Trends in cost of traditional and biologic psoriasis therapies from 2000 to 2008. A, Traditional therapies. Methoxsalen had the largest percentage increase in drug cost (315.7%), whereas acitretin increased in price by 157.5%. Costs for methotrexate and cyclosporine did not increase. B, Biologic therapies. Efalizumab increased in cost by 35.1% during a 4-year interval, whereas the average wholesale price (AWP) of adalimumab increased by 27.2% from 2003 until 2008. Etanercept increased steadily in cost by 46.8% from 2000 to 2008. *Years when corresponding medications were not yet available. †This drug was withdrawn from the US market in April 2009. However, since this study is an analysis of psoriasis drug costs from 2000 to 2008, data with regard to efalizumab were included for the sake of comparison to other therapies.
Biannual percentage changes in average wholesale price for systemic psoriasis therapies compared with Consumer Price Index–Urban (CPI-U) values. A, Methoxsalen increased in price by 184.6% during a 2-year interval. *Intervals when the cost did not change (percentage change equals zero). B, Increasing CPI values are often outpaced by increasing costs of adalimumab, efalizumab, etanercept, and infliximab. Asterisks correspond to intervals when the medication was not yet available (2000-2004 for adalimumab, efalizumab, and alefacept) or when the cost did not change (2004-2008 for alefacept). †This drug was withdrawn from the US market in April 2009. However, since this study is an analysis of psoriasis drug costs from 2000 to 2008, data with regard to efalizumab were included for the sake of comparison to other therapies.
Beyer V, Wolverton SE. Recent Trends in Systemic Psoriasis Treatment Costs. Arch Dermatol. 2010;146(1):46-54. doi:10.1001/archdermatol.2009.319
Copyright 2010 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2010
To analyze the current total cost of systemic therapy for psoriasis and to compare annual trends in the cost of both generic and brand-name therapies with trends in the Consumer Price Index–Urban since 2000.
A cost model was developed that includes costs for prescription drugs, office visits, and suggested laboratory tests and monitoring procedures. Annual trends in psoriasis drug costs from 2000 through 2008 were analyzed by calculating the percentage change in the average wholesale price from the previous year; these values were compared with changes in the yearly Consumer Price Index–Urban values.
The United States.
Main Outcome Measures
Total annual costs for systemic psoriasis therapies and trends in cost compared with the trends in Consumer Price Index–Urban values (equivalent to inflation).
Current total annual costs for systemic psoriasis therapies ranged from $1197 (methotrexate) to $27 577 (alefacept, two 12-week courses). Trends in the average wholesale price of brand-name psoriasis therapies from 2000 through 2008 demonstrate an average increase of 66% (range, −24% to +316%); thus, costs of several brand-name psoriasis drugs greatly outpaced the rates of inflation for all items and all prescription drugs.
Despite the higher monitoring costs associated with traditional systemic therapies, annual costs of biologics exceed those of other available therapies. Current trends demonstrate that systemic psoriasis therapy costs are increasing at a much higher rate compared with general inflation.
Psoriasis is a chronic autoimmune disease that affects 1% to 3% of the general population and involves an estimated 4.5 million to 7.5 million Americans.1 The severity of psoriasis varies significantly, with disease ranging from scattered papules to generalized scaly plaques.2 The effect of the disease on the quality of life or physical and emotional well-being of a patient differs for each patient and often relates to the extent and location of the lesions.2,3 Therapeutic options for psoriasis are also varied: they consist of topical agents, phototherapy, and systemic therapies, such as retinoids, cyclosporine, methotrexate, and the 5 biologic agents that are currently approved by the Food and Drug Administration (FDA) for psoriasis and/or psoriatic arthritis. Published consensus guidelines recommend topical therapies for patients with mild, localized disease that does not interfere with activities of daily living, whereas phototherapy and systemic therapies are used for more extensive disease.4
Up to one-third of Americans with psoriasis have moderate to severe disease that cannot be controlled with topical treatments alone.5 Despite the increased efficacy of systemic therapies relative to topical treatments in moderate to severe disease, these therapies appear to be underused. Studies6 have demonstrated that only 43% of patients with severe psoriasis are receiving systemic therapy of any kind. Reluctance by physicians to prescribe systemic therapies or by patients to adhere to systemic treatment regimens may be owing to several factors, such as intolerance of treatment, adverse effects, patient affordability, and fear of potential adverse effects.7,8 The issue of affordability is especially relevant, with changing insurance plans, rising copayments, and current trends in prescription drug prices increasing the cost burden of psoriasis. Newer, more expensive biologic therapies have increased the awareness of the cost of psoriasis therapy.9 Despite their significant impact on disease control and quality of life in patients with psoriasis, the high cost of biologic therapies relative to more traditional systemic therapies requires careful decision making when choosing among the therapeutic options discussed in this article.
Analysis of treatment cost is especially important with regard to a chronic disease such as psoriasis, which often requires lifelong management. The heavy economic burden of psoriasis has been estimated to exceed $3 billion to the health care industry annually.10 Previous estimates of the direct cost of psoriasis, including drugs and hospitalizations, have demonstrated total costs of $649.6 million for 1.4 million patients with psoriasis in 1997.11 However, recent approval of costly biologic therapies and changing health care costs call for updated cost analyses. In addition, awareness of current trends in prescription drug costs, which demonstrate that increasing prescription drug costs have outpaced Consumer Price Index–Urban (CPI-U) rates, which are generally considered synonymous with the rate of inflation,12 is also essential when making therapeutic decisions. We hypothesized that this current trend also applies to the cost of prescription medications for the treatment of psoriasis. This study compares the current cost of treatment among different systemic psoriasis therapies and compares recent trends in psoriasis drug costs to trends in general CPI-U rates and all prescription drug costs. Findings from this study will help physicians and employers understand the current costs and trends associated with psoriasis treatment.
A cost model based on continuous, year-long treatment was developed for each therapy, which includes costs of medications, office visits, laboratory tests, and monitoring procedures. The cost model was based on clinical experience and published and manufacturer's guidelines.13- 15 Specifically, costs for alefacept (Amevive; Biogen Idec, Cambridge, Massachusetts) were based on two 12-week treatment courses per year or one 16-week treatment course.16 Despite its withdrawal from the US market in 2009, we included data with regard to efalizumab for the sake of comparison with other available therapies in the analyzed period. Costs for UV-B and psoralen–UV-A (PUVA) therapy were based on a total of 42 treatments per year, with 12 weeks of thrice-weekly induction treatments and maintenance therapy of 2 treatments per month for 9 months per year. Costs for the first year of treatment with adalimumab include a 80-mg loading dose at week 1, followed by 40 mg at week 2, followed by 40 mg every other week. Estimates of cost for the first year of treatment for etanercept include a loading dose of 50 mg subcutaneously twice weekly for 12 weeks followed by 50 mg subcutaneously weekly. Costs for the first year of infliximab therapy include a dose of 5 mg/kg at weeks 0, 2, and 6 and then every 8 weeks. Because of the variability in guidelines for laboratory monitoring during treatment with biologic therapies, the cost model includes only those monitoring tests recommended by the FDA.
Costs for each therapy were assessed from the perspective of the third-party payer by use of the average wholesale price (AWP) of each drug.17 When calculating medication costs, we assumed a patient weight of 80 kg and, in the case of the biologics efalizumab and infliximab, assumed that vials of medication were fully used during treatment. Costs for infliximab were based on a 3-hour infusion time. Monitoring costs for methotrexate include a liver biopsy (Current Procedural Terminology [CPT]18 code 47000) to factor in the highest potential cost for this drug therapy (even though the liver biopsy would be performed less than yearly).
Costs of outpatient office visits, laboratory testing, infusions, and other suggested laboratory and related monitoring procedures were determined by means of the 2008 Medicare National Median Physician Reimbursement schedule and Clinical Laboratory Fee schedule (Table 1 and Table 2).19 Under the assumption that most patients with psoriasis are established patients, the costs of office visits were calculated in the same manner as level 3 return visits. We used CPT codes to search for costs of laboratory tests and procedures.20 All costs were calculated in US dollars. Additional direct costs, such as hospitalizations or costs associated with medication adverse effects, and indirect costs, such as time away from work, were not included in this analysis.
Annual trends in psoriasis drug costs from 2000 until 2008 were analyzed by calculating the percentage change in AWP from the previous year. In addition, total percentage change from 2000 to 2008 was calculated by means of the following formula: 100(AWP  − AWP )/AWP (2000). Changes in CPI-U rates for all items and for prescription drugs were determined by the use of CPI values for all urban consumers (US city average) because these values are generally considered equivalent to the rate of inflation.21 Therefore, any discussion of trends in inflation rates in this study is solely based on CPI-U values.
Our analyses were developed to answer 2 questions. First, what is the current, direct cost of systemic therapy for psoriasis? Second, what is the trend in these costs relative to general inflation?
To assess costs for psoriasis therapy, we used a cost model to compare the direct annual costs of phototherapy, systemic agents, and biologics (Tables 1 and 2). Results of our cost analysis are summarized in Table 3. The annual costs ranged from $1197 for methotrexate, 7.5 mg weekly, to $27 577 for two 12-week courses of alefacept. Phototherapy costs ranged from $3083 per year for UV-B therapy to $7288 for PUVA annually (including induction and maintenance costs). Costs for acitretin, 25 mg daily ($9163), were comparable to those for cyclosporine, 400 mg daily ($9999); however, some patients require 50 mg/d of acitretin, which translates to an annual cost of $17 613.
Annual costs of the biologics used for psoriasis therapy ranged from $18 384 to $27 577. The therapies that require loading doses (adalimumab, etanercept, and infliximab) were more costly during the first year of treatment compared with subsequent years ($23 538 vs $21 876 for adalimumab, $26 862 vs $21 876 for etanercept, and $23 639 vs $20 717 for infliximab). Overall, prices for the recommended dosing of the biologics were comparable. However, estimates of total costs of biologic regimens that have been studied in clinical trials22 but are not the current recommended regimen were found to vary considerably from our cost calculations. For example, patients treated with efalizumab, 2 mg/kg subcutaneously weekly, accumulate an annual cost $48 530, whereas treatment with etanercept, 50 mg, sustained at this dose twice weekly incurs an annual cost of $43 732.
We analyzed trends in AWP for the various brand-name and generic systemic psoriasis treatments from 2000 through 2008 (Table 4). The percentage changes in drug prices between 2000 and 2008 ranged from −24.1% for methotrexate to +316% for the brand-name version of methoxsalen (Oxsoralen-Ultra; Valeant Pharmaceuticals International, Aliso Viejo, California). Acitretin (Soriatane; Stiefel Laboratories Inc, Coral Gables, Florida) had the second largest increase (157.5%) during this period. Conversely, 1 brand-name version of cyclosporine (Neoral; Novartis Pharmaceuticals Corporation, Basel, Switzerland) did not increase in price during this 8-year interval, whereas another (Gengraf; Abbott Laboratories, Abbott Park, Illinois) decreased in price by 4.0%. When analyzing trends in drug costs for the biologics, the interval must be considered because a few of these therapies have only been available since 2003 or 2004. For example, efalizumab (Raptiva; Genentech Inc, South San Francisco, California) increased in cost by 35.1% during a 4-year interval. Similarly, the AWP of adalimumab (Humira; Abbott Laboratories) increased by 27.2% during a 5-year interval (from 2003 to 2008). Etanercept (ENBREL; Immunex Corporation, Thousand Oaks, California) increased in cost by 46.8% from 2000 to 2008, whereas infliximab increased by only 14.0%, and the cost of alefacept has not changed during the same interval. Annual percentage change in drug prices fluctuate widely (Table 4, Figure 1, and Figure 2); for example, a significant increase in the cost of the brand-name version of methoxsalen (Oxsoralen-Ultra) was observed between 2004 and 2006 (184.6%). Therapies such as acitretin, adalimumab, efalizumab, and etanercept have increased in cost steadily every year, whereas methotrexate, cyclosporine, and alefacept have seen minor, if any, increases in price since 2000.
To test our hypothesis that psoriasis treatment costs have been increasing at a rate greater than the CPI-U, trends in psoriasis treatment costs from 2000 through 2008 were compared with the CPI-U21 for all items and all prescription drugs during the same period. There was an increase of 25.8% for all items and a 30.1% increase in prescription drug costs (Table 5). The increase in CPI-U values, which is considered equivalent to general inflation, was greatest between 2007 and 2008 (4.2%), whereas prescription drug prices increased by a significant 6.0% between 2000 and 2001. Costs for all psoriasis medications except methotrexate, cyclosporine, alefacept, and infliximab have increased at a substantially greater rate than the CPI-U rate for all items and prescription drugs (Tables 4 and 5, and Figure 2A and B).
The results of our cost analysis, summarized in Table 3, revealed that the most costly of the currently recommended treatment regimens is alefacept (two 12-week treatments) at $27 577 per year, whereas the least costly treatment is methotrexate ($1197-$1393 annually, depending on the dose). Published cost-effectiveness analyses have demonstrated annual costs for psoriasis medication of up to $37 000; however, these analyses included costs for treatment regimens that are often prescribed but not currently published as recommended regimens, such as adalimumab, 40 mg weekly, or efalizumab, 2 mg/kg subcutaneously weekly.9,16 Indeed, using our cost model, we calculated an annual cost of $48 530 for efalizumab, 2 mg/kg subcutaneously weekly.
Our analysis demonstrated a greater annual cost for PUVA therapy ($7288) relative to UV-B therapy ($3083); this is in part owing to the increasing cost of methoxsalen therapy, as well as additional costs associated with required monitoring during PUVA therapy. Thus, our cost estimates for PUVA therapy are substantially greater than those previously published.23 The results of our cost analysis are similar to previously published cost analyses; discrepancies may be accounted for by the use of different monitoring guidelines and increasing costs of medications, procedures, or clinical laboratory fees. Although various studies24,25 have recommended laboratory screening and monitoring tests for patients treated with biologic therapies, our cost model included only those guidelines recommended by the FDA. Including additional screening tests would have caused a nominal increase in the annual cost of biologic therapies. Our analysis did not include indirect costs, such as time away from work, direct costs such as hospitalizations, or costs related to adverse effects. In addition, costs were calculated by means of the 2008 Medicare National Median Physician Reimbursement schedule and Clinical Laboratory Fee schedule; retail costs for patients who do not qualify for Medicare may be higher. However, because our cost model analyzes annual costs for therapies, many of which are not given for a full year, the model may overestimate costs of certain drug regimens.
Of note, annual costs were assessed by means of the AWP for each therapy, which may not accurately reflect market prices for medications. The AWP is a reference price reported in publications such as the Red Book17 and used by Medicaid programs and most third-party payers when deciding reimbursement formulas.26 The AWP is derived from manufacturer and supplier data; the final price negotiated by intermediaries and obtained by hospitals, pharmacies, insurance companies, and physicians is often significantly lower than the AWP.27 A report28 by the Office of the Inspector General of the Department of Health and Human Services stated that the average sales price of more than 2000 medications, based on actual sales transactions, is lower than the AWP by a median percentage of 49%. Although the AWP may therefore overestimate the true cost of therapy to the patient, it is an important drug-pricing benchmark and helps monitor trends in drug costs.
The importance of calculating all disease-related costs in a pharmacoeconomic analysis must be emphasized because limiting an analysis to a comparison of drug acquisition costs would significantly underestimate the total cost of psoriasis therapy.29 Additional costs include those of laboratory monitoring, office visits, procedures such as a liver biopsy, and costs related to complications of emergency department visits and hospitalizations.11,30 For example, despite the low cost of methotrexate relative to other psoriasis therapies, frequent office visits and laboratory tests, as well as recommended periodic liver biopsies, increase the annual cost somewhat.31 A cost-minimization analysis that compared methotrexate and cyclosporine showed that although the direct costs of cyclosporine were higher, largely because of higher drug acquisition prices, higher indirect and follow-up costs were observed in the methotrexate group.31
Although not included in this analysis, indirect costs should also be considered when making therapeutic decisions. Also referred to as “forgotten costs” or “lost opportunity costs,” indirect costs include time away from work, transportation costs, and time spent in personal care.29 The decreasing use of phototherapy, despite its low direct cost and favorable safety profile, has been partially attributed to the lost work productivity and inconvenience associated with treatment.5 Simpson and colleagues5 propose that decreasing physician reimbursement rates and high out-of-pocket expenses for patients, owing to copayments with every treatment, place disincentives on phototherapy use. Given the inconvenience and costs associated with office-based phototherapy, home UV phototherapy (not included in this cost analysis) has become a more attractive option in certain settings because it helps defray indirect and direct costs. Alternatively, patients who are discouraged with the logistics and frequent copayments associated with office-based phototherapy may be encouraged to use more expensive biologics, the cost of which is generally mostly covered by insurance companies.
The contribution of insurance companies to prescription drug cost, however, does not necessarily translate to lower cost for the patient. A report by the Kaiser Family Foundation32 showed that, among the 75% of workers covered by a 3-tier or 4-tier cost-sharing plan for prescription drugs, average copayments in 2007 ranged from $11 per generic drug to $71 for fourth-tier prescription drugs. Several insurance plans have recently added a fourth tier for expensive lifestyle drugs and biologics, and plans such as BlueCross BlueShield have shifted away from fixed copayments for the fourth-tier prescriptions; they require patients to pay coinsurance of up to 40% on higher-tier drugs.33 Workers covered by 3-tier or 4-tier cost-sharing plans had an average coinsurance of 36% in 2007 for fourth-tier drugs compared with 21% for generic drugs.29 Although these tiered and high-deductible plans may help slow increases in drug spending and encourage efficient use of health care resources,34 this may be happening at the expense of patients who receive the most appropriate treatment. For example, a patient with a coinsurance of 25% will pay $6716 or $5885 for the first year of treatment with etanercept or adalimumab, respectively; both treatments will cost $5470 annually thereafter. This cost may be prohibitive when we consider that those who need prescriptions the most often can afford it the least: according to a report by the Kaiser Family Foundation35 in March 2008, lower-income patients, often elderly and in poor health, use more prescription drugs than their wealthier counterparts. More specifically, 32% of patients who take 4 or more prescriptions earn an annual income of $25 000 or less.35 These facts are especially relevant with regard to a disease such as psoriasis because studies have suggested that patients with lower family incomes are more likely to have severe psoriasis and are even less likely to be able to afford the most efficacious treatments.36
We analyzed recent trends in costs of systemic psoriasis therapies and compared these trends to general trends in CPI-U rates. Current trends in prescription drug prices show that increasing costs of prescription drugs are outpacing the overall increase in CPI-U rates.12 Our results validate our hypothesis that this trend also applies to therapies for psoriasis and is not limited to the newer, more expensive therapies; indeed, the greatest increase in costs were calculated for brand-name versions of methoxsalen and acitretin. Annual trends in psoriasis drug prices vary widely, with brand-name psoriasis drugs increasing in cost by an average of 66.0% between 2000 and 2008 (range, −24.1% to +315.7%). During the same period, the total CPI-U values of all items and all prescription drugs increased by 25.8% and 30.1%, respectively. Although the tendency of psoriasis drug costs to outpace the CPI-U rate is shared by both traditional and biologic therapies, traditional therapies remain much more affordable than biologic therapies. When considering the expense of biologic therapies, even relatively small increases in price become significant, such as the 14.0% increase during an 8-year period seen with infliximab.
Several factors contribute to the increasing cost of prescription medications. According to a study by the Tufts Center for the Study of Drug Development,37 research and development costs are increasing: the average cost of developing a new prescription drug in 2003 was $897 million compared with $802 million in 2001. Additional contributing factors include increased spending on advertising, the efficacy of medications relative to others on the market, difficulty recruiting patients into clinical trials, and stringent drug safety regulations that result in longer, more costly clinical trials.37 In addition, the recent introduction of new therapies that have a complicated production process, such as the biologics for psoriasis, increases prescription drug spending. Physician awareness of pharmacoeconomic issues is especially important in our current economic climate, in which health care spending continues to increase disproportionately to the amount of funds available. In 2006, national health expenditures totaled $2.1 trillion, which corresponds to an increase of 6.7% from the previous year.38 Although this average annual increase in health care expenditures is expected to remain steady for the next 10 years, prescription drug spending is expected to increase by an average of 8.2% yearly, from $200.7 billion in 2005 to $497.5 billion in 2016.12 Accordingly, prescription drugs are expected to account for 12.0% of total health care spending in 2017, compared with 10.3% in 2006.
While they act in the best interest of the patient to limit disease morbidity, physicians must consider costs associated with treatment to ensure efficient health resource allocation.39 Factors such as efficacy and tolerability must also be considered. Cost-effectiveness analyses of systemic treatments for psoriasis, such as the biologic drugs, have recently been published.9,16,22,23 A study by Feldman and coworkers,23 which used a 75% improvement in Psoriasis Area and Severity Index (PASI-75) scores as a measure of treatment success, found that methotrexate is the most cost-effective treatment ($5400 per treatment success annually), whereas alefacept has the highest cost ($40 600) per treatment success. When they factored in considerations of safety, the authors suggested UV-B phototherapy as the first-line agent of choice for severe psoriasis because of the higher risk profile of methotrexate. Regarding the cost-effectiveness of biologic therapies, studies16,23 suggest that adalimumab and infliximab are the most cost-effective biologic therapies, whereas efalizumab and alefacept are the least cost-effective on the basis of the PASI-75 analyses. Similar results were obtained by Sizto and coworkers,40 who demonstrated that infliximab provided the greatest increase in quality-adjusted life-years (QALYs) relative to supportive care (0.18 QALY), followed by adalimumab (0.16 QALY). Although methotrexate (0.13 QALY) and cyclosporine (0.08 QALY) were less beneficial, they were found to be the most cost-effective treatments for moderate to severe psoriasis. Several studies of efficacy and tolerability of psoriasis therapies do not factor cost into their analyses. A study41 of 12 trials that compares PASI-75 reduction rates at 12 weeks of treatment revealed the following rates: infliximab, 80%; adalimumab (40 mg/wk), 80%; cyclosporine, 78.2% to 80.3%; PUVA, 63%; methotrexate, 60%; UV-B, 55%; acitretin, 52%; etanercept (50 mg/wk), 34%; efalizumab, 31.4%; and alefacept, 21%. Similarly, an analysis42 of 24 randomized controlled trials of systemic psoriasis treatments, including cyclosporine, methotrexate, fumaric acid esters, and biologic treatments, suggested that infliximab is the most efficacious agent (75.5%-87.9% of patients met the PASI-75 criterion after 10 weeks), followed by adalimumab (an average of 64% met the criteria after 16 weeks). The study indicated that these agents are superior in efficacy to cyclosporine, methotrexate (although long-term studies are lacking), etanercept, efalizumab, and fumaric acid esters. Analysis of short-term tolerability demonstrated that methotrexate had the highest rate of withdrawal (7.3% average monthly incidence rate) secondary to adverse effects, with gastrointestinal upset the most frequent symptom. Biologic treatments demonstrated similar overall adverse events and withdrawal rates (16.6%-28.3% and 0.3%-1.3%, respectively); however, the type of adverse event differed. Infusion reactions were seen more frequently with infliximab, whereas injection site reactions were more frequently associated with etanercept, and upper respiratory tract infections with adalimumab.42 Long-term studies and head to head comparisons of tolerability for the various systemic treatments available for psoriasis are lacking and remain an area for future study.
Although the economic analyses have limitations, overall we provide an updated cost analysis for moderate to severe psoriasis therapy and clearly demonstrate trends in psoriasis drug costs. Given the current psoriasis prescription drug market and the likely development of novel therapeutics for psoriasis in the near future, these trends are likely to continue. Thus, in addition to considerations of safety, efficacy, and patient convenience and preference, health care professionals should be aware of current costs and anticipate future costs when making therapeutic decisions.
Correspondence: Vivianne Beyer, MD, Department of Dermatology, Indiana University School of Medicine, 550 N University Blvd, Ste 3240, Indianapolis, IN 46202 (email@example.com).
Accepted for Publication: June 11, 2009.
Author Contributions: Both authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Beyer and Wolverton. Acquisition of data: Beyer and Wolverton. Analysis and interpretation of data: Beyer and Wolverton. Drafting of the manuscript: Beyer. Critical revision of the manuscript for important intellectual content: Wolverton.
Financial Disclosure: Dr Wolverton is a consultant for Eli Lilly regarding anti–interleukin-17 agent therapy for psoriasis. Dr Wolverton is also a consultant for the following Amgen advisory groups for etanercept safety: neurology, epidemiology, rheumatology, and dermatology. Dr Wolverton obtains royalties from Elsevier Publishing Company for Comprehensive Dermatologic Drug Therapy, 2nd ed. (Philadelphia, Pennsylvania: Sanders Elsevier; 2007).