Marapr07.vp

Richard M. Scheffler, Stephen P. Hinshaw, Sepideh Modrek and Peter Levine The online version of this article, along with updated information and services, is For Reprints, Links & Permissions:
is published monthly by Project HOPE at 7500 Old Georgetown Road, Suite 600, Health Affairs Bethesda, MD 20814-6133. Copyright 2007 by Project HOPE - The People-to-People HealthFoundation. As provided by United States copyright law (Title 17, U.S. Code), no part of Health Affairs may be reproduced, displayed, or transmitted in any form or by any means, electronic or mechanical,including photocopying or by information storage or retrieval systems, without prior written permission from the Publisher. All rights reserved.
Not for commercial use or unauthorized distribution The United States is an outlier among developed countries in its highusage rates of these medications among children.
by Richard M. Scheffler, Stephen P. Hinshaw, Sepideh Modrek, and ABSTRACT: Little is known about the global use and cost of medications for attention defi- cit hyperactivity disorder (ADHD). Global use of ADHD medications rose threefold from 1993 through 2003, whereas global spending (U.S. $2.4 billion in 2003) rose ninefold, ad- justing for inflation. Per capita gross domestic product (GDP) robustly predicted use across countries, but the United States, Canada, and Australia showed significantly higher-than- predicted use. Use and spending grew in both developed and developing countries, but spending growth was concentrated in developed countries, which adopted more costly, long-acting formulations. Promoting optimal prescription and monitoring should be a prior- ity. [Health Affairs 26, no. 2 (2007): 450–457; 10.1377/hlthaff.26.2.450] Attention deficit hyperactivity keyissuesaresalient.First,objectivetestsfor most other psychiatric conditions. Guidelines subject of great clinical interest and strong for proper diagnosis must be followed, so that scientific investigation.1 The prevalence of disorders with similar symptoms (such as con- ADHD is 5–8 percent of U.S. children and ad- duct disorder or bipolar disorder) do not re- olescents, and its impairments are likely to ceive inappropriate stimulant treatment.5 Sec- persist into adolescence and adulthood.2 A ond, prescription rates of stimulants (and syndrome with strong neurobiological ori- other psychotropic medications) have greatly gins, ADHD has major importance for public increased, at least in the United States, even health, given the marked academic, social, fa- for preschoolers.6 About half of U.S. children milial, and accidental injury–related impair- ceive stimulant medications or related agents.7 First used for youth in the 1930s, psycho- Third, unintended side effects may accompany stimulant medications enhance dopaminergic health concerns have arisen about the poten- tial for negative cardiovascular effects and sui- majority of people who receive them.4 Several cidal thoughts.8 Fourth, questions abound Richard Scheffler ([email protected]) is director of the Nicholas C. Petris Center on Health Care Markets andConsumer Welfare, and Distinguished Professor of Health Economics and Public Policy at the University ofCalifornia, Berkeley. Stephen Hinshaw is professor and chair, Department of Psychology, at that university.
Sepideh Modrek is an Agency for Healthcare Research and Quality Pre-Doctoral Fellow and a graduate student inhealth services and policy analysis at the university. Peter Levine is a physician in the Department of Pediatrics,Kaiser Permanente Walnut Creek Medical Center, in Walnut Creek, California. DOI 10.1377/hlthaff.26.2.450 2007 Project HOPE–The People-to-People Health Foundation, Inc. about the potential for “diversion” of prescrip- amphetamine, or other stimulant (Exhibit 1).
tion stimulants to people without ADHD, re- lated to improvement of study skills and pos- then generated a three-level categorization of short-acting stimulants, long-acting medica- Since the 1960s, methylphenidate and dex- tions that received approval from the Food and troamphetamine have been the mainstays of Drug Administration (FDA) during the study stimulant prescriptions for ADHD. New for- period, and other long-acting medications. As mulations, as well as efficacious nonstimu- a proxy for dose, IMS Health calibrated each lants, have taken up an increasingly large U.S.
product form into a common metric known as market share since the late 1990s.10 Given the a standard unit (SU). For example, one stan- increasing recognition that ADHD is a prob- dard unit is equivalent to a 5 mg tablet or 5 ml lem of worldwide scope, diffusion of ADHD medications outside the United States may To compare countries’ usage, we created a well be occurring, but the extent to which usage per capita measure (SU per child ages 5– 19) to examine the relationship between use of Countries with different diagnostic tradi- each country that is a member of the Organiza- tions, such as the United Kingdom, have had tion for Economic Cooperation and Develop- ment (OECD). We estimated a fixed-effects States.12 However, ADHD appears in multiple model to establish the relationship between nations and cultures at similar prevalence per capita (ages 5–19) use and per capita rates.13 Less well documented are international GDP.14 This method estimates parameters in treatment patterns, including rates of use and panel data, which controls for variability associated costs. Our objectives are to (1) de- scribe global trends in ADHD-related medica-tion use, the types of medications used, and medication spending over the past decade; (2) In 1993, thirty-one countries had adopted make a preliminary determination of the fac- the use of ADHD medications; by 2003, the tors that explain diffusion of ADHD medica- number had grown to fifty-five.16 Our analysis tions globally; and (3) discuss relevant policy aggregated data from individual countries from the year of adoption into a global sum.
Because the United States is the single largest market, we also show its use, which consti- tutes 83–90 percent of total market share (by to analyze trends in the global market for classified ADHD medications as those in the medications increased 274 percent during the “ATC=N6B Psychostimulants” category, along study period (Exhibit 2).17 From 1993 to 2000, with the nonamphetaminelike stimulant mo- global volume increased steadily (13.2 percent dafinil (Provigil, Cephalon) and the nonstim- per year); from 2000 to 2003, growth acceler- ulant atomoxetine HCL (Strattera, Lilly). Each ated to 16.8 percent per year.18 The U.S. share of the global market declined from 86.8 percent classified into one of four categories, along the in 1993 to 83.1 percent in 2003. A more detailed dimensions of (1) stimulant versus nonstimu- analysis (not shown) reveals marked variation lant medications and (2) long-acting (formu- lations that remain active for at least eight growth rates as high as 46 percent per year, hours) versus short-acting (fewer than eight whereas moderate-use countries had growth hours). We further classified stimulants by ac- rates of nearly 20 percent per year. If these tive pharmacologic agent: methylphenidate, rates continue, the U.S. market share should H E A L T H A F F A I R S ~ Vo l u m e 2 6 , N u m b e r 2 EXHIBIT 1Categories Of Attention Deficit Hyperactivity Disorder (ADHD) Medications SOURCE: Authors’ classification scheme, based on the pharmacologic literature.
NOTES: These medication names are those used commonly in the United States. The equivalent preparations were a Long-acting medications approved by the Food and Drug Administration (FDA) after January 1993.
fixed-effects model.20 Exhibit 3 illustrates the We then analyzed the relationship between relationship between per capita (ages 5–19) per capita utilization of ADHD medications use and per capita GDP. Data points on or near the diagonal line show countries using these tries, given that income is a well-known pre- medications at levels predicted by their per ca- dictor of health care spending.19 Use of ADHD pita GDP. The United States uses them at a medications is positively related to per capita level that is about four times higher than GDP, with a p value of less than 0.001 using a would be predicted by income alone, holding EXHIBIT 2U.S. And Global Volume Of Attention Deficit Hyperactivity Disorder (ADHD)Medications, 1993–2003 SOURCE: MIDAS database, IMS Health, 1993–2003.
NOTES: Volume adjusted to generate dosage equivalence between short- and long-acting medications. Long-acting medicationsare weighted twofold over short-acting medications. SU is standard units.
EXHIBIT 3Predicted And Actual Usage Of Attention Deficit Hyperactivity Disorder (ADHD)Medications In OECD Countries, By Income, 2003 Per capita GDP, in thousands of U.S.$ PPPs SOURCE: MIDAS database, IMS Health, 2003.
NOTES: Although the data are shown for a single year, other years look similar in overall patterns; 2003 is the most recent yearfor which data were available. Only twenty-seven of the thirty Organization for Economic Cooperation and Development (OECD)member states are represented; no data were available for Iceland, Denmark, or Slovakia.TUR is Turkey. MEX is Mexico. POL isPoland. HUN is Hungary. CZE is Czech Republic. PRT is Portugal. PRK is South Korea. GRC is Greece. NZL is New Zealand. ESP isSpain. ITA is Italy. DEU is Germany. JPN is Japan. FIN is Finland. FRA is France. SWE is Sweden. BEL is Belgium. GBR is UnitedKingdom. AUS is Australia. NLD is Netherlands. CAN is Canada. AUT is Austria. CHE is Switzerland. IRL is Ireland. NOR is Norway.
USA is United States. LUX is Luxembourg. Where space prohibits labeling each point, abbreviations are as follows: 1 = PRT, GRC,PRK; 2 = FIN, SWE, FRA, JPN; 3 = GBR, BEL. SU is standard units. GDP is gross domestic product. PPP is purchasing power parity.
a Countries have significantly different usage than the predicted at the 95 percent confidence interval.
all country-specific deviations constant. Can- cent growth rate per year. Other countries showed a slightly slower spending growth rate expected use. In contrast, Italy, Ireland, Aus- than the United States—about 21.0 percent tria, Japan, Sweden, and Finland use less than per year. A country-by-country analysis re- predicted by per capita GDP. A country-by- veals that spending increases were much more country analysis of the growth rate of use of ADHD medications over the past four years shows increases in both developed countries change in the medications used to treat ADHD (Exhibit 5). The volume of short-acting medi- cations plateaued and then steadily decreased to spending, the global expenses regarding after 1999, whereas the volume of long-acting ADHD medications were U.S.$2.4 billion by formulations increased during the second half 2003, representing a ninefold increase (ad- of the study period. The timing of this shift is justed for inflation) since 1993 (Exhibit 4).
related to FDA approval of Concerta (Johnson From 1993 to 2000, spending grew steadily and Johnson), a controlled-release form of (about 17.6 percent per year), but after 2000 methylphenidate, in August 2000. Although the annual growth rate increased more steeply the release of this and other relatively costly to 40.9 percent. This acceleration was largely long-acting formulations such as Adderall XR driven by the U.S. market, where newer medi- (Shire) and Strattera (Lilly) help explain some cations—primarily longer-acting formula- of the increase in use, their presence is particu- tions—became available. The United States larly related to the large spending increase. Us- dominates global spending on ADHD medica- ing the same data set, we found that the in- tions, making up approximately 92–95 per- crease in spending for ADHD medications in cent of the total expenditures, with a 22.6 per- the U.S. market was attributable mostly to H E A L T H A F F A I R S ~ Vo l u m e 2 6 , N u m b e r 2 EXHIBIT 4U.S. And Global Spending On Attention Deficit Hyperactivity Disorder (ADHD)Medications, 1993–2003 SOURCE: MIDAS database, IMS Health, 1993–2003.
NOTES: Spending is deflated to 2003 U.S. dollars using the U.S. Consumer Price Index. Cross-sectional variation from country tocountry was accounted for by IMS Health, which had converted all local currencies to U.S. dollars using purchasing power parity(PPP) methods. SU is standard units.
price increases.23 From 1994 to 2003, U.S. sales 2002, as long-acting formulations were intro- volume increased 80 percent, while prices in- duced worldwide. These data suggest that the creased 285 percent in real dollars. In contrast, pattern in these countries lags behind that of in OECD countries sales volumes increased 322 percent, while prices rose only 70 percent.
A sampling of prices from Consumer Reports Best Buy Drugs reveals a large differential between short- and long-acting formulations.24 Beyond United States has greatly increased during the the United States (data not shown), the use of past twenty years.25 Although the increase short-acting formulations began to plateau in might be leveling off for children, much con-cern has been raised regarding the potential EXHIBIT 5Global Volume Of Attention Deficit Hyperactivity Disorder (ADHD) Medications, ByCategory, 1993–2003 SOURCE: MIDAS database, IMS Health, 1993–2003.
NOTE: SU is standard units.
for overdiagnosis of the disorder, the potential for overuse of these medications, and the pos- function of per capita GDP showed that de- against their important clinical benefits under spite robust predictions of use from this indi- conditions of careful diagnosis and treatment cator of national income, other factors might monitoring.26 This study has confirmed that explain variation in use. For example, U.S. cli- the United States is by far the world’s largest nicians tend to recognize ADHD as a debilitat- ing disorder.30 Furthermore, changes in the countries have lagged behind is not well un- federal special education law (the Individuals derstood. Advertising in the United States is with Disabilities Education Act [IDEA]) en- clearly an important factor; the number of U.S.
acted in 1991 opened up special education ser- medical specialists who are able to diagnose and treat ADHD is also crucial. Clearly, na- greater diagnosis.31 Other factors include tional policies about the purchase of medica- widespread third-party medication coverage, tions by the health system—as well as the very marketing efforts by pharmaceutical firms, nature of different national health care sys- and a general increase in the use of all psycho- tems—are essential factors. It appears that lit- tropic medications for children.32 Finally, the tle difference exists in the rates of the disorder United States has the highest overall drug between the United States and other coun- tries, although rates of “diagnostic prevalence” (that is, cases actually diagnosed by clinicians) Despite major differences from the U.S.
lag well behind true prevalence outside the health care system and considerable regulation of pharmaceutical prices, Canada also exhibits Other countries, however, are beginning to high usage of ADHD medications.34 This pat- follow U.S. trends. Over the past decade, use of tern might be a function of its proximity to the United States, with exposure to U.S. advertise- tries has increased at rates even greater than ments and cultural norms. Within other coun- those in the United States. Furthermore, coun- tries (such as Sweden and France), strict gov- tries using the most ADHD medications per ernmental regulation against the prescription capita are those with the highest incomes.
of ADHD medications might also explain some Over the past five years, many developing of the variation in use.35 For example, in countries have had yearly growth rates in use France, use of methylphenidate, the only ap- proved ADHD medication, requires a hospital- initiated prescription from a neurology, psy- increasing globally, the cost of these medica- chiatry, or pediatric specialist.36 In the Nether- tions is escalating even more rapidly. With the lands, there are lower rates of prescription advent of longer-acting stimulants and non- than in the United States, but major increases stimulant agents, spending for ADHD medica- of stimulant use became evident in the late tions has risen dramatically. Note that a 1990s.37 Other potential factors related to dif- month’s supply of generic methylphenidate ferential use rates include differing diagnostic taken two or three times per day costs around criteria for ADHD, the professional training of U.S.$25, whereas the cost of one month’s sup- physicians and mental health care providers ply of Concerta 18 mg, taken once a day, is with respect to ADHD, differences in national U.S.$109.29 The change to long-acting agents health care systems, rates of overall pharma- began in the United States in 2000; other ceutical spending, perceptions of ADHD by countries are following the same trend but parents and teachers, and variation in educa- with a lag of several years. We can expect that the already burgeoning global costs for medi- ADHD medications, their costs, and their po- H E A L T H A F F A I R S ~ Vo l u m e 2 6 , N u m b e r 2 tential risks and benefits is now a global issue.
N.J.: Civic Research Institute, 2002), 5-1–5-21.
ADHD clearly involves biological, behavioral, 4. C. Bradley, “The Behavior of Children Receiving and environmental factors.39 Economic and Benzedrine,” American Journal of Psychiatry 94(1937): 577–585; and L.L. Greenhill and B.B.
cultural differences among countries are likely Osman, eds., Ritalin: Theory and Practice (Larch- to be salient for adoption and usage of medica- mont, N.Y.: Mary Ann Liebert, 2000).
tion treatments. Given the global diffusion of 5. Subcommittee on Attention-Deficit/Hyperactiv- ADHD medications as well as the prevalence ity Disorder and Committee on Quality Improve- ment, “Clinical Practice Guideline: Treatment of leading childhood disorder treated with medi- the School-Aged Child with Attention-Deficit/ cations across the globe. Clear priorities are (1) Hyperactivity Disorder,” Pediatrics 108, no. 4(2001): 1033–1044.
determining long-term benefits of pharmaco- 6. J.M. Zito et al., “Trends in the Prescribing of Psy- logic treatments and (2) ascertaining eco- chotropic Medications to Preschoolers,” Journal of nomic, professional training, and cultural fac- the American Medical Association 283, no. 8 (2000): tors that promote optimal prescription and 1025–1030; D.J. Safer, J.M. Zito, and E.M. Fine, “Increased Methylphenidate Usage for Attention Deficit Disorder in the 1990s,” Pediatrics 98, no. 6,Part 1 (1996): 1084–1088; and J.M. Zito et al., compare data on use and spending, to adjust “Rising Prevalence of Antidepressants among overuse or underuse, and that they weigh care- U.S. Youths,” Pediatrics 109, no. 5 (2000): 721–727.
fully the potential benefits versus the potential 7. Centers for Disease Control and Prevention, liabilities (side effects, addiction, diversion) of “Mental Health in the United States: Prevalence medication treatment for ADHD. During this of Diagnosis and Medication Treatment for At- era of global rises in medication use and ex- tention-Deficit/Hyperactivity Disorder—UnitedStates, 2003,” Morbidity and Mortality Weekly Report 8. G. Harris, “Warning Urged on Stimulants Like This study was funded in part by a grant from the Ritalin,” New York Times, 10 February 2006.
National Institute of Mental Health [1R01 9. H. Upadhyaya et al., “Attention-Deficit/Hyperac- MH067084-01 MH1 SRV-C (01)]; in part by a grant tivity Disorder, Medication Treatment, and Sub- from the Center for Child and Youth Policy, University stance Use Patterns among Adolescents and of California, Berkeley; and in part by the Nicholas C. Young Adults,” Journal of Child and Adolescent Psycho-pharmacology 15, no. 5 (2005): 799–809; and S.E.
Petris Center on Health Care Markets and Consumer McCabe, C.J. Teter, and C.J. Boyd, “The Use, Welfare, School of Public Health, University of Misuse, and Diversion of Prescription Stimulants California, Berkeley. Special thanks to Teh-Wei Hu, among Middle and High School Students,” Sub- Farasat Bokhari, Hui-Chu Lang, Mistique Felton, Tim stance Use and Misuse 39, no. 7 (2004): 1095–1116.
Brown, Laurie Habel, Susan Stone, Tom Ray, and Brent 10. Safer et al., “Increased Methylphenidate Usage.” Fulton for their helpful comments and reviews. 11. J. Biederman and S.V. Faraone, “Attention Deficit Hyperactivity Disorder: A Worldwide Concern,”Journal of Nervous and Mental Disease 192, no. 7 R.A. Barkley, “Attention-Deficit Hyperactivity 12. National Institute for Health and Clinical Excel- Disorder,” in Child Psychopathology, 2d ed., ed. E.J.
lence, “Guidance on the Use of Methylphenidate Mash and R.A. Barkley (New York: Guilford, (Ritalin, Equasym) for Attention Deficit/Hyper- 2003), 75–143; and M. Rappley, “Attention Defi- activity Disorder (ADHD) in Childhood,” Tech- cit-Hyperactivity Disorder,” New England Journal of nology Appraisal Guidance no. 13, October 2000, Medicine 352, no. 2 (2005): 165–173.
http://nice.org.uk/pdf/Methylph-guidance13.pdf 2. S. Mannuzza and R.G. Klein, “Long-Term Prog- nosis in Attention-Deficit/Hyperactivity Disor- 13. J. Nigg, What Causes ADHD: Understanding What der,” Child and Adolescent Psychiatric Clinics of North Goes Wrong and Why (New York: Guilford, 2006).
America 9, no. 3 (2000): 711–726.
14. J. Wooldridge, Econometric Analysis of Cross Section 3. S.P. Hinshaw, “Is ADHD an Impairing Condition and Panel Data (Cambridge, Mass.: MIT Press, in Childhood and Adolescence?” in Attention-Defi- cit Hyperactivity Disorder: State of the Science, Best Prac- 15. A detailed description of the data construction tices, ed. P.S. Jensen and J.R. Cooper (Kingston, and methods can be found in the online technical 30. R.A. Barkley et al., “International Consensus appendix, available at http://content.health Statement on ADHD: January 2002,” Clinical Child affairs.org/cgi/content/full/26/2/450/DC1.
and Family Psychology Review 5, no. 2 (2002): 89–111.
16. We separated Hong Kong and Taiwan from 31. J.M. Swanson, M. Lerner, and L. Williams, “More mainland China because this separation is made Frequent Diagnosis of Attention Deficit-Hyper- activity Disorder,” New England Journal of Medicine 17. We use the terms usage and volume interchange- 32. H.A. Huskamp et al., “Impact of Three-Tier For- 18. All growth rates are calculated using the least- mularies on Drug Treatment of Attention-Defi- squares method described in Section B of the on- cit/Hyperactivity Disorder in Children,” Archives line technical appendix. See Note 15.
of General Psychiatry 62, no. 4 (2005): 435–441; E.R.
19. R. Cooper, T.E. Getzen, and P. Laud, “Economic Berndt, “The U.S. Pharmaceutical Industry: Why Expansion Is a Major Determinant of Physician Major Growth in Times of Cost Containment?” Health Affairs 20, no. 2 (2001): 100–114; and J.M.
38, no. 2 (2003): 675–696; and T. Hitiris and J.
Zito et al., “Psychotropic Practice Patterns for Posnett, “The Determinants and Effects of Youth: A Ten-Year Perspective,” Archives of Pediat- Health Expenditure in Developed Countries,” rics and Adolescent Medicine 157, no. 1 (2003): 17–25.
Journal of Health Economics 11, no. 2 (1992): 173–181.
33. IMS Health, “World Pharma Sales 2001: US Still 20. See Section B of the online technical appendix, as Driving Growth,” http://www.imshealth.com/ web/content/0,3148,64576068_63872702_70260998_70328515,00.html (accessed 1 Novem- 21. See Section C of the online technical appendix; 34. A. Charach et al., “Correlates of Methylphenidate Use in Canadian Children: A Cross-Sectional 23. H.C. Lang, R.M. Scheffler, and T.W. Hu, “ADHD Study,” Canadian Journal of Psychiatry 51, no. 1 Medications in the U.S. and OECD Countries: Determinants of Quantity and Price,” Petris 35. L. Sizoo, “Swedish Doctors Wrote Illegal ADHD Working Paper no. 100/2006, http://petris.org/ Prescriptions,” Local, 5 February 2005; and C.
project-ADHD_medications_US_OECD.htm (ac- Frances et al., “Paediatric Methylphenidate (Ritalin) Restrictive Conditions of Prescription 24. “Evaluating Prescription Drugs Used to Treat: in France,” British Journal of Clinical Pharmacology 57, Attention Deficit Hyperactivity Disorder— Comparing Effectiveness, Safety, and Price,” Con- 36. Frances et al., “Paediatric Methylphenidate sumer Reports Best Buy Drugs, 2005, http://www .crbestbuydrugs.org/PDFs/ADHDfinal.pdf (ac-cessed 11 December 2006).
37. E. Schirm et al., “Psychotropic Medication in Children: A Study from the Netherlands,” Pediat- 25. J.L. Rushton and J.T. Whitmire, “Pediatric Stimu- lant and Selective Serotonin Reuptake InhibitorPrescription Trends,” 38. D. Bramble, “Annotation: The Use of Psycho- tropic Medications in Children: A British View,” lescent Medicine 155, no. 5 (2001): 560–565.
Journal of Child Psychology and Psychiatry 44, no. 2 26. National Center on Addiction and Substance (2003): 169; and C. Malacrida, “Medicalization, Abuse, “Under the Counter: The Diversion and Ambivalence, and Social Control: Mothers’ De- Abuse of Controlled Prescription Drugs in the scriptions of Educators and ADD/ADHD,” Health U.S.,” July 2005, http://www.casacolumbia.org/ Absolutenm/articlefiles/380-final_report.pdf (ac-cessed 1 November 2005); and G.T. Ray et al., “At- 39. Barkley, “Attention-Deficit Hyperactivity Disor- tention-Deficit/Hyperactivity Disorder in Chil- dren: Excess Costs Before and After Initial Diag- 40. J.M. Rey and M.G. Sawyer, “Are Psychostimulant nosis and Treatment Cost Differences by Ethnic- Drugs Being Used Appropriately to Treat Child ity,” Archives of Pediatrics and Adolescent Medicine 160, and Adolescent Disorders?” British Journal of Psychi- 27. Nigg, What Causes ADHD.
28. See Section C of the online technical appendix, 29. “Evaluating Drugs Used to Treat: Attention Defi- H E A L T H A F F A I R S ~ Vo l u m e 2 6 , N u m b e r 2

Source: http://gspp.berkeley.edu/assets/uploads/research/pdf/Health_Aff-2007-Scheffler-450-7.pdf