The Future
of Prescription Drug Expenditures
HA 601
The increasing cost of medical care has been in the
U.S. financial policy spotlight for decades, and has become an acute concern
over the past 15-20 years. The rate of
growth of medical expenditures have consistently increased, consuming 8.8% of
U.S. GDP in 1980 ($246 billion), 15.3% of U.S. GDP in 20035 ($1.7
trillion), and at current rates is projected to be 18.7% of U.S. GDP in 201418. Uses of the medical budget include physician
visits, hospitalization, rehabilitation, and prescription medications.
The
fastest growing use of the U.S. medical dollar is prescription
medications. In 2003, prescription
medications accounted for 11% of U.S. medical expenditures ($179.2 billion)3. Outpacing the growth of total medical
expenditures (which itself outpaces inflation), increasing use and cost of
prescription drugs far outpace inflation (increasing on average 12.2% annually
from 1990-19991, 15% from 2001 to 20022, and 11% from
2002-20033), and are a major contributor to the nation’s spiraling
healthcare costs. By 2011, prescription
medication sales alone are projected to be 2.5% of the GDP. Such growth is unsustainable in the
long-term and crippling to the economy.
This essay will describe the primary drivers of
prescription drug spending and describe how components of the health care
system influence these drivers. The
three most significant drivers, in order of descending magnitude, are price,
utilization, and changing prescribing habits.
During the 10 year period from 1994 to 2004, prices
for all prescriptions increased 8.3% annually, a total price increase of 222%
for the period, more than triple the U.S. inflation rate3. The price of brand name prescription medications
increased at 4.3 times the inflation rate in 200316. In recent years, the pharmaceutical
industry’s ability to increase prices well above the inflation rate and
maintain high profit margins despite a weak economy have not won many favors
with the public. A 2005 survey by the
Kaiser Family Foundation reported 71% of respondents thought high profits by
drug insurance companies were very important factors in increasing costs of
health care, and 35% thought they were the most important7. The pharmaceutical industry was the most
profitable in the U.S. from 1995-2002, and third most profitable in 20043.
On the one hand, it’s easy to revile drug companies as
they enjoy steady profits while the rest of the economy may go boom or
bust. Attributing steep increases in
drug prices to unchecked corporate greed is a simplistic analysis of a complex
market. Marcia Angell, former editor of
the New England Journal of Medicine, accuses major U.S. drug companies
of spending 2.5 times more on marketing and administration (selling, general,
and administrative expenses, or “SG&A”) expenses than R&D in her book The
Truth About Drug Companies. Ralph
Nader’s consumer rights group Public Citizen wrote that in 2002, the top
10 drug companies spent 30.8% of revenues into SG&A and invested 14.1% of
revenues into R&D8 (SG&A / R&D ratio = 2.2). By themselves, these figures may seem
appalling to the casual observer.
However, such data can be misleading when presented out of context. In fiscal 2004, Microsoft’s SG&A expense
was 32.2% of revenues and R&D was 15.5% (SG&A / R&D ratio = 2.1),
3M’s SG&A was 21.7% and R&D was 5.7% (SG&A / R&D ratio = 3.8),
McDonald’s SG&A was 35.4% and investments were 7.2% (SG&A / R&D
ratio = 4.9), Exxon Mobil’s SG&A was 27.6% and investments were 5%
(SG&A / R&D ratio = 5.5)9.
The pharmaceutical industry’s so called “imbalance” between R&D and
SG&A suddenly appears very ordinary compared to other large companies. Also of note, marketing expenses include the
$15.9 billion of drug samples the industry provided to physicians in 20043.
Only the office, computing, and accounting machine
industry invests more in R&D as a percentage of revenue than the U.S.
pharmaceutical industry10.
The fundamental law of risk and return for investment states that return
is directly proportional to risk.
Developing new drugs is time-consuming and expensive, but the paybacks
aren’t realized for years, if ever.
There are significant barriers to entry in new drug
design that drive up costs of new drugs.
NMEs (new molecular entities) are screened for therapeutic value, and
companies file patents for potential drug candidates. Animal studies are conducted to determine dosage, toxicity, and
carcinogenicity. Next, the company
submits an NDA (new drug application) with the FDA. FDA approval is a stringent and lengthy process of three human
clinical trial phases and extensive review of results before a drug can be
marketed in the United States.
These entry barriers are manifest in the industry’s
high financial risks. The pharmaceutical
industry develops over 60,000 new molecular entities to obtain a single
blockbuster drug10. In a
study of 63 new drugs with an average approval date of 1997, companies invested
$403 million out of pocket to bring one new drug to market, but one must also
consider the cost of capital (~15%) while the invested money is locked in the
lengthy R&D and approval process, which is $802 million11.
Companies are further challenged by patent
expirations, allowing competitors to sell the drugs at lower prices, free of
major R&D investment. In 2002, the
average time from discovery to market of a new drug was 14.7 years15. Assuming a company patents an NME upon
discovery, a 20-year patent allows a company to exclusively market the drug in
a 5-6 year window to recoup its investment and realize profit.
Utilization is the second factor in the growth of
prescription drug spending. From 1994
to 2004, the number of prescription drugs dispensed increased 68% (2.1 billion
to 3.5 billion annually) while the population grew at 12%. In 2004, 12 prescriptions were filled per
capita per year, with 51% of the population taking prescription drugs daily,
and 27% taking three or more daily7. Combined with the increasing cost of drugs, the increased
utilization has created significant financial burden on insurers and
individuals alike.
Employers and insurers began absorbing a higher
percentage of health care costs during the emergence of managed care in
exchange for more service restrictions.
In 1960, patients paid over 96% of prescription drug cost out of
pocket. The patient’s share has
gradually declined to just under 27% in 19986. The burden of healthcare costs is an issue
of significant concern to employers, and for good reason, as signs of impact on
the national economy have emerged. In
November 2005, General Motors announced 30,000 layoffs due to declining market
share and profitability. The company
cited employee healthcare costs as one of its financial challenges, saying they
added $1,500 to the cost of every automobile17.
As a result of high percentages paid by employers,
patients were insulated from awareness of the true cost of their treatment,
especially as prices increased. The
payers have begun shifting the cost burden back to patients. Strategies for cost containment include
implementation or restructuring of drug formularies, excluding medications,
higher deductibles, and shifting from co-pays to co-insurance (paying a
percentage of cost instead of a fixed co-payment). In 2005, 74% of employer-sponsored insurance beneficiaries
reported using 3 or 4 tier formularies to encourage higher utilization of
cheaper medications that are comparably efficacious to more expensive brand
name drugs3.
The combination of increased utilization, price, and
shift of financial burden from insurers is putting an increasing strain on
patients’ ability to afford medications.
Of the 51% of the population taking at least one prescription drug daily
in 2005, one-third report it is somewhat or very difficult to pay for them, and
about one-fourth of households report a member who did not fill a prescription,
reduced or skipped doses in the last year because of cost7.
Prescription drugs comprise the largest share of
out-of-pocket healthcare spending, 22.9% in 2002. The Medicare drug benefit will decrease this for
Medicare-eligible patients who participate, but not for all patients. Since utilization of medical services with
high out-of-pocket expenses is sensitive to patients’ amount of disposable
income, low-income patients are susceptible to poorer compliance23.
The prescribing habits of practitioners
play a significant role in prescription drug spending. In 2004, the pharmaceutical industry spent
$7.8 billion on physician-directed advertising and $15.9 billion on drug samples
to influence prescribing habits in favor of newer, more expensive brand name
medications3. However,
physicians are increasingly aware of patients’ difficulties affording
drugs. A 2005 survey found 75% of
physicians inquire about patients’ ability to pay for drugs, but only on a
case-by-case basis (not all patients).
Practitioners can prescribe accordingly, and can also direct patients to
drug company assistance programs. An
analysis of over 6,300 prescriptions filled in 1996 in 758 U.S. pharmacies
found that 63% were eligible for generic substitution. With the average brand name prescription
drug costing more than three times the average cost of generic drugs,
prescribers have significant opportunity to influence prescription drug
spending4.
Future Per Literature – “It’s
tough to make predictions, especially about the future.” - Yogi Beara
Prescription drug spending is expected to continue to
be the fastest growing component of total healthcare costs through 2014, but
the decrease in growth rates that began in 2000 is expected to continue
steadily, from a high of 19.7% growth in 1999 to 8.7% by 2014. Reasons for the slowdown include decreased
growth in drug prices, patent expiration of current blockbuster drugs, fewer new
blockbuster drugs expected, and cost containment and cost shifting strategies
by insurers that decrease patient demand.18
Since 1970, U.S. drug companies have been doubling
R&D funding roughly every 6-8 years.
The rate of new drug approvals is less steady, but has been decreasing
from a peak of 56 approvals in 1996, dipping to a 20 year low in 2003 (17
approvals), and up to 35 new approvals in 200419 (note: in 2004 the
FDA began including BLA [biological license application] approval data with NME
approvals, which makes 2004 data difficult to compare to previous years. BLAs include vaccines, monoclonal
antibodies, and blood products). The
decline in approvals is likely due to random variation, increasing costs of R&D
resources, and the lag time for new research (input) to yield new drug approvals
(output). However, some have said the
industry is suffering from decreased productivity and inability to
significantly overcome regulatory barriers and shorten approval times.
Meanwhile, the pharmaceutical industry is
rapidly changing, and it is too early to tell how these changes will play out
in the industry (for example, will the competitive environment of the new
industry structure enhance or hamper efficient use of limited R&D
funding?). Academic research has
increased in complexity and expense, forcing scientists to develop skills
necessary to manage business ventures.
Meanwhile, accelerating research and discovery of potential drug targets
created an unfilled niche now filled by entrepreneurial biotechnology
companies. In the pharmaceutical industry,
once comprised solely of academic institutions and “Big Pharma,” the lines
between the two began to blur.
Science-intensive research made drug company R&D organizations
behave more like universities.
Scientists, now more business-savvy, began pursuing ventures to exploit
the commercial potential of their discoveries.
Biotech firms have positioned themselves between
academic research and Big Pharma.
Smaller and less bureaucratic than large companies, biotech companies
are perhaps better equipped to explore the commercial potential of new
discoveries, but they lack the capital funding to survive the expensive
regulatory barriers and decade-long product development cycle of new drug
design. Thus, they contractually and
collaboratively partner with large drug companies to develop new products.
In the near- to long-term future, new drug research is
expected to proceed more quickly, but at higher costs. The aforementioned restructuring of the
research industry and FDA efforts to move NMEs through the approval pipeline
more quickly (1992 act) will help shorten approval times. Development will be more expensive due to
larger and more expensive research projects, increased cost of skilled labor,
and more rigorous clinical trial design.
Ironically, third party payer cost-containment strategies increase
clinical trial costs by challenging drug companies to show new drug
effectiveness against existing, cheaper medications (instead of placebo) as a
condition of coverage; this increased trial complexity drives up development
cost. Development of a new drug begun
in 2001 and FDA-approved 12 years later is projected to cost $970 million out
of pocket, and $1.9 billion when opportunity costs are included11.
Increased new drug development costs are a given, and
these will be passed along to patients in the form of higher prices for brand
name (innovator) drugs. Insurers and
patients will seek cost containment strategies to counter the price increases.
The Medicare Part D drug benefit begins coverage in
January 2006. The major effect of the
program will be a switch in payer contributions. Out of pocket, private health insurance, and Medicaid
contributions will decrease as Medicare picks up the payments. Medicare will pay 28% of the U.S.
prescription drug budget in 2006, up from 2% in 2005. Out of pocket payments will drop from 29.1% (2005) to 19.9%
(2006), reflecting the program’s benefits to seniors. Expectations are for the program’s price discounts to be offset
by utilization increases; overall, Medicare should have a minor effect on total
prescription drug expenditures, increasing the growth rate 0.5% for the year.18
Ninety percent of pharmacists in a 2004 survey
believed the Medicare drug benefit would not reduce interest in importing
medications20, because the price benefits will not be substantial
enough and pertain only to Medicare recipients. Prices for brand name prescription drugs are regulated in many
countries, and U.S. patients are increasingly asking pharmacists how to import
medications from Canada and other countries at lower prices. Proposals to legalize drug importation have
been introduced at federal, state, and local government levels, and
Springfield, Massachusetts has already setup a program to facilitate access to
Canadian drugs21. Drug quality
and counterfeiting are the major patient safety concerns expressed by the FDA
and pharmacists, but a majority of pharmacists believe the risks of imported
drugs could be mitigated if they were able to monitor patients’ progress.
There is considerable controversy over the effect of
imported drugs on U.S. drug prices. One
camp contends allowing U.S. patients to purchase medications from foreign
pharmacies will foster free market competition and force drug companies to
lower domestic prices. The second camp
argues that drug prices are lower in other countries for two reasons: 1.)
citizens in some countries are unwilling or unable to afford U.S. drug prices,
so drug companies charge lower prices in those countries to reflect the lower
demand; 2.) many countries regulate drug prices. In the first case, allowing U.S. patients to import drugs would
increase demand and increase prices in those countries. In the case of countries with price
controls, exporting drugs to the U.S. would deplete drug supplies in price
control countries and force them to restrict exports. Canada, a country with price controls, has already announced
plans to restrict drug exports to the U.S. to prevent shortages22. It’s also likely drug companies would
restrict sales to any country that exported a significant amount of low-cost
drugs to the U.S.; in fact, Pfizer has already pledged to do just that26. In the end, importation would not
appreciably affect U.S. drug prices.
Furthermore, in the view of the pharmaceutical industry, the U.S. is the
most favorable economy for new drug introductions, because profits from U.S.
sales effectively support new drug research, and U.S. patients demand
groundbreaking drugs and are willing to pay higher prices for them.
In the future, an increasing aging population and
limited financial resources will force payers to prioritize spending by
increasingly using evidence-based medicine to select preferred drug
treatments. As the population ages, the
system will encounter more patients with chronic conditions that require
expensive, long-term treatment. It’s in
the best interests of insurers to pursue such treatment if it will help avoid
progression of disease and more expensive treatments in the future. Before agreeing to pay for disease
management, however, insurers want to ensure the highest level of treatment
efficacy per dollar with the lowest level of adverse effects. This is the concept of rational
pharmacotherapy.
A pure application of rational pharmacotherapy is the
concept of “no cure, no pay.” In such
an arrangement, drug companies justify the cost of their medications by
guaranteeing a quantifiable therapeutic outcome. If the outcome is not met, the insurer and/or patient are
reimbursed for the cost of the medication.
Several “no cure, no pay” strategies have already been offered by drug
companies. Merck offers to refund
patients prescribed Zocor® if their LDL cholesterol is not lowered to a
predetermined target level. Lilly
offered Cialis® customers unsatisfied with the erectile dysfunction treatment
vouchers for the oral ED treatment of their choice24.
As stated earlier, U.S. patients demand groundbreaking
drugs and will pay more for them. This
was evidenced by the unpopularity of closed formularies and utilization
restrictions under managed care. The
trend for the future is open formularies.
Tiered formularies will become more complex, expanding to 4 and 5 tier
systems (a few plans have already introduced such structures). The additional tiers will allow insurers to
categorize drugs by their cost and therapeutic value in addition to
brand/generic status. Following the
trend of rational pharmacotherapy, formularies may structure cost sharing so
that patients pay based on drugs’ overall clinical and economic value, instead
of a fixed co-payment6 that fails to reflect drugs’ true cost or
value. The growing prevalence of
“consumer-based healthcare,” where employers provide high deductible health
insurance plans and a spending account that patients manage, will also enhance
patient awareness of drug costs.
My career goal as a pharmacist is to apply my skills
in pharmacoeconomics. I selected
prescription drug costs to learn more about it. My exposure to the topic consisted of media reports portraying a
hopelessly broken system with no fix in sight, nearly an “end of the world is
nigh” perspective. I had a negative
view of “Big Pharma” drug companies and their financial motivations, and believed
the Heinz Dilemma would be the eventual fate of U.S. healthcare as masses of
patients fell through the safety net of a bankrupt national welfare
system. (The Heinz Dilemma is a
hypothetical scenario for ethical discussions.
It describes a patient is unable to afford an expensive medication to
treat a life-threatening chronic illness.
The patient must steal the medication to survive or she will die.) After completing this project, I’d like to
think I have a more informed (if not realistic) understanding of the topic.
If you only take one thing away from my opinion
section, let it be this: in my opinion,
pharmacoeconomics and a national healthcare policy are key to managing future
prescription drug costs.
I think there are two primary goals for drug therapy. First, it should lower total healthcare
spending. Second, it should improve
patient quality of life. Increased
prescription drug expenditure that achieves these goals is productive
spending. Medications are an effective
cost avoidance tool and increase longevity; healthcare costs would be higher
without them. For example, diabetes and
its complications cost $130 billion annually.
Proper diabetes management saves the healthcare system between $35 and
$50 billion annually. In addition to
lifestyle changes, a host of medications (oral antidiabetic, insulin,
lipid-lowering, antihypertensive, and renal protective agents) prevent
detrimental sequelae such as neuropathy, vascular and renal disease, blindness,
and loss of limbs. Quality of life is more
difficult to quantify.
Pharmacoeconomics has several analysis methodologies to evaluate both
economic and humanistic effects of drug therapy.
The most disturbing realization I’ve had in HA-601 and
my research for this essay is that the United States lacks organized healthcare
policies. Our political leaders seem
more concerned about reelection than making difficult choices. So in the absence of political leadership,
the healthcare system’s stakeholders play tug of war as they negotiate their
own path.
One unmet policy need I see is new drug
development. Developing new drugs is
expensive, and it won’t be cheaper in the future; by extension, the cost of new
drugs will also increase. As research
reveals more drug targets and possibilities, how will we ensure limited R&D
resources will be used for the greatest good?
As I said earlier, increased prescription drug costs are fine if it
results in lower total healthcare spending.
We need to direct R&D where it is needed; infectious disease and
antibiotic resistance are real problems (tuberculosis requires triple
antibiotic therapy; multi-drug resistant staphylococcus aureus is
beginning to defy treatment with all available agents), but no new antibiotic
classes have been discovered in over 30 years.
“Lifestyle” and “me too” drugs needlessly consume
R&D dollars and increase expenditures.
Earlier I defended drug companies for high drug prices, but here is a
case where drug company behavior is indefensible. Lifestyle drugs treat conditions such as erectile dysfunction,
overactive bladder, and male pattern baldness.
“Me too” drugs duplicate the mechanism of existing drugs that are
cheaper and comparably effective. The
worst offenders are purified enantiomers (where a drug molecule has a mirror
image that’s purified for marginally therapeutic improvement); Nexium is the
purified enantiomer of Prilosec for peptic ulcers; Lexapro is the purified
enantiomer of Celexa for depression; and Lunesta is the purified enantiomer of
Ambien for insomnia are examples. Some “me
too” drugs do have important clinical value, however – genetic variations in an
enzyme responsible for cholesterol synthesis result in variable effectiveness
of “statin” drugs in different people; that is, what works for one individual
may not work for another.
DTC (direct-to-consumer) advertising increased 15-fold
between 1994 and 2004 to $4 billion3. The pharmaceutical industry argues DTC creates public awareness
of treatable diseases and encourages patients to seek treatment and obtain
positive health outcomes. On the other
hand, DTC promotes newer, highly profitable brand name drugs and exaggerates
their effectiveness compared to cheaper drugs.
Nexium was demonstrated to be “significantly” better than Prilosec for
treating erosive esophagitis. This is
misleading, because “significant” meant Nexium consistently had a higher cure
rate (90%) than Prilosec (87%), but does the marginal improvement justify
paying four times the cost?
What is the pharmacist’s role in the
future of prescription drug expenditures?
As healthcare costs inevitably rise and consume a larger proportion of
GDP, they will be consuming resources once designated for other uses. As a result, there will be increased
priority on using healthcare dollars more efficiently. Policy leaders will need to place a value on
drug therapies, evaluate which are most therapeutically- and cost-effective,
find the “happy medium,” and justify their findings to patients and
prescribers, which is, by definition, the goal of pharmacoeconomics. Such analysis is becoming evident in
emerging healthcare plans:
Pharmacoeconomic analysis makes it possible to structure multi-tier
formularies according to drugs’ therapeutic value. Health-spending accounts encourage patients and practitioners to
rely on evidence based medicine and make their own pharmacoeconomic decisions.
References
1
Kreling DH, Mott DA,
Wiederholt JB, Lundy J, Levitt L.
Prescription drug trends: a
chartbook update. Kaiser Family
Foundation 2001 Nov.
2
Prescription drug trends
factsheet. Kaiser Family Foundation
2004 Oct.
(http://www.kff.org/rxdrugs/upload/Prescription-Drug-Trends-October-2004-UPDATE.pdf)
3
Prescription drug trends
factsheet. Kaiser Family Foundation
2005 Nov.
(http://www.kff.org/insurance/upload/3057-04.pdf)
4
Mott DA, Cline RR. Exploring generic drug use behavior: the role of prescribers and pharmacists in
the opportunity for generic drug use and generic substitution. Medical Care 40(8):662-674.
5
Trends and indicators in
the changing health care marketplace.
Kaiser Family Foundation 2005 Feb.
(http://www.kff.org/insurance/7031/index.cfm)
6
Olson BM. Approaches to pharmacy benefit management
and the impact of consumer cost sharing.
Clinical Therapeutics 2003 Jan; 25(1):250-272.
7
Health care costs
survey. USA Today, Kaiser Family
Foundation, Harvard School of Public Health 2005 Aug. (http://www.kff.org/newsmedia/upload/7371.pdf)
8
2002 drug industry
profits: hefty pharmaceutical company margins dwarf other industries. Public Citizen 2003 Jun. (http://www.citizen.org/documents/Pharma_Report.pdf)
9
Fiscal 2004 SEC filings
(profit and cash flow statements) of listed companies provided by
finance.yahoo.com
10
Schweitzer SO. Pharmaceutical Economics and Policy,
1st edition. New
York: Oxford University Press, 1997; 21.
11
DiMasi JA, Hansen RW, Grabowski HG.
The price of innovation: new
estimates of drug development costs. Journal
of Health Economics; 22(2003):151-185.
12
Thompson CA. Drug importation not the right answer, Nobel
economist says. American Journal of
Health Systems Pharmacists 2004 Mar 15; 61:546-547.
13
Cockburn IM. The changing structure of the pharmaceutical
industry. Health Affairs 2004
Jan-Feb; 23(1):10-22.
14
Hoffman JM, Shah ND,
Vermeulen LC, Hunkler RJ, Hontz KM.
Projecting future drug expenditures – 2005. American Journal of Health Systems Pharmacists 2005 Jan
15; 62:149-167.
15
Tufts Center for the
Study of Drug Development.
16 Gleason PP, Gunderson BW, Gericke KR. Are incentive-based formularies inversely
associated with drug utilization in managed care? The Annals of Pharmacotherapy 2005; 39:339-345.
17 Berliner U.
How General Motors got stuck in reverse. National Public Radio Morning Edition broadcast: November 28, 2005. (http://www.npr.org/templates/story/story.php?storyId=5028937)
18 Heffler S, Smith S, Keehan S, Borger C, Clemens MK,
Truffer C. Trends: U.S. health spending
projections for 2004-2014. Health Affairs, Web Exclusive,
February 23, 2005; 74-85.
19
CDER approval times for
priority and standard NMEs and new BLAs calendar years 1993-2004. U.S. Food and Drug Administration.
(http://www.fda.gov/cder/rdmt/NMEapps93-04.htm)
20 Pickard AS, Nau DP, McKercher PL, Schumock GT. Importation of prescription medications:
experiences, opinions, and intended behaviors of U.S. community pharmacists. Journal of the
American Pharmacists Association 2004; 44:666-672.
21
Benjamin GC, Levinson
R. Importing drugs: a viable
solution? The Physician Executive
2004 Jan-Feb; 55-57.
22 Learner N.
Canada plans to restrict drug exports; Springfield saves millions. Washington Drug Letter 2005 Jul 11;
37(27).
23
Heffler S, Smith S,
Keehan S, Clemens MK, Zezza M, Truffer C.
Health spending projections through 2013. Health Affairs, Web Exclusive, 2004; W4-79 – W4-93.
24 Møldrup C. No cure, no pay. British Medical Journal 2005 May 28; 330:1262-1264.
25 Physicians stymied by
prescription costs – Pri-Med survey.
Business Wire: May 25, 2005.
26 Fine A. Drug
importation: the process of importing less expensive drugs is no solution for
U.S. Managed Care Quarterly 2005; 13(3):26-27.
27 Minshall ME, Roze S, Palmer AJ, Valentine WJ, Foos V,
Ray J, Graham C. Treating diabetes to
accepted standards of care: a 10-year projection of the estimated economic and
health impact in patients with type 1 and type 2 diabetes mellitus in the
United States. Clinical
Therapeutics 2005 Jun; 27(6):940-950.
28
Gladwell M. How to think about prescription drugs. The
New Yorker: October 25, 2004.