22217 hpo no. 12 2007new

No. 12 • September 2007
Reform without Reason: What’s Wrong with
the FDA Amendments Act of 2007
By John E. Calfee
The recently passed Food and Drug Administration Amendments Act (FDAAA) actually combines atleast two laws. One renews the Prescription Drug User Fee Act (PDUFA), which sets deadlines forFDA action on new drug applications and assesses user fees to cover the salaries of extra FDA employees.
PDUFA expires on September 30. Hardly anyone wants to see the massive FDA layoffs that wouldensue if user fees cease to flow, so FDAAA was must-pass legislation. The second part of FDAAA isCongress’s response to the perceived drug safety crisis that burst forth in October 2004 after the arthritispain reliever Vioxx was pulled from the market because of excess heart attacks (Calfee 2005). With thetwo parts intertwined, reform of FDA’s drug safety oversight will also arrive by October 1. As far as thisHealth Policy Outlook is concerned, FDAAA is an FDA reform law. It commands attention becauseit promises to bring the most important changes in FDA regulation in at least a decade and probablysince the landmark 1962 amendments that created the modern FDA. FDAAA rests upon two foundations. One is the FDA’s competence and culture while suggesting assumption that we are in the middle of a drug numerous legislative and regulatory solutions safety crisis. The other is that there is something (e.g., Topol 2004b; Furberg et al. 2006; Strom seriously wrong at the FDA in the sense that the 2006; Avorn 2007; Hennessy and Strom 2007).
agency has become too friendly to the pharma- By all accounts, however, a central influence ceutical industry and has downplayed drug safety has been a September 2006 report, The Future of Drug Safety: Promoting and Protecting the Health of the Public (NAS 2006; all citations are to the The Institute of Medicine Report and
“uncorrected proofs” released at that time), pub- Other Influences
lished by the Institute of Medicine (IOM),which is part of the National Academy of Sci- Before explaining why these assumptions are mis- ences. The IOM report was actually commissioned taken, a few words about their origins will be use- by the FDA itself, but it offered harsh criticism of ful. FDAAA and FDA reform have been supported the agency and recommended sweeping changes.
by the editorials, a wide array of members of Con- gress, a highly visible dissident FDA staffer (Kohn mendations. Reform proponents, especially in and Bor 2004), and, especially, high-profile medical academia, routinely cite the IOM report (e.g., journals, which have mounted an assault on the Curfman, Morrissey, and Drazen 2006; New YorkTimes 2006; Smith 2007).
Despite its influence and provenance, the John E. Calfee ([email protected]) is a resident scholarat AEI.
IOM report was deeply flawed. One problem is that none of the authors were drug development special- evidence. Sifting through the rapidly accumulating ists, which virtually guaranteed that the exigencies of literature on the episodes that animated the IOM and new drug development would be slighted in comparison Congress, one is struck by two things. First, there is no to drug safety. Worse, the report was remarkably evidence of a drug safety crisis or even a worsening of unscholarly and provided only cursory scientific support drug safety. And second, there is little in the most telling for much of its analysis and recommendations. On drug safety episodes for which the FDA could be held direct-to-consumer advertising of prescription drugs, for example, the authors cited a few largely irrelevant olderarticles while ignoring a flood of recent econometric Vioxx, SSRIs, and More. The triggering event for
research (Berndt 2006; Calfee 2007) and a much-cited FDAAA was Merck’s withdrawal of Vioxx on Septem- randomized trial that revealed large health benefits from ber 30, 2004, after an ongoing clinical trial revealed antidepressant advertising (Kravitz et al. 2005). On excess heart attacks among Vioxx users (Psaty and another crucial topic—“off-label” prescribing for uses Furberg 2005). As the FDA presciently pointed out at not explicitly approved by the FDA—the only citation the time, it was far from clear that Vioxx or its competing (pp. 2–7) was to a Washington Post article (Boodman Cox-2 inhibitor, Celebrex, was significantly riskier than 2006) instead of the Archives of Internal Medicine article the much older non-steroidal anti-inflammatory drugs (Radley et al. 2006) that the Post article was about. On (NSAIDs) they replaced, given that these older drugs yet another central issue—whether FDA drug warnings had never been subjected to rigorous clinical trials like are effective—the report (pp. 2–16) cited a trade press the one that brought Vioxx down. Subsequent research report (Medical News Today 2006) rather than the has largely vindicated that view, with the entire class of Archives of Internal Medicine study it was about (Lasser NSAIDs (old and new, Cox-2s or not) now bearing heart et al. 2006). That article actually found that medical attack warnings (Calfee 2005; Kearney et al. 2006).
harm from prescriptions that violated warnings was The second-ranking triggering event was controversy extremely rare (an estimated total of sixteen such events over previously non-public clinical trial results in which among the 324,548 patients in the study).
children and adolescents taking one of the SSRI class of antidepressants (Prozac or Zoloft, for example) weremore likely to exhibit suicidal “ideation” or thoughts (but not to attempt or commit suicide). Faced withrelentless criticism from litigators, politicians, popular press editorialists, and elite medical journals, the FDAimplemented its strongest warning (a “black box,” Amazingly, the IOM report avoided the most impor- which appears on the FDA-approved label) for all tant question: is there a drug safety crisis at all? The report antidepressants, not just SSRIs (because again, there stated at the outset, “The committee did not attempt to was little reason to think that older drugs, which can document whether or not a drug safety crisis exists, and cause fatal overdoses, are safer). Subsequent research this report should not be interpreted as commenting on taking a variety of approaches has found that SSRI use is that claim one way or the other” (p. 1-1). The authors strongly associated with lower, not higher, suicide rates, argued, however, that various events—notably the with- and that the highly publicized warnings probably did drawal of Vioxx and a controversy over clinical trial more harm than good by reducing antidepressant use. results suggesting a suicide risk from the widely prescribed In particular, a series of reports has found that there is a selective serotonin reuptake inhibitor (SSRI) class of anti- striking, inverse relationship between SSRI prescriptions depressants—had “contributed to a public perception that and youth suicides in a variety of data sets and that the the drug safety system is in crisis” (p. 1-1). The rest of the imposition of new FDA warnings (beginning with public report essentially assumed that this perception is true.
health alerts) is strongly associated with reduced anti-depressant prescribing for children (and younger adults) The Drug Safety Crisis That Wasn’t There
and higher suicide rates (Shogren 2004; McKeown,Cuffe, and Schulz 2006; Ludwig, Marcotte, and Norberg The widespread perception of a drug safety crisis has 2007; Brent 2007; Gibbons et al. 2007; Lubell et al. 2007; been driven by elaborate anecdotes rather than systematic Other frequently cited examples of FDA failure also 2007, an editorial in the journal Nature Clinical Practice do not withstand scrutiny. An example is the pioneer vigorously attacked the Avandia meta-analysis on antibiotic Ketek. The FDA was actually slow in approving methodological grounds and criticized the New England this drug, mainly because of fraud in one of the pivotal Journal for rushing it into print and causing confusion clinical trials, and it finally approved Ketek partly on the among patients (Fuster and Farkouh 2007).
basis of five years of experience in Europe (where it Academic controversy over the New England Journal remains in use). This unorthodox process provoked much meta-analysis and its meaning has continued. Some of criticism, but any new class of reasonably safe and effec- this appeared in the New England Journal itself, where tive antibiotics is a valuable addition, and despite the communications focused on methodological weaknesses controversy, there is little reason to think this particular that strongly undermined the basic results of that analysis drug is especially dangerous (Usdin 2006a, 2006b).
(see the letters by Bracken; Brett; Diamond and Kaul;and Mannucci et al. 2007). Nonetheless, the New England Avandia and the New England Journal’s FDA Reform
Journal published another editorial endorsing stronger Agenda. Finally, there is Avandia, a popular diabetes drug
FDA reform (Psaty and Furberg 2007b). On September first approved in 1999. On May 21, 2007, the New 11, 2007, the Journal of the American Medical Association England Journal of Medicine published a meta-analysis of (JAMA) published two more meta-analyses. One found, adverse cardiovascular events in clinical trials of Avandia again, excess heart attacks (but slightly fewer deaths) for (Nissen and Wolski 2007). Coauthored by a prominent Avandia users (Singh et al. 2007). The other revealed critic (Nissen) of the FDA’s handling of Vioxx, the meta- slightly reduced cardiovascular risk for Actos, Avandia’s analysis revealed excess heart attacks and strokes among chief competitor (Lincoff et al. 2007). A JAMA editorial Avandia users. Accompanying the meta-analysis was an largely echoed the earlier New England Journal editorials editorial by two well-known advocates of FDA reform (Psaty and Furberg 2007a), one of whom (Psaty) wasamong the IOM authors. They declared that the Avandia meta-analysis revealed FDA neglect and was reasonenough to implement one of the more radical reforms, the creation of an independent drug safety board (some- strongly associated with lower, not higher, thing the IOM report declined to recommend).
Events since publication of the New England Journal suicide rates, and that the highly publicized meta-analysis have been instructive. Very soon, the pub-lic learned that the meta-analysis authors and the journal editors were politically motivated (Usdin 2007b, Gottlieb good by reducing antidepressant use.
2007c [May 29]), with Nissen having worked closely withRepresentative Henry Waxman’s (D-Calif.) pro-reformstaff. Dissent quickly emerged from the academic medical In the middle of all this came a July 30, 2007, joint community regarding both policy and research methods.
meeting of two FDA advisory committees. The members Endocrinologists (who research and treat diabetes) and voted 23–1 that Avandia involves an elevated risk of the editors of The Lancet, a leading British journal often (minor) heart attacks compared to placebos, but they critical of the FDA, declared themselves deeply troubled offered no consistent view on cardiovascular risks com- by the precipitous May 21 New England Journal editorial pared to alternative drugs, including Actos (Hampton and its reliance upon what was clearly very tentative evi- 2007; Usdin 2007d). The committee also voted 21–3 in dence (Lancet 2007). FDA staff, researchers at Glaxo- favor of keeping Avandia on the market and in support of Smith-Kline (GSK, the manufacturer of Avandia), and new label warnings, but not the FDA’s strongest “black others pointed out that significant cardiovascular risk box” warning. In the meantime, the liability bar began its had not been revealed by large randomized trials, includ- inevitable work on Avandia litigation (Usdin 2007c).
ing the ongoing Rosiglitazone Evaluated for Cardiac These results are very mixed. The imprecision and Outcomes and Regulation of Glycaemia in Diabetes questionable reliability of meta-analyses have been (RECORD) trial, which was designed to address cardio- made clear, as articles have demonstrated how modest vascular safety (Krall 2007; Home et al. 2007). In July changes in data and statistical methods (which involve considerable judgment even on such basic matters as materials) amidst vast regions of opacity (cf. Calfee whether to include trials in which no adverse events 2006). It is surely plagued by dysfunctions of all sorts, but occurred) can dramatically alter the results. Many, if reform advocates have fixed upon the utterly implausible not most, endocrinologists clearly want to maintain argument that the FDA staff tends to slight drug safety treatment options for a condition (diabetes) that is while catering to the industry. Part of their argument is notoriously variable among patients and correspond- that the FDA’s partial reliance on industry user fees ingly difficult to treat. Avandia clinical trials continue, undermines safety (Okie 2005). But user-fee deadlines under FDA oversight as always. Obviously, experts can only require faster decisions, not decisions that are favor- disagree over such matters as the nature and timing of able to the applicant. The IOM report reviewed the lit- cardiovascular warnings for Avandia, but that is very erature on user fees (pp. 3; 5–8) and found no evidence different from a finding that the FDA tended to down- of a diminution in drug safety from faster new drug play risks while approving Avandia and monitoring its approvals since 1992 (the report later ignored that evi- dence, however). The IOM authors also could havecited an estimate that the advent of user fees saved at Are New Drugs Riskier than Old Ones? Running
least several times as many lives through faster drug through the drug safety debate is a common assumption approvals as might have been lost through safety prob- that new drugs tend to be riskier than old ones. The IOM report, for example, recommends that all newlyapproved drugs and indications carry a prominent black Running through the drug safety debate is triangle, and that direct-to-consumer advertising berestricted for the first two years (NAS 2007, p. S-10).
This attitude has little basis. Certainly it is true that much of the most important safety information aboutdrug risks is revealed only after FDA approval, but thisdoes not mean that new drugs are riskier. They are often What is far more important—but was ignored by the designed (successfully) to avoid the side effects caused by IOM—is the biased incentive structure that causes the older drugs. The first non-sedating antihistamines, for FDA staff to put too much weight on drug safety rather example, probably began to prevent fatal automobile than too little. The avalanche of criticism thrown at the accidents as soon as they came into use (Weiler et al.
agency since 2004 is the latest illustration of the single 2000). Also relevant is that many new drugs are studied most powerful and enduring force impinging upon the more intensely than older ones both before and soon FDA staff: a profound disparity in how types of errors are after approval. The effect is to reveal more risk informa- penalized. When deciding whether the benefits of a pro- tion faster, which can make newer drugs appear less safe posed new drug exceed its risks, FDA staff know that if relative to older ones than they really are. In each of the they commit what is often called a Type I error—the three major drug safety episodes that captured the atten- approval of a drug that turns out to be insufficiently safe tion of Congress (Vioxx, SSRI antidepressants, and once marketing begins—their error will usually become Avandia, discussed below), the safety crisis was triggered known (a “public error”). This can and often does lead mainly by post-approval clinical trials, and additional to impassioned criticism of the agency and correction of research and analysis soon revealed either a lack of a the error (although more often than not, critics fix upon safety problem (in the case of SSRIs) or risks for the something that was probably not an error at all). On the entire classes to which these drugs belonged. Thus, other hand, a Type II error—the failure to permit mar- slowing down the uptake of newer drugs would not keting of a drug that would in fact provide benefits in necessarily improve safety, although it would sacrifice excess of harms—is typically detected by relatively few people (a “private error”), and its deleterious effects canpersist more or less indefinitely.
The FDA Problem That Was Not There
The net effect of this asymmetry in publicity over Type I and Type II errors is to bias even the best- The FDA is an immensely powerful bureaucracy with intentioned FDA regulators toward excessive caution islands of transparency (cf. advisory committee meeting and excessive drug testing. Research on the “drug lag” of the 1960s and 1970s, when FDA approvals trailed far accepted views of such basic matters as, for example, the behind those in European nations, revealed no consumer benefits of lowering serum cholesterol or the safety of all benefit in terms of safer drugs (Peltzman 1973, 1974; NSAID pain relievers (Topol 2004a; Wadman 2007). Wardell and Lasagna 1975). Subsequent research FDAAA is best seen as fixing upon a target of oppor- revealed that similar patterns persisted long after tunity. The relative ease with which post-approval drug (Katin and Brown 1995). Yet slow drug approvals here safety problems become widely known (“public errors”) did not bring extra safety. An analysis of the United guarantees a steady drumbeat of criticism and proposals States, Spain, and the United Kingdom yielded essen- for reform (for example, Friedman et al. 1999; Wood, tially identical drug-withdrawal rates despite the more Stein, and Woosley 1998; Kleinke and Gottlieb 1998; rapid drug-approval timelines in the European countries Moore, Psaty, and Furberg 1998). Although there is lit- tle, if any, systematic evidence of real declines in FDAscrutiny or effectiveness in dealing with drug safety,individual episodes can create pressure for change, as occurred after the Vioxx withdrawal and the SSRI anti-depressant safety debate. In this context, the prospect of drug safety provisions of FDAAA will work user-fee renewal in 2007 provided a potent vehicle for FDA reform. It is no surprise that in its final paragraph,the IOM report proclaimed: “Now is the time to renew and transform Center for Drug Evalution and Research’s ultimately, the welfare of patients.
culture, its authorities, its scientific capacity, and its abil-ity to communicate with health care providers and thepublic.” Chiming in were the New England Journal of There is considerable anecdotal evidence that the Medicine (Hennesy and Strom 2007) and the American FDA quickly became even more cautious in approving Medical Association’s Archives of Internal Medicine new drugs and indications after the Vioxx uproar began in October 2004 (Harris 2005). Earlier this year, forexample, the FDA refused to approve the pain reliever The Perils of Passing Unnecessary
Arcoxia and the weight-loss drug Accomplia even FDA Reform
though both had been approved by the European Unionand many other nations (Gottlieb 2007b [April 17]; Trying to fix something that is not broken provides a Wadman 2007). The FDA has also been unreceptive to reliable path to unintended consequences. There are some promising new drugs for advanced cancer, includ- many reasons to think that the drug safety provisions ing Provenge, Genasense, and others (Usdin 2007a; of FDAAA will work badly from the standpoint of drug Miller and Henderson 2007; Miller 2007).
development, new drug approvals, and ultimately, the Neither Congress nor the IOM report has paid much attention to another potent force: market-driven manu-facturer incentives to maintain drug safety. Such incen- Unnecessary Additions to FDA Power. FDAAA will
tives operate with powerful effect in far less regulated increase FDA authority in a number of ways. This is high-tech industries such as automobiles, petroleum, and unfortunate because the FDA can easily persuade pharma- electronics. As in other industries, pharmaceutical ceutical firms to do virtually anything it asks, up to and manufacturers rely heavily upon maintaining their repu- including the imposition of severe warnings, controls tation among customers (especially physicians) for prod- over prescribing, cessation of advertising campaigns, and uct safety and efficacy. Post-approval clinical trials play a the removal of drugs. The FDA will be expected to use central role in this process. These trials are undertaken the new powers granted by FDAAA. The staff will be to expand markets, but they necessarily open the door to under considerable pressure to take stronger measures new and possibly alarming (as well as reassuring) safety even when those measures would be inappropriate in the information. Often, post-approval trials are bigger, complex and obscure circumstances that surround virtu- longer, and more informative than the trials under- ally all important regulatory decisions. An example is girding drug approvals. Often, they force revisions in the power to require manufacturers to complete so-called phase 4 trials for newly approved drugs. The IOM report drug development (as pointed out by Brett 2007). Critics (pp. 4-8–4-9) explained very clearly why last-minute who support FDAAA or stronger measures also want the negotiations in the drug-approval process often cause the FDA to avoid approving drugs that are only on a par with FDA and manufacturers to agree to post-approval trials (rather than superior to) existing drugs (Avorn 2005).
that turn out to be unnecessary or even harmful. FDAAA They applauded when the FDA refused to approve Prexige, will worsen the dilemma already faced by FDA when they a Cox-2 pain reliever for arthritis, on exactly those grounds.
are criticized for failing to do things that should not havebeen committed to in the first place.
Making Clinical Trial Data Public. FDAAA will expand
upon a recent trend toward revealing unpublished clini-
Inappropriate Responsibilities for Post-Approval Drug
cal trial results to which the FDA staff is already privy.
Use and Safety. FDAAA will be a powerful force in mak-
This public registry of undigested clinical trial data will ing the FDA responsible for post-approval drug use and probably reinforce the tendency for FDA staff to give safety. Today’s question—“Why didn’t the FDA catch the disproportionate weight to risks vs. benefits of drugs. It problem with that drug?”—will be replaced by a very dif- will also tempt medical journals and researchers to cherry- ferent one: “Why did the FDA let doctors do that with pick the data to support political agendas, at the cost that drug?” This will trigger a series of unfortunate actions.
of alarming the public and scaring patients away from Faced with the possibility of being blamed when physi- ongoing clinical trials. Finally, it can undermine R&D cians and others fail to use drugs as it had intended, the incentives by alerting competitors to R&D and market- FDA will have to be far more intrusive in its regulation of ing strategies (Lassman 2006). This provision may turn medical practice, but the FDA possesses few tools and out to be far more important and troublesome than one even less knowledge or other preparation for regulating would infer from its noncontroversial status in the how doctors go about their work. Forced again to choose among unattractive alternatives, the agency will employcrude measures such as onerous tracking requirements and FDA Reform is a One-Way Street. Finally, we should
restrictions on who may treat certain patients.
give considerable weight to the fact that this FDA All this is a logical result of what the IOM report and reform will probably be a one-way street. FDAAA will many other FDA reform advocates have propounded as increase FDA power, extend its reach beyond normal the necessity of a “life-cycle” approach to drug safety bounds, and expose FDA personnel to yet more scrutiny monitoring and regulation. This principle offers virtually and criticism for safety problems no matter how unpre- no guidance beyond the general principle of a steady dictable. There will be no tests or benchmarks for how expansion of FDA authority and responsibility. We have well this new regime will work. In a world of ever more to assume that if the FDA is to be held responsible for extensive post-approval clinical trials and database whether especially risky drugs are used properly (as sug- dredging, there is no reason to think drug safety data will gested, for example, by the IOM report’s recommendation become more reassuring or less alarming as time passes.
5.1 and 5.2), it must be able to exercise commensurate The FDA will find it very difficult to retreat from any power over physicians and other health care providers.
The impetus for regulating medical practice will also Perhaps the only hope is that the FDA, whose arise from a more general aspect of FDAAA. It will resources include some leaders of exceptional ability, reinforce and strengthen the bias toward preventing will take due account of its inability to regulate usefully Type I errors (a safety problem with an approved drug) medical practice and of the ease with which the agency at the cost of Type II errors (such as moving too slowly can slow down the most innovative kind of R&D. If the to facilitate new drug development and approval). For FDA leadership proceeds with an exceptional measure example, a crucial issue in adapting regulation to scientific of wisdom and restraint, they may be able to prevent advances is revising and expanding the use of surrogate FDAAA from doing great harm. We shall see.
markers (such as LDL cholesterol levels rather than heart AEI research assistant Walton Dumas and web editor Laura Drink- attacks for cardiovascular drugs, or glycemic levels for wine worked with Mr. Calfee to edit and produce this Health Pol- diabetes drugs). Many FDA critics see reform as a tool for icy Outlook. Additional research assistance was provided byRobert Ward, an AEI intern and George Washington University reining in or nearly eliminating surrogate markers (Psaty medical student who attended the July 30, 2007, joint meeting of and Furberg 2007a), which could drastically retard new References
myocardial infarction and cardiovascular death. Annals of Avorn, Jerry. 1995. Editorial: Drug regulation and drug infor- Diamond, George A., and Sanjay Kaul. 2007. Letter: Rosiglita- mation—who should do that to whom? American Journal of zone and cardiovascular risk. New England Journal of Medi- Public Health 85 (1): 18–19.
———. 2005. FDA standards—Good enough for government Fontanarosa, Phil B., Drummond Rennie, and Catherine D.
work? New England Journal of Medicine 353 (10): 969–72.
DeAngelis. 2004. Editorial: Postmarketing surveillance: ———. 2007. Paying for drug approvals: Who’s using whom? Lack of vigilance, lack of trust. Journal of the American New England Journal of Medicine. April 13.
Medical Association 292: 2647–50.
Bakke, Olav M., Michael Manocchia, Francisco de Abajo, Friedman, Michael A., Janet Woodcock, Murray M. Lumpkin, Kenneth I. Kaitin, and Louis Lasagna. 1995. Drug safety dis- Jeffrey E. Shuren, Arthur E. Hass, and Larry J. Thompson.
continuations in the United Kingdom, the United States, 1999. The safety of newly approved medicines: Do recent and Spain from 1974 through 1993: A regulatory perspec- market removals mean there is a problem? Journal of the tive. Clinical Pharmacology and Therapeutics 58: 108–17.
American Medical Association 281: 1728–34.
Berndt, Ernst R. 2006. The United States experience with direct- Furberg, Curt D., Arthur A. Levin, Peter A. Gross, Robyn S.
to-consumer advertising of prescription drugs: What have we Shapiro, Brian L. Strom. 2006. The FDA and drug safety: learned? in Frank A. Sloan and Chee-Ruey Hsieh, eds., Pro- A proposal for sweeping changes. Archives of Internal Medi- moting and coping with pharmaceutical innovation: An interna- tional perspective. Cambridge: Cambridge University Press.
Fuster, Valentin, and Michael E. Farkouh. 2007. Editorial: Boodman, S. G. 2006. Many drug uses don’t rest on strong Faster publication isn’t always better. Nature Clinical Prac- science. Washington Post. May 23.
Bracken, Michael B. 2007. Letter: Rosiglitazone and cardio- Gibbons, Robert D., C. Hendricks Brown, Kwan Hur, Sue M.
vascular risk. New England Journal of Medicine 357 (9): Marcus, Dulal K. Bhaumik, Joëlle A. Erkens, Ron M. C.
Herings, and J. John Mann. 2007. Early evidence on the Brent, David. 2007. Editorial: Antidepressants and suicidal effects of regulators’ suicidality warnings on SSRI prescrip- behavior: Cause or cure? American Journal of Psychiatry 164 tions and suicide in children and adolescents. American Journal of Psychiatry 164: 1356–63.
Brett, Allan S. 2007. Letter: Rosiglitazone and cardiovascular Gottlieb, Scott. 2007a. Drug safety proposals and the intrusion risk. New England Journal of Medicine 357 (9): 937–40.
of federal regulation into patient freedom and medical Bridge, Jeffrey A., Satish Iyengar, Cheryl B. Salary, Rémy P.
practice. Health Affairs 26 (3): 664–77.
Barbe, Boris Birmaher, Harold Alan Pincus, Lulu Ren, and ———. 2007b. Drug danger. Wall Street Journal. April 17.
David A. Brent. 2007. Clinical response and risk for ———. 2007c. Journalistic malpractice. Wall Street Journal.
reported suicidal ideation and suicide attempts in pediatric antidepressant treatment: A meta-analysis of randomized Hampton, Tracy. 2007. MI risks linked to Rosiglitazone. Jour- controlled trials. Journal of the American Medical Association nal of the American Medical Association 298 (10): 1149–51.
Harris, Gardiner. 2005. F.D.A. responds to criticism with new Calfee, John E. 2005. The Vioxx fallout. AEI Health Policy caution. New York Times. August 6.
Hennessy, Sean, and Brian L. Strom. 2007. Perspective: ———. 2006. Playing catch-up: The FDA, science, and drug PDUFA reauthorization—Drug safety’s golden moment of regulation. AEI Health Policy Outlook. March.
opportunity? New England Journal of Medicine. April 13.
———. 2007. An assessment of direct-to-consumer advertis- Home, Philip D., et al., for the RECORD Study Group. 2007.
ing of prescription drugs. Clinical Pharmacology and Thera- Rosiglitazone evaluated for cardiovascular outcomes—an interim analysis. New England Journal of Medicine 357 (1): Curfman, Gregory D., Stephen Morrissey, and Jeffrey M.
Drazen. 2006. Editorial: Blueprint for a stronger Food and Kaitin, Kenneth, and Jeffrey Brown. 1995. A drug lag update.
Drug Administration. New England Journal of Medicine 355 Drug Information Journal 29 (2): 361–74.
Kearney, Patricia M., Colin Baigent, Jon Godwin, Heather Diamond, George A., Leon Bax, and Sanjay Kaul. 2007. Per- Halls, Jonathan R. Emberson, and Carlo Patrono. 2006. Do spective: Uncertain effects of Rosiglitazone on the risk for selective cyclo-oxygenase-2 inhibitors and traditional non-steroidal anti-inflammatory drugs increase the risk of Administration drug advisory committee meetings. Journal atherothrombosis? Meta-analysis of randomised trials.
of the American Medical Association 295 (16): 1921–28.
British Medical Journal 332: 1302–08.
Mannucci, Edoardo, Matteo Monami, and Niccolò Kessler, David A. 1978. Regulating the prescribing of human Marchionni. 2007. Letter: Rosiglitazone and cardiovascular drugs for nonapproved uses under the Food, Drug and Cos- risk. New England Journal of Medicine 357 (9): 937–40.
metic Act. Harvard Journal on Legislation 15 (4): 693–760.
Marks, Harry M. 1995. Revisiting the origins of the compul- Kleinke, J. D., and Scott Gottlieb. 1998. Editorial: Is the FDA sory drug prescriptions. American Journal of Public Health 85 approving drugs too fast? British Medical Journal 317: McClellan, Mark. 2007. Drug safety reform at the FDA— Kohn, David, and Jonathan Bor. 2004. Scientist’s warnings on pendulum swing or systematic improvement? New England drugs stir debate: Some doctors question FDA’s assessment Journal of Medicine. April 13.
of 5 popular prescriptions. Baltimore Sun. November 19.
McKeown, Robert E., Steven P. Cuffe, and Richard M. Schulz.
Krall, Ronald L. 2007. Cardiovascular safety of Rosiglitazone.
2006. U.S. suicide rates by age group, 1970–2002: An examination of recent trends. American Journal of Public Kravitz, R. L., Ronald M. Epstein, Mitchell D. Feldman, Carol E. Franz, Rahman Azari, Michael S. Wilkes, Ladson Hin- Medical News Today. 2006. Black box warning violations in ton, and Peter Franks. 2005. Influence of patients’ request outpatient settings putting patients at risk. February 16.
for direct-to-consumer advertised antidepressants. Journal of Cited in Institute of Medicine FDA drug safety report on the American Medical Association 293 (4): 1995–2002.
February 16 as Public Health Newswire.
Laine, Christine, et al. 2007. Editorial: Clinical trial registration— Miller, Henry I., and David R. Henderson. 2007. Govern- looking back and moving ahead. New England Journal of mental influences on drug development: Striking a better Medicine 356 (26): 2734–36.
balance. Nature Reviews Drug Discovery 6: 532–39.
Lancet. 2007. Editorial: Rosiglitazone: Seeking a balanced per- Miller, Richard. 2007. Cancer regression. Wall Street Journal.
Lasser, Karen E., Diane L. Seger, D. Tony Yu, Andrew S.
Moore, Thomas J., Bruce M. Psaty, and Curt D. Furberg. 1998.
Karson, Julie M. Fiskio, Andrew C. Seger, Nidhi R. Shah, Time to act on drug safety. Journal of the American Medical Tejal K. Gandhi, Jeffrey M. Rothschild, and David W.
Bates. 2006. Adherence to black box warnings for prescrip- Institute of Medicine, Committee on the Assessment of the tion medications in outpatients. Archives of Internal Medi- U.S. Drug Safety System. 2006. The future of drug safety: Promoting and protecting the health of the public. Washington, Lassman, Scott M. 2006. Transparency and innuendo: An alternative to reactive overdisclosure. Law and Contempo- ———, Committee on Identifying and Preventing Errors.
Philip Aspden, Julie Wolcott, Linda R. Cronenwett, eds.
Lincoff, A. Michael, Kathy Wolski, Stephen J. Nicholls, and 2007. Preventing medication errors. Washington, DC: Steven E. Nissen. 2007. Pioglitazone and risk of cardiovas- cular events in patients with type 2 Diabetes mellitus: A Nature Biotechnology. 2007. Editorial, Probity gone nuts: The meta-analysis of randomized trials. Journal of the American thinking behind the US Food and Drug Administration’s Medical Association 298 (10): 1180–88.
draft guidance on financial conflicts of interest for outside Lubell, K. M., S. R. Kegler, A. E. Crosby, and D. Karch. 2007.
experts is deeply flawed. 25 (5): 483.
Suicide trends among youths and young adults aged 10–24 New York Times. 2006. Editorial: Prescription for a stronger years—United States, 1990–2004. Morbidity and Mortality Weekly Reports 56 (35): 905–08.
Nissen, Steven E., and Kathy Wolski. 2007. Effect of Rosiglita- Ludwig, Jens, Dave E. Marcotte, and Karen Norberg. 2007.
zone on the risk of myocardial infarction and death from car- Anti-depressants and suicide. NBER Working Paper diovascular causes. New England Journal of Medicine. May 21.
No. 12906, National Bureau of Economic Research, Okie, Susan. 2005. What ails the FDA? New England Journal of Lurie, Peter, Cristina M. Almeida, Nicholas Stine, Alexander Pardoll, Drew, and James Allison. 2004. Cancer immuno- R. Stine, and Sidney M. Wolfe. 2006. Financial conflict of therapy: Breaking the barriers to harvest the crop. Nature interest disclosure and voting patterns at Food and Drug Peltzman, Sam. 1973. An evaluation of consumer protection inadequate response to the IOM. New England Journal of legislation: The 1962 drug amendments. Journal of Political Solomon, Daniel H., and Wolfgang C. Winkelmayer. 2007.
———. 1974. Regulation of pharmaceutical innovation: The 1962 Editorial: Cardiovascular risk and the Thiazolidinediones: amendments. Washington, DC: AEI Press.
Déjà vu all over again? Journal of the American Medical Pfeffer, Cynthia R. 2007. Editorial: The FDA pediatric Association 298 (10): 1216–18.
advisories and changes in diagnosis and treatment of pedi- Strom, Brian L. 2006. How the US drug safety system should atric depression. American Journal of Psychiatry 164: be changed. Journal of the American Medical Association 295 Philipson, Tomas J., Ernst R. Berndt, Adrian H. B. Gottschalk, Topol, Eric J. 2004a. Editorial: Intensive Statin therapy: A sea and Matthew W. Strobeck. 2005. Assessing the safety and change in cardiovascular prevention. New England Journal efficacy of the FDA: The case of the prescription drug user of Medicine 350 (15): 1562–64.
fee acts. NBER Working Paper 11724, National Bureau of ———. 2004b. Failing the public health: Rofecoxib, Merck, and the FDA. New England Journal of Medicine 351 (17): Psaty, Bruce M., and Curt D. Furberg. 2005. Cox-2 inhibitors—lessons in drug safety. New England Journal of Usdin, Steve. 2006a. Ketek politics. BioCentury. June 26, A1.
———. 2006b. Science friction. BioCentury. December 18, ———. 2007a. Editorial: Rosiglitazone and cardiovascular risk. New England Journal of Medicine 356: 2522–24.
———. 2007a. Blockbusted at ODAC. BioCentury. May 14, A1.
———. 2007b. Editorial: The record on Rosiglitazone and the ———. 2007b. Political defibrillator. BioCentury. May 28, risk of myocardial infarction. New England Journal of Medi- ———. 2007c. Scripting Avandia suits. BioCentury. July 23, Radley, David C., Stan N. Finkelstein, and Randall S.
Stafford. 2006. Off-label prescribing among office-based ———. 2007d. A vote against panic. BioCentury. August 7, physicians. Archives of Internal Medicine 166: 1021–26.
Shogren, Elizabeth. 2004. FDA sat on report linking suicide, Wadman, Meredith. 2007. The pain game. Nature 448: 400–01.
drugs: Officials ordered more studies after their own Wardell, William M., and Louis Lasagna. 1975. Regulation expert found children on antidepressants were twice as and drug development. Washington, DC: AEI Press.
likely to show suicidal behavior. Los Angeles Times. Weiler, John M., John R. Bloomfield, George G. Woodworth, Angela R. Grant, Teresa A. Layton, Timothy L. Brown, Simon, Gregory E. 2006. The antidepressant quandary— David R. McKenzie, Thomas W. Baker, and Ginger S. Wat- considering suicide risk when treating adolescent son. 2000. Effects of Fexofenadine, Diphenhydramine, and depression. New England Journal of Medicine 355: alcohol on driving performance: A randomized, placebo- controlled trial in the Iowa driving simulator. Annals of Inter- Singh, Sonal, Yoon K. Loke, and Curt D. Furberg. 2007. Long- nal Medicine 132 (5): 354–63.
term risk of cardiovascular events with Rosiglitazone: A Wood, Alastair, C. Michael Stein, and Raymond Woosley.
meta-analysis. Journal of the American Medical Association 1998. Making medicines safer: The need for an independ- ent drug safety board. New England Journal of Medicine 339 Smith, Sheila Weiss. 2007. Sidelining safety—the FDA’s

Source: http://www.1789mass.com/files/2007/09/26/20070927_HPOcorrected_3.pdf