Design and Reporting Modifications in Industry-Sponsored Comparative Psychopharmacology Trials DANIEL J. SAFER, M.D.1 This review of recently published pharmaceutical industry–sponsored comparative psychotropic drug trials aims to classify apparent design and reporting modifications that favor the sponsor’s product. The modifications have been grouped into 13 discrete categories, and representative examples of each are presented. Strong circumstantial evidence suggests that marketing goals led to these adjustments. The consequences of marketing influences on comparative psychopharmacology trials are discussed in terms of conflicts of interest, the integrity of the scientific literature, and costs to consumers, as well as their impact on physician practice. —J Nerv Ment Dis 190:583–592, 2002 Occasionally, a pharmaceutical product enters the market that has some distinct advantages over its predecessors. Soon thereafter, other companies manufacture and patent a similar drug, referred to as a me-too drug. Then, an intense competition for market share develops for which the drug’s price is not the central issue (Kessler et al., 1994; Tarabusi and Vickery, 1998). An important aspect of this com- petition is the publication of clinical trials showing the superiority of one drug over another that has a niche in the same market. The purpose of that re- search is not only to show the relative superiority of the company’s product, but also to show the draw- backs of the competing drug. The resulting publication of pharmaceutical com- pany–sponsored comparative drug research now oc- cupies a big share of the medical literature. It is fueled by marketing aims, so it is no wonder that an estimated 89 to 98% of comparative drug treatment studies funded by pharmaceutical companies yield results that are favorable to their company’s product (Cho and Bero, 1996; Davidson,1986). Furthermore, such sponsored research generally shows their product to be safer than its rival (Mandelkern, 1999; Rochon et al., 1994b). To achieve the aim of demonstrating comparative superiority, certain design adjustments and reporting modifications are frequently utilized. These can in- clude the following: 1) using a dose of the comparable drug that is outside of the standard clinical range, 2) altering the usual dosing schedule of the competing drug, 3) using misleading research measurement scales, 4) picking endpoints post hoc, 5) masking un- favorable side effects, 6) repeatedly publishing the same or similar findings for impact, 7) selectively high- lighting findings favorable to the sponsor, 8) editorial- izing in the abstract, 9) publishing the obvious, 10) statistical obfuscation, 11) selecting subjects and a time frame designed to achieve a favorable outcome, 12) withholding unfavorable results, and 13) masked sponsorship. These research modifications will be presented in this review in more detail with examples. Then, the broader and untoward ramifications of competitive drug company–sponsored clinical research will be presented. This review of the psychopharmacology literature, primarily from the U.S., is selective and does not aim to present a comprehensive or a balanced perspec- tive of all pharmaceutical company–sponsored psych- otropic research. Some have argued that bias in the literature is not infrequent and includes placebo- controlled registration trials and some government- sponsored research (Conley, 2001; Gorelick,1998; Moynihan, 2002). Nonetheless, this report focuses on industry-sponsored comparative trials because these contain far more numerous and prominent examples of design and reporting modifications and because industry-sponsored studies now dominate the compar- ative drug trial literature (Angell, 2001; Davidoff, 2002). Design Adjustments and Reporting Modifications Using Doses Outside the Usual Range for Competitive Advantage An obvious example of drug company–sponsored research designed to prove a point is that of a sec- 1Departments of Psychiatry and Pediatrics, Johns Hopkins University School of Medicine, Baltimore, 7702 Dunmanway, Dundalk, MD 21222. Send reprint requests to Dr. Safer. 0022-3018/02/1909 –583 Vol. 190, No. 9 THE JOURNAL OF NERVOUS AND MENTAL DISEASE Printed in U.S.A. Copyright © 2002 by Lippincott Williams & Wilkins DOI: 10.1097/01.NMD.0000030522.74800.0D 583 ond-generation neuroleptic compared to a relatively high dose of haloperidol. At least eight studies spon- sored by three different drug companies have com- pared their second-generation neuroleptic drug to a fixed high dose of haloperidol of 20 mg per day (Chouinard et al., 1993; Deo et al., 1990; Emsley 1999; Goldstein et al., 1999; Marder and Meibach, 1994; Simpson and Lindenmeyer, 1997) or to an av- erage haloperidol dose greater than 20 mg per day (Lapierre et al., 1990; Patris et al., 1990). Using such an unusually high dose virtually ensures that the second-generation product will have fewer extrapy- ramidal side effects (EPS) than haloperidol. In com- parative studies between second-generation neuro- leptics, unusually high fixed or average doses of risperidone (7.2 and 8 mg/day) have also been used to achieve similar results (Peuskens et al., 1999; Tran et al., 1997b). Doses of haloperidol exceeding the customary lev- els of 4 to 10 mg/day produce no better clinical results than do doses above that range (Baldessarini et al., 1988; Mc Evoy et al., 1991; Rifkin et al., 1991), but they induce more EPS (Mc Evoy et al., 1991; Rosebush and Mazurek, 1999: Stone et al., 1995; Wong et al., 1999) and lead to far more treatment drop-outs (Beasley et al., 1999; Chouinard et al., 1993; Tollefson et al., 1997b; Van Putten et al., 1990). Likewise, doses of risperidone above 6 mg/day pro- duce more side effects and no greater therapeutic results than standard doses of 4 to 5 mg/day (Love et al., 1999; Nyberg et al., 1999). Furthermore, a recent meta-analysis clearly revealed that doses of haloper- idol greater than 12 mg/day produced significantly fewer favorable symptom scores than haloperidol at doses of less than 12 mg/day (Geddes et al., 2000). Substantially Altering the Dose Schedule of the Comparison Drug for Competitive Advantage Commercially sponsored studies comparing two antidepressant drugs often schedule an unusually rapid and substantial dose increase in the one not manufactured by the sponsoring company. For ex- ample, in comparative studies of antidepressants, the dose of fluoxetine was increased to 40 mg/day in over one half of the subjects at 3 weeks (Geerts et al., 1999), to an average of 47 mg/day during weeks 4 to 8 (Rudolph and Feiger, 1999), and to an average of 40 mg/day in 10 of the 21 subjects at 4 weeks (Armitage et al., 1997). Likewise, sertraline was in- creased to an average of 148 mg/day at 6 weeks (Feiger et al., 1996), and paroxetine was increased to 50 mg/day in one third of the cases by 7 weeks (Kiev and Feiger, 1997). Doses and dose schedules beyond the usual range, particularly early in treatment, char- acteristically bring an increased rate of side ef- fects—as was the case in the studies referred to above. Sometimes in comparative studies, the company- supported protocol used a relatively low dose of its own sponsored drug or the competing drug to accen- tuate efficacy or side effect differences from its rival. Examples include a slightly high but reasonable mean dose of 15.7 mg/day of olanzapine compared with a clearly substandard mean dose of 201 mg/day of clo- zapine for weeks 2 through 18 (Bitter et al., 2000), and a low mean dose of 102 mg/day of fluvoxamine compared to a moderately high mean dose of 34 mg/day of fluoxetine after week 7 (Rapaport et al., 1996). Another skewed design can be achieved by altering the customary timing of the drug adminis- tration. For example, in one study comparing parox- etine with amitriptyline, the amitriptyline was ad- ministered twice daily, which led to a prominent degree of daytime sleepiness (Christiansen et al., 1996). Using Self-Serving Measurement Scales and Making Misleading Conclusions From Measurement Findings Some researchers use unpublished rating scales, and as a group, these are said to constitute “a major source of bias in randomized controlled clinical tri- als” (Marshall et al., 2000). Clearly, industry-spon- sored researchers use rating scales that will attain the most favorable result for their product (Sheehan, 1999). In studies of risperidone versus placebo and haloper- idol, drug company–sponsored researchers utilized the “worst EPS score” to compare differential EPS changes (Chouinard et al., 1993; Lemmens et al., 1999; Marder and Meibach, 1994; Simpson and Lindenmeyer, 1997). They reported that after a brief washout period, risperidone at doses of 2 to 6 mg/day given to chroni- cally medicated schizophrenic patients evidenced the same degree of EPS as placebo. The company’s mar- keting personnel widely circulated this result (Janssen Pharmaceutica Research Foundation, 1996; Borison et al., 1992), which misled many to assume that risperi- done at doses of 2 to 6 mg/day did not cause EPS, because obviously a placebo given to treatment-naive subjects would not cause them. However, risperidone can indeed cause EPS in customary doses (Ho et al., 1999; Lane et al., 2000; Mullen et al., 1999; Rosebush and Mazurek, 1999). Using a “worst EPS score” can give quite different and potentially misleading results compared to final total EPS scores or last recorded SAFER584 observations (Tollefson et al., 1997a). Fortunately, most investigators at this time now use total and last observation scores to evaluate EPS rating changes from baseline (Bondolfi et al., 1998; Marder et al., 1997; Wirshing et al., 1999). Another example involves defining treatment suc- cess using a 20% or greater decline in a symptom severity rating score. Patients with schizophrenia may achieve a 20% decline in the Brief Psychiatric Rating Scale scores and yet remain “profoundly dis- abled” (Gilbody et al., 2002). A further example in- volves defining treatment success as at least a 25% decline from baseline on a rating scale score for symptom severity for youths with attention deficit hyperactivity disorder (ADHD). In an industry-spon- sored study, the majority of the groups that received the company product and methylphenidate achieved this goal and were thus deemed equivalent in their response. However, the cutoff score was so modest that 32% of the placebo subjects also met this stan- dard (Heilgenstein et al., 2000), which was more than double the customary placebo response in ADHD stimulant studies (Greenhill et al., 2001). Pre- sumably, if a standard symptom reduction cutoff score had been used, the reported drug equivalency would not have been achieved. Selecting the Major Findings and Endpoints Post Hoc Kessler (1992) refers to comparative drug studies wherein “the putative advantage is related to an endpoint that was not the primary hypothesis tested by the study” (p. 950). He adds, “such claims of superiority often depend on an analysis of multiple endpoints, a practice often referred to as data dredging.” Thus, it is unclear whether a seven-week study had been originally designed to be an eight- week study. More suggestive is the side effect com- parison of venlafaxine and fluoxetine that was done one week into treatment (Silva, 1998), which sug- gests that something was left out that would apply to a more meaningful assessment period. As Schooler (2000) pointed out in her review of two industry-sponsored comparative drug trials, “perhaps in each case the researchers examined all the scores and reported only those that looked best for their drug” (p. 13). Carpenter (2001) likewise noted that industry studies “showing a superior effect of the spon- sor’s drug may be the result of scores of analyses searching multiple variables and reporting those that show an advantage” (p. 13). Tamminga (2000) similarly observed that in industry-sponsored stud- ies “the distinction between the a priori hypothesis and the secondary analyses is often lost” and a “whole rash of secondary analyses are used to present the results” (p. 5). Masking Unfavorable Side Effects Reports of sexual side effects from selective sero- tonin reuptake inhibitor (SSRI) antidepressants range from 2% to 73% depending primarily on whether side effects were elicited merely by open- ended questioning or by a detailed inquiry (Modell et al., 1997; Montejo-Gonzalez et al., 1997; Zajecka et al., 1999). Reports by sponsoring pharmaceutical companies tend to downplay sexual side effects of SSRIs by using open-ended or nonspecific questions about side effects (Zajecka et al., 1999). In one in- stance, a drug company–sponsored review covering over 3,000 subjects treated with SSRIs simply did not list any sexual side effects on its 23-item side effect table (Preskorn, 1997). At the opposite ex- treme, researchers from competing companies em- phasize the high rate of sexual side effects with SSRIs when their drug has a lower rate. These in- vestigators ask specific questions in order to bring out the high rate of sexual side effects that charac- terize SSRI treatment so as to favor their product’s side effect profile (Croft et al., 1999). Repeatedly Publishing the Same or Similar Positive Studies to Increase the Impact Obviously, pharmaceutical manufacturers want to make sure that reports of their drug’s advantage over competing drugs in sponsored research receive wide circulation. Their supported researchers consequently submit relatively similar findings to different journals, each with a slightly different emphasis but stressing the main themes. They also publish reviews of these same findings (Gilbody and Song; 2000; Huston and Moher, 1996). Examples repeatedly showing an EPS advantage with one second-generation neuroleptic over high-dose haloperidol include prominently over- lapping articles by one pharmaceutical company–sup- ported team (Beasley et al., 1999; Lane et al., 2000; Tollefson et al., 1997b; Tran et al., 1997a,1998) and a very similar duplication of articles from another firm’s second-generation neuroleptic—which is well described by Huston and Moher (1996). Still another pharmaceutical company followed this same pat- tern, sponsoring five studies comparing sertraline or fluoxetine with bupropion SR in order to reiterate the point about their drug having fewer sexual side effects than SSRI drugs (Coleman et al., 1999; Gil- DESIGN AND REPORTING MODIFICATIONS 585 body and Song, 2000; Kavoussi et al., 1997; Segraves et al., 2000; Walker et al., 1993). An editorial in JAMA by Rennie (1999) discussed in depth this problem of publication duplication. Selectively Highlighting Findings Favorable to the Sponsor The selective reporting of the findings in industry- sponsored trials has merited criticism from experi- enced investigators (Carpenter, 2002; Davidoff et al., 2001). In this regard, Schooler is quoted as follows: “They emphasize those parts of the data that are most favorable to their drug” (Moukheiber, 2001, p. 268). As an example, in the placebo-controlled clin- ical trial comparing the effect of two antidepres- sants with each other to reduce anxiety symptoms, the abstract totally ignored the fact that the placebo effect for anxiety was in most respects comparable to that of the two antidepressants and that it re- sulted in far fewer side effects (Trivedi et al., 2001). Likewise, two studies mentioned an early treatment response of the sponsor’s drug in their abstract but did not mention that this relative benefit was no longer present by the end of the study (Claus et al., 1992; Wheatley et al., 1998). Editorializing for the Sponsor in the Abstract In the seven-week study comparing fluvoxamine and paroxetine, the patients receiving paroxetine were given (at week 7) a substantial mean dose of 36 mg/day, whereas patients receiving fluvoxamine were given a relatively low mean dose (at week 7) of 102 mg/day. Certain side effects were understand- ably higher in the paroxetine group. In their conclu- sions, the authors advised, “when a patient has dif- ficulty tolerating one SSRI, the clinician may choose to change to a different agent within the same class” (Kiev and Feiger, 1997). In another study, risperidone treatment was shown to be comparable in efficacy and side effects to pimo- zide in the treatment of Tourette’s disorder. Two rela- tively minor differences favored risperidone, leading the authors to conclude that “risperidone may become the first-line treatment of Tourette’s disorder, owing to a more favorable efficacy and tolerability profile” (Bruggeman et al., 2001, p. 50). Publishing the Obvious to Emphasize a Point It is known from placebo-controlled clinical trials that certain second-generation antidepressants have far fewer sexual side effects than SSRIs. Thus, it is not necessary to repeatedly compare that side effect between these drugs; the result is obvious before- hand. Likewise, it has long been known that tricyclic antidepressants (TCAs) are a risk for older adults with heart disease (Roose, 1992). Thus, the compar- ison of paroxetine with nortriptyline in older adults showing more cardiovascular problems with the lat- ter drug (Nelson et al., 1999) does not add new information to the literature. In this respect, Shimm and Spece (1991) express concern about “scientifi- cally uninteresting and inconsequential clinical tri- als” (p. 150). Kessler et al. (1994) view such scien- tifically unnecessary trials as “thinly veiled attempts to entice doctors to prescribe a new drug being marketed by the company. . .” (p. 1351). Touting Nonsignificant but Favorable Differences and Negating Dropout Differences Statistically Medical journal supplements provide numerous other examples of studies with small and modest- sized samples that report an advantage of one drug over another but avoid analyzing the difference statistically (Bero et al., 1992; Greene et al., 2000; Hotopf et al., 1997; Rochon et al., 1994a). One exam- ple of this frequent pattern is from a paroxetine- clomipramine comparative study. A statistically in- significant difference was used to justify the author’s conclusion that paroxetine was “better tol- erated” (Lecrubier and Judge, 1997). One technique to accentuate a nonsignificant but supposedly favorable finding in a comparative study is to view it only in terms of relative risk (without pre- senting a range or confidence interval). This technique applied to a small sample often misleads and has re- portedly influenced physician prescribing patterns (Bobbio et al., 1994; Brett, 1989, Feinstein, 1992). When there is a substantial dropout of subjects before the end of a clinical trial, industry-sponsored researchers commonly use the last observation car- ried forward method to include the dropout data in the total results. This approach assumes a steady trajectory of the available findings from the drop- outs—which is seldom the case (Allison and Casey, 2001). Consequently, it is usually preferable to avoid this form of statistical analysis (Schooler, 2000). Selecting Subjects and Altering the Duration of Trials to Achieve a Favorable Outcome Treatment resistance in schizophrenia research is not well defined (Brenner et al., 1990). Comparisons of SAFER586 second-generation neuroleptics with one another and with haloperidol in short-term drug company–spon- sored clinical trials of so-called treatment-resistant schizophrenic patients have resulted in findings that suggest that a number of second-generation neurolep- tics (aside from clozapine) are very beneficial for these patients. Surprisingly, after only 6 and 8 weeks of treatment, 35% to 65% of patients so identified were found to respond to each of the new medications. However, these studies included a sizable number of patients whose treatment resistance was highly ques- tionable (Marder, 1999). Bondolfi et al. (1998), for ex- ample, included both treatment-resistant and treat- ment-“intolerant” patients, and one half of the patients studied by Breier and Hamilton (1999) were outpa- tients (presumably less emotionally disturbed than inpatients). The relatively brief clinical trials with olanzapine and risperidone suggested that these drugs were as effec- tive with treatment-resistant patients as clozapine, the present “gold standard” of such treatment (Conley et al., 1999a). However, it must be noted that clozapine treatment may take up to one year to achieve maxi- mum benefit (Lieberman et al., 1994). When drug treat- ment comparisons of an even longer duration (1 to 2 years) are made with chronically impaired schizo- phrenic patients, clozapine to an even greater extent produces the best results (Conley et al., 1999b). Withholding Unfavorable Results Drug company–sponsored clinical drug trials are for the most part not publicly recorded (Rennie, 1999), so that unfavorable trials can be kept from publication (Friedberg et al., 1999; Goldberg, 1999; Gwynne, 1999; Hillman et al., 1991; Lauritsen et al., 1987). This unfortunately is still a fairly common practice (Quick, 2001). The sponsoring company’s control of publication is aided by confidentiality agreements signed by the investigators before any project is financed (Goldberg, 1999; Gwynne, 1999; Dickersin et al., 1987; Rennie, 1997; Shenk, 1999). Masking Sponsorship In multisite clinical trial studies, the pharmaceuti- cal company designs the research— usually in con- sultation with some of the outside team—and then all the clinical investigators carry out the protocol (Council on Scientific Affairs, 1990). In some in- stances, the designers of the research do not put their names on the study, which makes the role of the pharmaceutical company less clear. This prac- tice has been referred to as “ghost authorship” (Flanagin et al., 1998; Rennie and Flanagin, 1994). Related to this concern is the fact that some authors still do not disclose the financial sponsorship of the research (Krimsky et al., 1998). Broader Drawbacks of Research Linked to Promotion Lowers the Scientific Quality of Published Research Friis (1999), in a lead editorial for Acta Psychiat- rica Scandinavica, wrote that “the need of a com- pany to increase sales of a new drug easily creates a climate in which it is virtually impossible for the company’s own research department to produce un- biased studies of that drug” (p. 157). The research department, in the view of some, might then “. . .rep- resent a subgroup of the marketing division, with the aim of producing scientific arguments for promoting sales.” Drug company–sponsored symposia (published as supplements) are far less carefully peer reviewed than standard research reports, and their papers are consistently appraised as more misleading than are others in the published medical literature (Bero et al., 1992; Cho and Bero, 1996; Rochon et al., 1994a). Leads Physicians to Alter Their Prescribing Options Studies show that drug company promotion of their own sponsored published research findings— often containing a selective presentation of the relative mer- its of their products and the drawbacks of competitive products—successfully influences physician prescrib- ing practices (Chren and Landefeld, 1994; Spingarn et al., 1996; Ziegler et al., 1995). The use of “design- adjusted,” drug company–sponsored medical publica- tions that superficially appear like more scientifically solid research has been shown specifically to mislead physicians, resulting often in the prescription of rela- tively expensive and sometimes less appropriate drugs (Bero et al., 1992; Caudill et al., 1996; Spingarn et al., 1996). Drug company–sponsored articles reporting positive clinical trials in medical journals frequently convey to physicians the impression that their prescribed phar- maceuticals produce more favorable results than is the case in community practice. For example, studies of brief and extended drug company–sponsored clinical trials with olanzapine (N �2000) report a dropout rate due to weight gain of less than one half of one percent (Littrell et al., 2000; Tollefson et al., 1997b; Tran et al., 1999), which suggests that it was not a problem. Yet a DESIGN AND REPORTING MODIFICATIONS 587 nationwide survey of 277 psychotic outpatients treated primarily with atypical neuroleptics reported that weight gain caused at least 25% of the medication nonadherence (Weiden et al., 2000), particularly for the obese, who were three times more nonadherent for this reason than were those of normal weight (Weiden, 2001). A second example of a misleading dependence on sponsored clinical trial data pertains to desmopres- sin for the treatment of nocturnal enuresis. In a com- pany-supported comparative clinical trial, both a bed- wetting alarm and desmopressin had relatively similar results during the treatment period, but no follow-up assessments were made (Longstaffe et al., 2000). How- ever, two independent studies reporting the aftermath of similar comparative trials revealed that 3 to 12 months after the treatment ended only 10% to 18% who were treated with desmopressin ceased their enuresis, versus 42% to 56% who had been treated with the alarm (Monda and Husmann, 1995: Schulman et al., 2000; Wille, 1986). Adds Unnecessarily to the Cost of Prescription Drugs A me-too drug is a drug that has slight chemical differences from a frequently prescribed innovator drug so that the latter’s patent is not infringed. Me- too drugs represent over 75% of all new drugs and are among the most heavily marketed (Garattini, 1997). The cost of marketing prescribed drugs in the U.S. was over 12 billion dollars in 1993, which ex- ceeds the pharmaceutical industry’s expenditure for basic research (Woosley, 1994). The cost of pharma- ceuticals has risen far more than other products in the U.S. economy and is a particular burden for those senior citizens whose only medical coverage is Medicare (Soumerai and Ross-Degnan, 1999). Tends to Have a Corrupting Effect on Researchers The enticement of making a large amount of money by participating in drug company–sponsored clinical trials is not to be underestimated (Eichenwald and Kolata, 1999a; Eichenwald and Kolata, 1999b). Along with this involvement and financial aid go ever-present pressures to endorse or support the company’s prod- uct (Nemeroff, 1998). Tends to Distort the Knowledge Base and the Terminology in the Field The findings from a large number of comparative clinical trials that are financed by pharmaceutical companies are commonly included in literature re- views and meta-analyses. Many are scientifically weak. Their inclusion in a review may well alter the conclusions of a reviewer (and therefore the read- er). For example, Stahl (1999) and Ho et al. (1999), citing a drug company–sponsored study, accept the reported but misleading finding that “low”-dose ris- peridone has the same likelihood as placebo to in- duce EPS. Likewise, certain terms that reflect a drug compa- ny’s claim may enter into common medical usage. For example, the term ‘mood stabilizer’ is commonly applied in pharmaceutical company–sponsored studies to certain anticonvulsant and neuroleptic drugs as though they benefited mood beyond their acute and short-term anti-manic effects (Sobo, 1999). The sheer weight of the company’s many suppos- edly favorable studies has its own impact. The pos- itive comparative findings of company-financed clin- ical trial research are widely disseminated and aggressively promoted. The fact that these findings are published in nationally circulating medical jour- nals suggests to physicians and to the public that they have high-level approval from within the med- ical establishment. Limits Research Autonomy Confidentiality agreements between the researcher and the sponsor delay research reporting and occa- sionally impede publication of findings deemed unde- sirable by the sponsoring company (Blumenthal, 1996; Chalmers, 1990; Dickersin et al., 1987; Phillips and Hoey, 1998; Rennie, 1997, Shuchman, 1998). Influences Physicians to Decrease Their Prescriptions of Less Costly Medications That Have Equivalent Efficacy Once second-generation neuroleptics became viewed as first-line medication treatments, physi- cians increasingly bypassed their neuroleptic prede- cessors— ones that, though still very useful, are no longer promoted. First-generation neuroleptics as a group cause far less weight gain than their succes- sors (Allison et al., 1999) and their side effect pro- files are much better established. A recent meta- analysis of available data comparing conventional with second-generation neuroleptics by the National Schizophrenia Guideline Development Group of Great Britain concluded that they had equivalent efficacy and tolerability. Therefore, the group rec- SAFER588 ommended that “conventional antipsychotics should usually be used in the initial treatment of an episode of schizophrenia unless the patient has previously not responded to these drugs or has unacceptable extrapyramidal side effects” (Geddes et al., 2000). In support of this, a four-week, independently funded study recently revealed that modest average doses of haloperidol (3.7 mg/day; N � 212) were equally as effective as average doses of risperidone (3.2 mg/ day; N � 34) for neuroleptic-naive psychotic pa- tients, and that at these doses, the two neuroleptics had similar results on the development of EPS (Rosebush and Mazurek, 1999). In like manner, independent reviewers have con- cluded that tricyclic antidepressants—whose pro- portional use has prominently decreased in the 1990s (Hirshfeld, 1998)—are as efficacious for the treatment of depression as SSRIs (Anderson, 2000), although they are far more hazardous in overdose. Furthermore, for certain depressive subtypes (e.g., melancholia), the first-generation antidepressants appear to be more effective than SSRIs (Parker et al., 2001). Nonetheless, both second-generation an- tidepressants and second-generation neuroleptics are routinely reported to be “first-line agents” by opinion leaders who wield great influence when they present industry-sponsored studies at profes- sional meetings. Conclusion The concerns raised herein relate largely to the very expensive, often misleading, and commonly wasteful drug trial comparisons of similarly effec- tive drugs that are sponsored by major pharmaceu- tical companies. In response to these concerns, nu- merous articles and editorials written over the last two decades in medical journals have focused on: “conflicts of interest,” “scientific misconduct” (Chalmers, 1990), “is medicine for sale?” (Angell, 2000a), “commercial sources of influence” (Avorn et al., 1982), and “the industrialization of clinical re- search” (Rettig, 2000). Furthermore, ethical practice guidelines have been written by the Food and Drug Administration and professional medical organiza- tions to limit conflicts of interest between physi- cians and pharmaceutical company representatives (Goldfinger, 1990; Zappala, 1998). Although legal restraints have curtailed some exces- sive industry financial incentives to physicians, the financial power of the big pharmaceutical companies continues to make them far and away the dominant vendors of clinical research (Bodenheimer, 2000; Silversides, 1998). So, however forceful, the overall influence of the editorials and guidelines of organized medicine appears to be meager. The major pharmaceutical companies are ex- tremely successful financially (Angell, 2000b), such that they can afford to spend an average of 23 to 30 cents of every dollar of the wholesale cost of trade name prescription medications on promotional mar- keting (Barry, 2000; Stolberg and Gerth, 2000; Wolfe, 1996). Likewise, fully one third of the American Psychiatric Association’s annual income is derived from pharmaceutical companies (Borenstein, 2000). Such financial resources allow the industry to suc- ceed handily in influencing physician practice and determining the bulk of the U.S. drug research agenda (Avorn et al., 1982; Lexchin, 1993; Peterson et al., 2000). With these financial advantages, the adage continues to apply: those who pay the piper call the tune. References Allison DB, Casey DE (2001) Antipsychotic-induced weight gain: a review of the literature. J Clin Psychiatry 62 (Suppl 7):22– 31. Allison DB, Mentore JL, Heo M, Chandler LP, Cappelleri JC, Infante MC, Weiden PJ (1999) Antipsychotic-induced weight gain. Am J Psychiatry 156:1686 –1696. Anderson IM (2000) Selective serotonin reuptake inhibitors ver- sus tricyclic antidepressants. J Affect Disord 58:19 –36. Angell M (2000a) Is academic medicine for sale? N Engl J Med 342:1516 –1518. Angell M (2000b) The pharmaceutical industry—to whom is it accountable? N Engl J Med 342:1902–1904. Angell M (2001) Medicine in the noise age: what can we believe? Accountability Res 8:189 –195. Armitage R, Yonkers K, Cole D, Rush J (1997) A multicenter, double-blind comparison of the effects of nefazodone and fluoxetine on sleep architecture and quality of sleep in de- pressed outpatients. J Clin Psychopharmacol 17:161–168. Avorn J, Chen M, Hartley R (1982) Scientific versus commercial sources of influence on the prescribing behavior of physicians. Am J Med 73:4 – 8. Baldessarini RJ, Cohen BM, Teicher MH (1988) Significance of neuroleptic dose and plasma level in the pharmacological treatment of psychoses. Arch Gen Psychiatry 45:79 –91. Barry P (2000, April) What’s behind high drug prices in the U.S.? AARP Bulletin 4:4 –10. Beasley CM, Dellva MA, Tamura RN, Morgenstern H, Glazer WM, Ferguson K, Tollefson GD (1999) Randomized double-blind comparison of the incidence of tardive dyskinesia in patients with schizophrenia during long-term treatment with olanzap- ine or haloperidol. Br J Psychiatry 174:23–30. Bero LA, Galbraith A, Rennie D (1992) The publication of spon- sored symposiums in medical journals. N Engl J Med 327: 1135–1140. Bitter I, Dossenbach M, Martenyi F, Slabber M (2000) Olanzapine versus clozapine in patients non-responsive to standard ac- ceptable treatment of schizophrenia. Poster presented at the American Psychiatric Association Annual Meeting, Chicago, Ill. Blumenthal D (1996) Ethics issues in academic-industry relation- ships in the life sciences. Acad Med 71:1291–1296. Bobbio M, Demichelis B, Giustetto G (1994) Completeness of reporting trial results. Lancet 343:1209 –1211. Bodenheimer T (2000) Uneasy alliance: clinical investigators and the pharmaceutical industry. N Engl J Med 342:1539 –1544. DESIGN AND REPORTING MODIFICATIONS 589 Bondolfi G, Dufour H, Patris M, May JP, Billeter U (1998) Ris- peridone versus clozapine in treatment resistant chronic schizophrenia. Am J Psychiatry 155:499 –504. Borenstein D (2000, November 17) Pharmaceutical companies. Psychiatric News 35:3. Borison RL, Pathiraja AP, Diamond BI, Meibach RC (1992) Ris- peridone: clinical safety and efficacy in schizophrenia. Psycho- pharmacol Bull 28:213–218. Breier A, Hamilton SH (1999) Comparative efficacy of olanzapine and haloperidol for patients with treatment-resistant schizo- phrenia. Biol Psychiatry 45:403– 411. Brenner HD, Dencker SJ, Goldstein MJ, Hubbard JW, Keegan DL (1990) Defining treatment refractoriness in schizophrenia. Schizophrenia Bull 16:551–561. Brett AS (1989) Treating hypercholesterolemia: how should prac- ticing physicians interpret the published data for patients. N Engl J Med 321:676 – 680. Bruggeman R, Linden CV, Buitelaar JK, Gericke GS, Hawkridge SM, Temlett JA (2001) Risperidone versus pimozide in Tourette’s disorder: a comparative double-blind, parallel- group study. J Clin Psychiatry 62:50 –56. Carpenter WT (2001) Industry phase IV trials are of little or no value: pro. J Psychotic Disorders 5(4):3&13. Carpenter WT (2002) From clinical trial to prescription. Arch Gen Psychiatry 59:282–285. Caudill TS, Johnson MS, Rich EC, McKinney P (1996) Physicians, pharmaceutical sales representatives and the cost of prescrib- ing. Arch Fam Med 5:201–206. Chalmers I (1990) Underreporting research is scientific miscon- duct. JAMA 263:1405–1408. Cho MK, Bero LA (1996) The quality of drug studies in sympo- sium proceedings. Ann Intern Med 124:485– 489. Chouinard G, Jones B, Remington G, Bloom D, Addington D, MacEwan GW, Labelle A, Beauclair L, Arnott W (1993) A Canadian multi-center placebo-controlled study of fixed doses of risperidone and halperidol in the treatment of chronic schizophrenic patients. J Clin Psychopharmacol 13:25– 40. Chren MM, Landefeld CS (1994) Physicians’ behavior and their interactions with drug companies. JAMA 271:684 – 689. Christiansen PE, Behnke K, Black CH, Ohrstrom JK, Bork- Rasmussen H, Nilsson J (1996) Paroxetine and amitriptyline in the treatment of depression in general practice. Acta Psychi- atr Scand 93:158 –163. Claus A, Bollen J, De Cuyper H, Eneman M, Malfroid M, Peuskens J, Heylen S (1992) Risperidone versus haloperidol in the treatment of chronic schizophrenic inpatients. Acta Psy- chiatr Scand 85:295–305. Coleman CC, Cunningham LA, Foster VJ, Batey SR, Donahue RM, Houser TL, Ascher JA (1999) Sexual dysfunction associ- ated with the treatment of depression: a placebo-controlled comparison of bupropion sustained release and sertraline treatment. Ann Clin Psychiat 60:205–215. Conley RP (2001) Industry phase IV trials are of little or no value: con. J Psychotic Disorders 5(4):3, 13. Conley RR, Tamminga CA, Kelly DL, Richardson CM (1999a) Treatment-resistant schizophrenic patients respond to cloza- pine after olanzapine non-response. Biol Psychiatry 46:73–77. Conley RR, Love RC, Kelly DL, Bartko JJ (1999b) Rehospitaliza- tion rates of patients recently discharged on a regimen of risperidone or clozapine. Am J Psychiatry 156:863– 868. Council on Scientific Affairs (1990) Conflicts of interest in med- ical center/industry research relationships. JAMA 263:2790 – 2793. Croft H, Settle E, Houser T, Batey SR, Donahue MJ, Ascher JA (1999) A placebo-controlled comparison of the antidepressant efficacy and effects on sexual functioning of sustained-release bupropion and sertraline. Clin Therap 21:643– 658. Davidoff F (2002) Between the lines: navigating the uncharted territory of industry-sponsored research. Health Affairs 21: 235–242. Davidoff F, De Angelis CD, Drazen JM, Hoey J, Hojgaard L, Horton R (2001) Sponsorship, authorship, and accountability JAMA 286:1232–1234. Davidson RA (1986) Source of funding and outcome of clinical trials. J Gen Intern Med 1:155–158. Deo R, Soni S, Rastogi SC, Levine S, Plant I (1990) Remoxipride and haloperidol in the acute phase of schizophrenia: a double- blind comparison. Acta Psychiatr Scand 82(Suppl. 358):120 – 124. Dickersin K, Chan S, Chalmers TC, Sacks HS, Smith H (1987) Publication bias and clinical trials. Controlled Clin Trials 8:343–353. Eichenwald K, Kolata G (1999a, May 16) Drug trials hide con- flicts for doctors. New York Times, pp.1, 28. Eichenwald K, Kolata G (1999b, May 17) A doctor’s drug studies turn into fraud. New York Times, pp. A1, A16. Emsley RA (1999) Partial response to antipsychotic treatment: the patient with enduring symptoms. J Clin Psychiatry 60(Suppl 23):10 –13. Feiger A, Kiev A, Shrivastava RK, Wisselink PG, Wilcox CS (1996) Nefazodone versus sertraline in outpatients with major depression. J Clin Psychiatry 57(Suppl 2):53– 62. Feinstein AR (1992) Invidious comparisons—and unmet clinical challenges. Am J Med 92:117–120. Flanagin A, Carey LA, Fontanarosa PB, Phillips SG, Pace BP, Lundberg MD, Rennie D (1998) Prevalence of articles with honorary authors and ghost writers in peer-reviewed medical journals. JAMA 280:222–224. Friedberg M, Saffran B, Stinson TJ, Nelson W, Bennett CL (1999) Evaluation of conflict of interest in economic analyses of new drugs used in oncology. JAMA 282:1453–1457. Friis S (1999) To what extent should papers submitted from drug companies be published in medical journals? Acta Psychiatr Scand 99:157–159. Garattini S (1997) Are me-too drugs justified? J Nephrol 10:283– 294. Geddes J, Freemantle N, Harrison P, Bebbington P (2000) Atyp- ical antipsychotics in the treatment of schizophrenia: system- atic overview and meta-regression analysis. BMJ 321:1371– 1376. Geerts S, Lacante P, Mignon A (1999) Venlafaxine versus fluox- etine in the treatment of depressed patients with concomitant anxiety. Paper presented at the 12th European College of Neuropsychopharmacology Congress. London. Gilbody SM, Song F (2000) Publication bias and integrity of psychiatric research. Psychol Med 30:253–258. Gilbody S, Wahlbeck K, Adams C (2002) Randomized controlled trials in schizophrenia: a critical perspective on the literature. Acta Psychiatr Scand 105:243–251. Goldberg C (1999, April 6) Urging a freer flow of scientific ideas. New York Times. Goldfinger SE (1990) Physicians and the pharmaceutical indus- try. Ann Intern Med 112:624 – 626. Goldstein J, Emsley R, Raniwalla J, Bailey P, Jones M (1999) Efficacy and tolerability of quetiapine compared to haloperi- dol in schizophrenic patients partially responsive to conven- tional antipsychotic treatment. Poster presented at the 39th Annual NCDEU Meeting, Boca Raton, Fl. Gorelick KJ (1998) Industry affiliations and scientific conclu- sions. JAMA 280:1141–1142. Greene WL, Concato J, Feinstein AR (2000) Claims of equiva- lence in medical research: are they supported by the evidence? Ann Intern Med 132:715–722. Greenhill LL, Swanson JM, Vitiello B, Davies M, Clevenger N, Wu M, Arnold LE (2001) Impairment and deportment responses to different methylphenidate doses in children with ADHD. J Am Acad Child Adolesc Psychiatry 40:180 –187. Gwynne P (1999) Corporate collaborations: scientists can face publishing constraints. Clientist 13(11):1, 6. Heilgenstein JH, Spenser TJ, Faries DE, Biederman, Kratochvil IC, Conners CK (2000) Efficacy of tomoxetine vs placebo in pediatric outpatients with ADHD. Presented at the American Academy of Child and Adolescent Psychiatry Annual Meeting, New York, NY. Hillman AL, Eisenberg JM, Pauly MV, Bloom BS, Glick H, Knosian B, Schwartz JS (1991) Avoiding bias in the conduct of SAFER590 cost-effectiveness research sponsored by pharmaceutical companies. N Engl J Med 324:1362–1365. Hirshfeld RM(1998) American health care systems and depres- sion. J Clin Psychiatry 59(Suppl 20):5–10. Ho BC, Miller D, Nopoulos P, Andreasen NC (1999) A compara- tive effectiveness study of risperidone and olanzapine in the treatment of schizophrenia. J Clin Psychiatry 60:658 – 663. Hotopf M, Lewis G, Normand C (1997) Putting trials on trial—the costs and consequences of small trials in depression. J Epi- demiol Community Health 51:354 –358. Huston P, Moher D (1996) Redundancy, disaggregation, and the integrity of medical research. Lancet 347:1024 –1026. Janssen Pharmaceutica Research Foundation (1996) Risperidol: a first choice in psychosis. J Clin Psychiatry 57:62. Kavoussi RJ, Segraves T, Hughes AR, Ascher JA, Johnston JA (1997) Double-blind comparison of bupropion sustained re- lease and sertraline in depressed outpatients. J Clin Psychia- try 58:332–337. Kessler DA (1992) Addressing the problem of misleading adver- tising. Ann Intern Med 116:950 –951. Kessler DA, Rose JL, Temple RJ, Sharpiro R, Griffin JP (1994) Therapeutic-class wars-drug promotion in the marketplace. N Engl J Med 331:1350 –1353. Kiev A, Feiger A (1997) A double-blind comparison of fluvoxam- ine and paroxetine in the treatment of depressed outpatients. J Clin Psychiatry 58:146 –152. Krimsky S, Rothenberg LS, Stott G, Kyle G (1998) Scientific journals and their authors’ financial interests: a pilot study. Psychother Psychosom 67:194 –204. Lane HY, Chiu WC, Chou JC, Wu ST, Su MH, Change WH (2000) Risperidone in acutely exacerbated schizophrenia. J Clin Psy- chiatry 61:209 –214. Lapierre YD, Nair NP, Chouinard G, Awad AG, Saxena B (1990) A controlled dose-ranging study of remoxipride and haloperi- dol in schizophrenia—a Canadian multicentre trial. Acta Psy- chiatr Scand 82(Suppl 358):72–76. Lauritsen K, Havelund T, Laursen LS, Madsen JR (1987) With- holding unfavorable results in drug company sponsored clini- cal trials. Lancet 1:1091. Lecrubier Y, Judge R (1997) Long-term evaluation of paroxetine, clomipramine and placebo in panic disorder. Acta Psychiatr Scand 95:153–160. Lemmens P, Brecher M, Van Baelen B (1999) A combined anal- ysis of double-blind studies with risperidone vs. placebo and other antipsychotic agents. Acta Psychiatr Scand 99:160 –170. Lexchin J (1993) Interactions between physicians and the phar- maceutical industry. CMAJ 149:1401–1407. Lieberman JA, Safferman A, Pollack S, Szymanski S, Johns C, Howard A (1994) Clinical effects of clozapine in chronic schizophrenia. Am J Psychiatry 151:1744 –1752. Littrell KH, Johnson CG, Peabody CD, Hilligoss NM, Petty RG (2000) Olanzapine treatment for patients with schizophrenia and substance abuse. Poster at the 40th annual meeting of NCDEU, Boca Raton, Fl. Longstaffe S, Moffatt ME, Whalen JC (2000) Behavioral and self-concept changes after six months of enuresis treatment. Pediatrics 105:935–940. Love RC, Conley RR, Kelly DL, Bartko JJ (1999) A dose-outcome analysis of risperidone. J Clin Psychiatry 60:771–775. Mandelkern M (1999) Manufacturer support and outcome. J Clin Psychiatry 60:122. Marder SR, Davis JM, Chouinard G (1997) The effects of risperi- done on the five dimensions of schizophrenia derived by factor analysis. J Clin Psychiatry 58:538 –546. Marder SR (1999) Newer antipsychotics in treatment-resistant schizophrenia. Biol Psychiatry 45:383–384. Marder SR, Meibach RC (1994) Risperidone in the treatment of schizophrenia. Am J Psychiatry 151:825– 835. Marshall M, Lockwood A, Bradley C, Adams C, Joy C, Fenton M (2000) Unpublished rating scales: a major source of bias in randomized controlled trials of treatments for schizophrenia. Br J Psychiatry 176:249 –252. Mc Evoy JP, Hogarty GE, Steingard S (1991) Optimal dose of neuroleptic in acute schizoprenia. Arch Gen Psychiatry 48: 739 –745. Modell JG, Katholi CR, Modell JD, DePalma RL (1997) Compar- ative sexual side effects of bupropion, fluoxetine, paroxetine, and sertraline. Clin Pharmacol Ther 61:476 – 487. Monda JM, Husmann DA (1995) Primary nocturnal enuresis: a comparison among observation, imipramine, desmopressin ac- etate, and bed-wetting alarm systems. J Urol 154:745–748. Montejo-Gonzalez AL, Liorca G, Izquierdo JA, Ledesma A (1997) SSRI-induced sexual dysfunction. J Sex Marital Ther 23:176 – 183. Moukheiber Z (2001) Conflicting claims, manipulated data, snip- ing medical ads. Forbes 167:267–268. Moynihan R (2002) Drug firms hype disease as sales ploy, indus- try chief claims. BMJ 324:867. Mullen J, Reinstein M, Bari M, Ginsberg L, Sandler N (1999) Quetiapine and risperidone in outpatients with psychotic dis- orders. Poster presented at the 39th Annual Meeting of the NCDEU, Boca Raton, Fl. Nelson JC, Kennedy JS, Pollock BG, Thode FL, Narayan M (1999) Treatment of major depression with nortriptyline and parox- etine in patients with ischemic heart disease. Am J Psychiatry 156:1024 –1028. Nemeroff CB (1998) The escalating pharmaceutical company wars: Where is the academic to hide? CNS Spectrums 3:17. Nyberg S, Eriksson B, Oxenstierna G, Halldin C, Farde L (1999) Suggested minimal effective dose of risperidone based on PET-measured D2 and 5-HT2A receptor occupancy in schizo- phrenic patients. Am J Psychiatry 156:869 – 875. Parker G, Roy K, Wilhelm K, Mitchell P (2001) Assessing the comparative effectiveness of antidepressant therapies: a pro- spective clinical practice study. J Clin Psychiatry 62:117–125. Patris M, Agussol P, Alby JM, Brion S, Burnat G, Castelnau D (1990) A double-blind multicentre comparison of remoxipride, at two dose levels, and haloperidol. Acta Psychiatr Scand 82(Suppl 358):78 – 82. Peterson TJ, Dording CM, Kornbluh RA, Alpert JE, Nierenberg AA, Rosenbaum JF, Fava M (2000) A survey of prescribing in the treatment of depression. Presented at the 153rd Annual Meeting of the American Psychiatric Association. Chicago, IL. Peuskens J, Bech P, Moller HJ, Bale R, Fleurot O, Rein W (1999) Amisulpride vs. risperidone in the treatment of acute exacer- bations of schizophrenia. Psychiat Res 88:107–117. Phillips RA, Hoey J (1998) Constraints of interest: lessons at the Hospital for Sick Children. CMAJ 159:955–957. Preskorn SH (1997) Clinically relevant pharmacology of selective serotonin reuptake inhibitors. Clin Pharmacokinet 32(Suppl 1):1–21. Quick J (2001) Maintaining the integrity of the clinical evidence base. Bull World Health Org 79:1093. Rapaport M, Coccaro E, Sheline Y, Perse T, Holland P, Fabre L, Bradford D (1996) A comparison of fluvoxamine and fluox- etine in the treatment of major depression. J Clin Psychophar- macol 16:373–378. Rennie D (1997) Thyroid storm. JAMA 277:1238 –1243. Rennie D (1999) Fair conduct and fair reporting of clinical trials. JAMA 282:1766 –1768. Rennie D, Flanagin A (1994) Authorship! Authorship! Guests, ghosts, grafters, and the two sided coin. JAMA 271:469 – 471. Rettig RA (2000) The industrialization of clinical research. Health Aff 19:129 –146. Rifkin A, Doddi S, Karajgi B, Borenstein M, Wachspress M (1991) Dosage of haloperidol for schizophrenia. Arch Gen Psychiatry 48:166 –170. Rochon PA, Gurwitz JH, Cheung M, Hayes JA, Chalmers TC (1994a) Evaluating the quality of articles published in journal supplements compared with the quality of those published in the parent journal. JAMA 272:108 –113. Rochon PA, Gurwitz JH, Simms RW, Fortin PR, Felson DT, Minaker KL, Chalmers TC (1994b) A study of manufacturer- supported trials of nonsteroidal anti-inflammatory drugs in the treatment of arthritis. Arch Intern Med 154:157–163. DESIGN AND REPORTING MODIFICATIONS 591 Roose SP (1992) Modern cardiovascular standards for psycho- tropic drugs. Psychopharmacol Bull 28:35– 43. Rosebush PI, Mazurek MF (1999) Neurologic side effects in neuroleptic-naïve patients treated with haloperidol or risperi- done. Neurology 52:782–785. Rudolph RL, Feiger AD (1999) A double-blind, randomized, pla- cebo-controlled trial of once-daily venlafaxine extended re- lease (XR) and fluoxetine for the treatment of depression. J Affect Disord 56:171–181. Schooler NR (2000) Evaluating clinical trial data from schizo- phrenia research. J Clin Psychiatry Audiographic Series, 3(3):13, 18. Schulman SL, Colish Y, von Zuben FC, Kodman-Jones C (2000) Effectiveness of treatment for nocturnal enuresis in a hetero- geneous population. Clin Pediatrics 39:359 –364. Segraves RJ, Kavoussi R, Hughes AR, Batey SR, Johnston JA, Donahue R, Ascher JA (2000) Evaluation of sexual functioning in depressed outpatients. J Clin Psychopharmacol 20:122–128. Sheehan D (1999) Anxiety and mood disorders. Presented at the 152nd Annual Meeting of the American Psychiatric Associa- tion, Washington, D.C. Shenk D (1999, March 22) Money & science�ethics problems on campus Nation 268:11–18. Shimm DS, Spece RG (1991) Industry reimbursement for enter- ing patients into clinical trials. Ann Intern Med 115:148 –151. Shuchman M (1998) Legal issues surrounding privately funded research cause furore in Toronto. CMAJ 159:983–986. Silva JC (1998) Randomized, double blind comparison of ven- lafaxine and fluoxetine in outpatients with major depression. J Clin Psychiatry 59:352–357. Silversides A (1998) Private sector becoming the key to research funding in Canada. CMAJ 159:397–398. Simpson GM, Lindenmeyer JP (1997) Extrapyramidal symptoms in patients treated with risperidone. J Clin Psychopharmacol 17:194 –201. Sobo S (1999, October) Mood stabilizers and mood swings: In search of a definition. Psychiatric Times pp. 36 – 42. Soumerai SB, Ross-Degnan D (1999) Inadequate prescription- drug coverage for medicare enrollees—a call to action. N Engl J Med 340:722–727. Spingarn RW, Berlin JA, Strom BL (1996) When pharmaceutical manufacturers’ employees present grand rounds, what do res- idents remember? Acad Med 71:86 – 88. Stahl SM (1999) Selecting an atypical antipsychotic by combining clinical experience with guidelines from clinical trials. J Clin Psychiatry 60(Suppl 10):31– 41. Stolberg SG, Gerth J (2000, July 23) How companies stall gener- ics and keep themselves healthy. New York Times pp.1,12 &13. Stone CK, Garver DL, Griffith J, Hirschowitz, Bennett J (1995) Further evidence of a dose-response threshold for haloperidol in psychosis. Am J Psychiatry 152:1210 –1212. Tamminga CA (2000) Evaluating clinical trial data from schizo- phrenia research. J Clin Psychiatry Audiographic Series, 3(3):4 –5. Tarabusi CC, Vickery G (1998) Globalization in the pharmaceu- tical industry, part 1. Int J Health Serv 28:67–105. Tollefson GD, Beasley CM, Tamura RN, Tran PV, Potvin JH (1997a) Blind, controlled, long-term study of the comparative incidence of treatment-emergent tardive dyskinesia with olan- zapine or haloperidol. Am J Psychiatry 154:1248 –1254. Tollefson GD, Beasley CM, Tran PV, Street JS, Krueger JA, Tamura RN, Graffeo KA, Thieme ME (1997b) Olanzapine ver- sus haloperidol in the treatment of schizophrenia and schizo- affective and schizophreniform disorders. Am J Psychiatry 154:457– 465. Tran P, Creanga D, Zhang F, Wang J, Vangala S, Cousins L, Tollefson G (1999) Clinical experience with olanzapine in ethnic subgroups. Poster at the 39th Annual Meeting of the NCDEU, Boca Raton, Fl. Tran PV, Dellva MA, Tollefson GD, Wentley AL, Beasley CM (1998) Oral olanzapine versus oral haloperidol in the mainte- nance treatment of schizophrenia and related psychoses. Br J Psychiatry 172:499 –505. Tran PV, Dellva MA, Tollefson GD, Beasley CM, Potvin JH, Kiesler GM (1997a) Extrapyramidal symptoms and tolerability of olanzapine versus haloperidol in the acute treatment of schizophrenia. J Clin Psychiatry 58:205–211. Tran PV, Hamilton SH, Kuntz AJ, Potvin JH, Andersen SW, Beasley C, Tollefson GD (1997b) Double-blind comparison of olanzapine versus risperidone in the treatment of schizophre- nia and other psychotic disorders. J Clin Psychopharmacol 17:407– 418. Trivedi MH, Rush AJ, Carmody TJ, Donahue RM, Bolden-Watson C, Houser TL, Metz A (2001) Do bupropion SR and sertraline differ in their effects on anxiety in depressed patients? J Clin Psychiatry 62:776 –781. Van Putten T, Marder SR, Mintz J (1990) A controlled dose comparison of haloperidol in newly admitted schizophrenic patients. Arch Gen Psychiatry 47:754 –758. Walker PW, Cole JO, Gardner EA, Hughes AR, Johnston A, Batey SR, Lineberry CG (1993) Improvement in fluoxetine-associated sexual dysfunction in patients switched to bupropion. J Clin Psychiat 54:459 – 465. Weiden PJ (2001) Optimizing optimistically: getting the most out of the atypical antipsychotics. Presented at the Annual Meet- ing of the American Psychiatric Association, New Orleans, La. Weiden P, Mackell J, McDonnell DD (2000) Obesity as a risk factor for antipsychotic noncompliance. Poster at the 40th Annual Meeting of the NCDEU, Boca Raton, Fl. Wheatley DP, Moffaert M, Timmerman L, Kremer CM (1998) Mirtazapine: Efficacy and tolerability in comparison with flu- oxetine in patients with moderate to severe major depressive disorder. J Clin Psychiat 59:306 –312. Wille S (1986) Comparison of desmopressin and enuresis alarm for nocturnal enuresis. Arch Dis Childhood 61:30 –33. Wirshing DA, Marshall BD, Green MF, Mintz J, Marder SR, Wirshing WC (1999) Risperidone in treatment-refractory schizophrenia. Am J Psychiatry 156:1374 –1379. Wolfe SM (1996) Why do American drug companies spend more than $12 billion a year pushing drugs? J Gen Int Med 11:637– 639. Wong JZ, Zipursky RB, Beiser M, Bean G (1999) Optimal halo- peridol dosage in first-episode psychosis. Can J Psychiatry 44:164 –167. Woosley RL (1994) A prescription for better prescritions. Iss Sci Technol 10:59 – 66. Zajecka J, Amsterdam JD, Quitkin FM, Reimherr FW, Rosenbaum JF (1999) Changes in adverse events reported by patients during 6 months of fluoxetine therapy. J Clin Psychiatry 60:389 –394. Zappala FN (1998) FDA’s Rx for keeping educational programs free of industry bias. Am J Health-Syst Pharm 55:2472–2474. Ziegler MG, Lew P, Singer BC (1995) The accuracy of drug information from pharmaceutical sales representatives. JAMA 273:1296 –1298. SAFER592