eLetters

Health Affairs encourages readers to engage in debate via electronic letters to the editor.

• To RESPOND to a particular article: Click on the link "Submit a response to this article" in the box at the top right-hand corner of the article.
• To READ responses to a particular article: Click on the link "View responses" in the box at the top right-hand corner of the article.

Electronic Letters to:

Stephen B. Soumerai, Fang Zhang, Dennis Ross-Degnan, Daniel E. Ball, Robert F. LeCates, Michael R. Law, Tom E. Hughes, Daniel Chapman, and Alyce S. Adams Use Of Atypical Antipsychotic Drugs For Schizophrenia In Maine Medicaid Following A Policy Change Health Affairs, May/June 2008; 27(3): w185-w195. [Abstract] [Full Text] [Figures Only] [PDF] [Online Appendix] [Reprints & Permissions]

eLetters:Submit a response to this article

Electronic letters published:

Atypical Antipsychotic Study Misses Mark

Timothy S. Clifford, M.D.   ( 9 April 2008 )

Evidence Of Adverse Effect Of PA Policy In Maine Is Weak

Robert A. Rosenheck   ( 9 April 2008 )

Letter Writers On Antipsychotic PA Policy Study Corroborate Our Evidence

Stephen B. Soumerai   ( 21 April 2008 )

Atypical Antipsychotic Study Misses Mark 9 April 2008
Timothy S. Clifford, M.D., Medical Consultant GHS Data Management Send letter to journal: Re: Atypical Antipsychotic Study Misses Mark tclifford{at}ghsinc.com Timothy S. Clifford, M.D.	Soumerai and colleagues' atypical antipsychotic study provides a cautionary example of how the best study design can be undone by inadequate preparation. In this case, there was a failure to interview any of the key Maine Medicaid pharmacy benefit administrators, including myself, the Medical Consultant for the state’s Preferred Drug List (PDL). Due diligence might have prevented several major unaddressed study limitations that render their conclusions either factually incorrect or unsupported. The most important study limitation is that it was inadequately controlled to support their conclusions and therefore damages any attributions concerning cause and effect or policy implications. The atypical antipsychotic PDL was not implemented in isolation. Virtually every drug category was affected simultaneously, so there were a large number of potentially confounding factors present in any given member’s case. Atypical drug discontinuance could have been influenced by a host of other PDL drug changes exclusive of the atypical category itself. If this were true, one could then offer the very different policy conclusion that rather than exempting the schizophrenia population from an atypical PDL, the state should instead lengthen the implementation schedule and delay the atypical drug class onset. Interestingly enough, the authors now have an ideal opportunity to redo their study and largely correct this limitation. Maine reinstituted prior authorization (PA) for new starters of Abilify and Zyprexa in July of 2007 at a time when practically no other changes were made to the PDL. The second limitation concerns discontinuity. This is problematic, since an analysis of Medicaid pharmacy claims only allows you to reach conclusions about what drugs Medicaid paid for, and not necessarily what additional drugs the patient actually received. We know from our PDL experience that many physicians maintained their Medicaid patients on samples of nonpreferred atypical drugs. We were able to confirm this by collecting chart records. The atypical PDL increased the use of atypical drug samples in an attempt to either circumvent or at least delay the use of preferred drugs. This obscured use of nonpreferred atypical drugs is crucial, since the member might in fact have never discontinued therapy. One last note on Medicaid pharmacy cost: If you do not have Medicaid rebate data, then you have no clue as to a state’s net costs. All Medicaid programs receive federally mandated drug rebates, and most states also negotiate for supplemental rebates. State Medicaid programs receive anywhere from 1% rebates on generics to 100% rebates on some brand-name drugs. This study only makes inaccurate observations about actual drug costs.
Evidence Of Adverse Effect Of PA Policy In Maine Is Weak 9 April 2008
Robert A. Rosenheck, Professor of Psychiatry Yale Medical School Send letter to journal: Re: Evidence Of Adverse Effect Of PA Policy In Maine Is Weak Robert.Rosenheck{at}yale.edu Robert A. Rosenheck	Soumerai et al. conclude that implementation of prior authorization (PA) for atypical antipsychotics (AAs) in Maine increased the risk of medication discontinuation with potentially harmful consequences. However, the increased risk of discontinuation in Maine as compared to New Hampshire was not statistically significant on the primary endpoint, and was significant on only one of two secondary endpoints (Exhibit 4). Furthermore, the magnitude of the PA effect in Maine was small, reaching about 10% at 90 days (Exhibit 2). More alarming is that, unrelated to PA, about 80% of new medication users in both states discontinued their medication after only five months. Small differences in attributable risk can yield misleadingly large hazard ratios when surviving proportions are small. The suggestion that a 10% increase in drug discontinuation caused meaningful adverse clinical consequences is not supported by data from the Maine sample, e.g., showing significantly increased hospitalization. The papers cited by Law et al. and Weiden et al. are associational in nature and thus cannot distinguish between discontinuation causing relapse and relapse causing discontinuation. The 18-month Clinical Antipsychotic Trial of Intervention Effectiveness (CATIE) trial [1], which the authors indicate was the model for their outcome measure, found up to 18% differences in attributable discontinuation risk between randomly assigned drugs (twice that attributed to PA in Maine) but found no associated differences in quality of life and most other outcomes.[2] Maine had reason to try to control AA expenditures. While AAs cost 10-20 times as much as generic antipsychotics, recent research has shown them to have few, if any, advantages and increased risk of metabolic syndrome.[3] Coauthorship of this study by employees of an AA manufacturer (and study cosponsor) may also raise conflict-of-interest concerns for some readers. Nevertheless, while PA may be problematic for clinicians, the evidence of harm from the Maine PA policy seems weak. References 1. Lieberman JA, Stroup S, McEvoy J et. al., Effectiveness of Antipsychotic Drugs in Patients with Chronic Schizophrenia: Primary Efficacy and Safety Outcomes of the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) Schizophrenia Trial. New England Journal of Medicine 2005; 353 (12): 1209-1223. 2. Swartz M, Stroup S, McEvoy J, Davis S, Rosenheck RA, Keefe RSE, Hsaio J, Lieberman J (in press). What CATIE Found: Results from the Schizophrenia Trial. Psychiatric Services. 3. Lewis S, Lieberman J. CATIE and CUtLASS: can we handle the truth? British Journal of Psychiatry 2008;192: 161-163. Dr. Rosenheck reports having received research funding from AstraZeneca Pharmaceuticals LP, Bristol-Myers Squibb, and Eli Lilly and Co.; and consulting fees from Bristol-Myers Squibb, Eli Lilly and Co., Organon Inc. and Janssen Pharmaceutica Products. He provided expert testimony for the plaintiffs in UFCW Local 1776 and Participating Employers Health And Welfare Fund, et al. v. Eli Lilly and Company.
Letter Writers On Antipsychotic PA Policy Study Corroborate Our Evidence 21 April 2008
Stephen B. Soumerai, Professor of Ambulatory Care and Prevention Harvard Medical School and Harvard Pilgrim Health Care Send letter to journal: Re: Letter Writers On Antipsychotic PA Policy Study Corroborate Our Evidence ssoumerai{at}hms.harvard.edu Stephen B. Soumerai	We thank Drs. Clifford and Rosenheck for their thoughtful comments, which ultimately reinforce our concerns that prior authorization (PA) of atypical antipsychotic agents can disrupt treatment of vulnerable patients with schizophrenia. We concur with Dr. Clifford that interactions between policy researchers and government officials are desirable, and we will try to increase such linkages in the future. We disagree, however, that our conclusions are “factually incorrect or unsupported.” Dr. Clifford makes the excellent point that Maine Medicaid implemented a Preferred Drug List affecting many drugs simultaneously. We pointed out in the discussion that schizophrenia patients, with many medical comorbidities, might be especially susceptible to challenges accessing other drugs, potentially exacerbating the effects of PA of their primary antipsychotic medications. So the “confounding” may be better understood as interacting with the effects of the atypical antipsychotic PA. Interestingly, we have presented initial results at a national scientific meeting of a non-industry-funded study of PA policies for both atypical antipsychotic medications and anticonvulsants among 6,712 patients with bipolar illness in Maine and New Hampshire.[1] Perhaps due to the effects of PA of multiple drug classes, we found a highly significant 2.28 greater risk of treatment gaps among bipolar patients in Maine, a stronger effect than that among patients with schizophrenia. Dr. Clifford’s case reports of physicians attempting to provide drug company-provided free drug samples to fill gaps in recommended treatment caused by PA, like the many Maine NAMI case reports of adverse outcomes of the policy, provide additional corroborating evidence of unintended effects of the PA policy. Drug samples are neither a reliable nor long-lasting fix for policy-induced disruptions in treatment, and may be used for marketing in addition to patient needs.[2] Dr. Clifford’s statement concerning unspecified rebates for use of preferred atypicals was already acknowledged in our paper. His economic argument, however, is unconvincing and does not mention the substantial administrative costs and hassle of PA for physicians, pharmacists, patients, and the state. Moreover, two-thirds of the study patients would now be in the Medicare Part D program where PA policies are prevalent, but the rebate amounts are as much as two-thirds lower than in state Medicaid programs.[3] The limited savings do not justify the observed treatment disruptions. As we report elsewhere,[4] other drug classes are better suited for PA. Several of Dr. Rosenheck’s arguments seem to be at odds with his recently published statements about sensible antipsychotic policy:[5] “We further outline a cautious implementation procedure that relies on standardized documentation and feedback, without a restrictive formulary, that would limit physician choice. . .. Because medication nonadherence can be a serious problem for people taking antipsychotic medication, procedures that might discourage adherence should be avoided and therapy should be individualized when indicated.” A close examination of the complex Maine PA policy indicates that it does not meet Dr. Rosenheck’s standards, or those of the recent consensus policy of State Mental Health Program Directors.[6] With regard to a few of the technical points, we never portrayed the observed PA effects as large, but they are worrisome. Observing even a modest effect is notable given that the PA was in effect for only 8 months, and its effects on therapy were increasing at the end of the policy. We agree that many schizophrenia patients discontinue or change treatment on their own (as also documented in Dr. Rosenheck’s studies), but our effects controlled for this threat to validity. Regarding the significance of the observed effects, Exhibit 2 shows a clearly visible effect, coincident with the timing of the policy, and using a strong controlled research design. Confidence intervals also back up the conclusions. We did not conclude that the policy affected clinical outcomes (such as hospitalization), as stated by Dr. Rosenheck. The new evidence of treatment disruptions in Maine Medicaid patients with bipolar illness [7] leads us to strongly support Dr. Rosenheck’s encouragement for nonrestrictive formularies, flexibility of physician choice,[8] and the avoidance of a panoply of unintended consequences.[9] References 1. Y. Zhang, A.S. Adams, D. Ross-Degnan, F. Zhang, and S.B. Soumerai, “Economic and clinical impacts of prior authorization for antipsychotic and anticonvulsant medications among Medicaid beneficiaries with bipolar disorder,” Society of General Internal Medicine accepted scientific abstract. In Supplement to the Journal of Internal Medicine, April, 2008 (in press). 2. S.L. Cutrona, S. Woolhandler, K.E. Lasser, D.H. Bor, D. McCormick, D.U. Himmelstein, “Characteristics of recipients of free prescription drug samples: a nationally representative analysis,” Am J Public Health 98 (2008): 284-9. 3. U.S. House of Representatives Committee on Oversight and Government Reform Majority Staff, “Private Medicare drug plans: High expenses and low rebates increase the costs of Medicare drug coverage,” October, 2007. Available at: http://www.allhealth.org/BriefingMaterials/HouseMajorityStaff-965.pdf (accessed on April 15, 2008). 4. S.B. Soumerai, “Benefits and risks of increasing restrictions on access to costly drugs in Medicaid,” Health Affairs 23 (2004): 135-146. 5. R.A. Rosenheck, D.L. Leslie, S. Busch, E.S. Rofman, M Sernyak, “Rethinking antipsychotic formulary policy,” Schizophrenia Bulletin 34 (2008): 375-80. 6. J. Parks, A. Radke, G. Parker, M-E. Foti, R. Eilers, M. Diamond, D. Svendsen, R. Tandon, “Principles of antipsychotic prescribing for policy makers, circa 2008. Translating knowledge to promote individualized treatment,” Schizophrenia Bulletin (advance access published online April 2, 2008), doi:10.1093?schbul/sbn019 7. Zhang et al, “Economic and clinical impacts of prior authorization.” 8. Rosenheck et al, “Rethinking antipsychotic formulary policy.” 9. S.B. Soumerai, T.J. McLaughlin, D. Ross-Degnan, C.S. Casteris, and P. Bollini, “Effects of limiting Medicaid drug-reimbursement benefits on the use of psychotropic agents and acute mental health services by patients with schizophrenia,” New England Journal of Medicine 331 (1994): 650-655; and Soumerai, “Benefits and risks of increasing restrictions on access to costly drugs in Medicaid.”