 m€ 600 m€ 700 m€ 800 m€ 19 90 19 92 19 94 19 96 19 98 20 00 20 02 20 04 20 06 20 08 20 10
C) Adjusted to 2010 prices, Health care Price Index
Biological drugs Drugs, excl bio Outpatient Inpatient Sick leave
25
Figure 2 The proportion of total costs for RA in 2010 prices represented by each cost type.
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 199 0 199 1 199 2 199 3 199 4 199 5 199 6 199 7 199 8 199 9 200 0 200 1 200 2 200 3 200 4 200 5 200 6 200 7 200 8 200 9 201 0 Disability pension Sick leave Inpatient Outpatient Drugs, excl bio Biological drugs
study in Sweden
Unit costs used for calculating total costs are depicted in table 4.
Table 4 Costs of inpatient and outpatient care, annual income, income per calendar day and annual cost of drugs. Current prices €.
€/Inpatient
day €/Outpatient visit* Average annual income, ages 55-65** Average income per calendar day, all ages*** Annual cost of drugs, million
1990 NA 177 17613 54 8 1991 NA 188 19914 54 8 1992 265 199 20829 55 8 1993 278 211 21866 57 9 1994 290 223 22700 58 10 1995 302 235 23224 59 10 1996 314 247 24330 61 11 1997 346 256 25344 63 12 1998 377 265 26326 65 11 1999 409 274 27622 68 12 2000 440 283 28756 72 27 2001 472 275 30085 76 35 2002 506 299 31503 79 40 2003 541 324 33168 81 54 2004 575 327 34372 84 78 2005 609 298 35478 86 96 2006 644 312 36463 89 115 2007 678 341 37689 92 136 2008 713 368 39095 94 156 2009 713 384 40059 97 177 2010 660 401 41954 99 198
* Visit to physician at hospital. The total costs of outpatient care include costs for primary care which constituted less than 1% of total costs for RA.
** Including vacation and social fees of 40%. Used for calculating the costs of DP. Weighted average to adjust for the larger amount of women in the RA-population.
*** Including vacation and social fees of 40%. Used for calculating the costs of sick leave. Weighted average to adjust for the larger amount of women in the RA-population.
study in Sweden
A sensitivity analysis was performed changing the assumptions for the cost domains that lack data for many years; outpatient care, drugs excluding biological and sick leave. Table 5 depicts the sensitivity analysis. At most, total costs are increased by 9% which occurs when sick leave costs are augmented by 50% (Sens 6).
Table 5 Sensitivity analysis. Total costs are shown in 2010 prices € after changing assumptions (Sens 1-6) for outpatient care, drugs and sick leave that lack data for some of the years. Change compared to baseline in brackets (%).
Outpatient care Drugs Sick leave
Baseline Sens 1 Sens 2 Sens 3 Sens 4 Sens 5 Sens 6
1990 454 443 (-3) 468 (3) 454 460 (1) 454 (0) 495 (9) 1991 445 435 (-2) 458 (3) 445 450 (1) 446 (0) 481 (8) 1992 489 480 (-2) 503 (3) 489 495 (1) 486 (-1) 528 (8) 1993 494 486 (-2) 508 (3) 494 500 (1) 486 (-2) 534 (8) 1994 480 472 (-2) 494 (3) 480 486 (1) 480 (0) 516 (7) 1995 459 453 (-2) 474 (3) 459 466 (1) 464 (1) 493 (7) 1996 447 441 (-1) 462 (3) 447 453 (1) 459 (3) 477 (7) 1997 453 448 (-1) 468 (3) 453 459 (2) 470 (4) 481 (6) 1998 450 446 (-1) 466 (4) 450 457 (2) 469 (4) 480 (7) 1999 463 461 (-1) 480 (4) 463 470 (2) 471 (2) 500 (8) 2000 482 482 498 (3) 482 489 (1) 492 (2) 518 (7) 2001 502 502 518 (3) 502 509 (1) 502 (0) 544 (8) 2002 518 518 535 (3) 519 (0) 525 (1) 510 (-2) 564 (9) 2003 535 535 553 (3) 537 (0) 542 (1) 526 (-2) 581 (9) 2004 577 577 595 (3) 581 (1) 584 (1) 566 (-2) 624 (8) 2005 591 591 607 (3) 596 (1) 598 (1) 591 632 (7) 2006 597 597 614 (3) 603 (1) 604 (1) 597 632 (6) 2007 607 607 625 (3) 614 (1) 613 (1) 607 639 (5) 2008 604 604 623 (3) 612 (1) 611 (1) 604 634 (5) 2009 605 605 626 (3) 615 (2) 612 (1) 605 634 (5) 2010 598 598 618 (3) 608 (2) 605 (1) 598 624 (4)
Sens 1: 45000 outpatient visits in 1990, linear trend is assumed to 90000 in 2000. Sens 2: 50% more outpatient visits throughout the period.
Sens 3: Drugs excl. biological doubled in 2010, a linear trend is assumed between 2001 and 2010. Sens 4: 50% increased drug costs, excluding biological, throughout the period.
Sens 5: A linear trend is assumed for years missing sick leave data. Sens 6: 50% increased sick leave costs throughout the period.
