National Alliance on Mental Illness
page printed from
(800) 950-NAMI;

June 5, 2006

NAMI Comments to Agency for Healthcare Quality and Research on Comparative Effectiveness of Pharmacological Treatment of Depression Report

On behalf of the National Alliance on Mental Illness (NAMI), I am pleased to submit the following comments to the Agency for Healthcare Research and Quality (AHRQ) on the agency’s draft report “Comparative Effectiveness of Pharmacological Treatment of Depression.”

As the nation’s largest organization representing people with severe mental illness and their families, NAMI supports the AHRQ’s goal of validating evidence regarding the comparative efficacy the available treatments for depression.  In NAMI’s view it is critically important that consumers, family members, treatment professionals and payors (both private and public) be able to make informed decisions about treatment alternatives.  Comparative effectiveness reviews – when undertaken carefully, with full view of all the scientific evidence – offer tremendous promise in promoting better quality and outcomes.  These reviews can also generate new scientific evidence, support new research and promote translation of findings into forms that are more useful for consumers and families.

While NAMI supports the concept of Comparative Effectiveness Reviews, as well as AHRQ’s goals in undertaking them, this review of the current treatments for depression contains a number of conclusions that appear unsupported by the available evidence.  Of particular concern to NAMI is the report’s conclusion that there are no substantial differences among the second generation anti-depressants with respect to efficacy, effectiveness and other important measures.  NAMI feels strongly that there is a considerable body evidence rebutting this conclusion (discussed in detail below). 

NAMI urges that AHRQ amend and supplement this report to take all of this evidence into account, especially the recently released first and second stages of the STAR*D clinical trial on treatment resistant depression undertaken by your colleagues at the National Institute of Mental Health (NIMH).  Further, NAMI recommends that this draft be amended and supplemented to include a review of research related to the important issue of anti-depressants and their link to flight or switch to mania.  This research is clinically essentially and is not developed in this AHRQ review.   

NAMI is particularly concerned about the enormous potential for this report, and its conclusions, to be used by payors to justify policies that restrict access to the full range of available treatments.  This is especially the case with public sector programs such as Medicare and Medicaid, which typically serve a more disabled and vulnerable population.  In NAMI’s view, it would be tragic consequence if this report were used as a basis for additional policies such as “fail first,” step therapy and mandatory therapeutic substitution that are already commonly used in programs such as Medicare and Medicaid. 

Individuals living with major depression – especially those eligible for Medicare and Medicaid – are already dealing with a complex disease, and typically multiple co-morbidities.  Placing their treatment and opportunities at recovery at risk through restrictive policies is certainly inconsistent with AHRQ’s expressed goals of quality improvement and scientific advance.

NAMI would like to offer the following comments on the AHRQ draft report “Comparative Effectiveness of Pharmacological Treatment of Depression.”

1. Despite a lack of data comparing different drugs on key measures such as effectiveness, the report nevertheless took specific and limited findings regarding a few drugs and generalized them to the entire class

It is important to note that of the 66 possible comparisons of included second generation anti-depressants, the study authors found direct head-to-head evidence for only 30 comparisons (report at page 50).  Further, on the key question of comparative effectiveness, the report analyzed only three trials involving four medications (citalopram to sertraline, fluoxetine to sertraline and fluoxetine to sertraline and paroxetine).  It is noteworthy that the report’s authors did not find any evidence on the comparative effectiveness for other drug comparisons.  Despite this, the authors nonetheless concluded, without explanation “results from efficacy trials can be viewed as indirect evidence, indicating no substantial differences in effectiveness exists among second generation anti-depressants.”[1]

2. The data in the report that identified differences between drugs was either ignored, minimized or lacked scientific justification

As noted above, the report reaches a conclusion that there are no substantial differences in efficacy among the second generation anti-depressants.  However, the body of the report actually identifies specific examples where such differences exist.

  • A comparison of paroxetine and venlafaxine that found different response and remission at all points in time that reached statistical significance at week 12,[2]
  • A meta-analysis conducted in the study comparing the effects of flouxeine to sertraline found statistically significant effect rates,[3] and
  • A sensitivity analysis comparing venlafaxine and fluoxetine found significantly different response rates, a finding that appears consistent with another meta-analysis in the report.

3. The report does not reflect the limitations in both the quality and quantity of data in its conclusions and fails to caution policymakers, payors  and clinicians about these limitations as a basis for coverage, treatment and payment decisions

The authors of the report acknowledge that “most efficacy trials were equivalence studies; they were not powered to establish a greater efficacy of a particular drug but were designed to present equivalency in efficacy between the compared drugs.”[4]  As a result, the report states that “clinically significant differences might not reach statistical significance.”[5]  Despite this, the report fails to caution policymakers or payors (both private sector and public sector) that using the report to inform coverage, treatment or payment policies could easily result in restrictions on access to medications.  As noted above, the report cites several instances in which statistically significant differences were observed between medications, yet the report repeatedly minimizes the effects of such differences on clinical outcomes.[6]

At one point, the report concludes that based on the strength of evidence, second generation anti-depressants offer no differences in improving quality of life.[7]  At the same time, the report later specifically notes that “investigators rarely assessed quality of life and functional capacity; if they did; they typically considered these only as secondary outcomes.”[8]  At best, these findings and conclusions (the former contained in the report’s Executive Summary, the latter in the full report) are inconsistent with each other.

NAMI would also note that clear differences in patient specific information for medications relevant to prescribing are presented in this draft.  Differences in speed of onset, sexual functioning, withdrawal symptoms, weight change, and sleep impact are all well demonstrated in this review – even with the incomplete literature available.  These differences are very relevant to prescribing decisions and are good examples of why a wide range of access to medications should be preserved.

4. The report excludes a substantial body of scientific evidence demonstrating that anti-depressants are not interchangeable due to individual responses of individual patients to different compounds

SSRI anti-depressants all produce their therapeutic effect by acting on a single protein known as the serotonin transporter.   However, a significant body of research demonstrates that the response in individual patients is far more complex.  First, on a physiological level, each of the medications in the SSRI class has a unique profile of activity in various neurotransmitter systems.  Thus, there is no clinical or scientific basis to support a claim that “one drug fits all,” even in cases where patients are experiencing similar symptoms. 

For years, clinicians have known – and research has verified – that individuals with major depression that failed to achieve a good clinical response to one SSRI, either due to lack of efficacy, or tolerability, can nevertheless respond to a second SSRI with response rates for the second agent as high as 50%.[9]  This is the basis for numerous clinical treatment guidelines that have concluded that anti-depressants are not interchangeable.  In NAMI’s view, prescribers must do a careful assessment of a broad range of factors (treatment history, co-morbid conditions, family history, etc.) in working with the patient to find the best medication.  This is in stark contrast the unjustified conclusion of interchangeability that flows from the AHRQ report.

Further, NAMI believes that the review AHRQ has undertaken here actually demonstrates the need for policymakers and payors to move toward a policy of open access to the full range of anti-depressant treatments.  The report in fact identifies gaps in existing research studies associated with depression such as anxiety, insomnia, pain, fatigue, motor activity and responses based on race.  NAMI would urge AHRQ to use additional Comparative Effective Reviews on depression to prod research in these important areas. 

NAMI would also commend this report for highlighting the paucity of research on anti-depressant combinations at treatment initiation.  This is a good example of the limits of the current literature.  

5. Basing the report’s conclusions on drug to drug comparisons obscures the problem that the statistical power to make such comparisons may be limited

NAMI is concerned that the report casually reaches conclusions of “no difference” when weak statistical power in many comparisons should more accurately by categorized as “insufficient data” for the purpose of drawing any conclusion.  NAMI would suggest that meta-analyses included in the report be accompanied by estimates of the statistical power of each comparison.       


Thank you for the opportunity to submit comments on this important draft report.  NAMI urges that AHRQ consider amending and supplementing this report to take into account all the available scientific and clinical research on major depression.

Michael J. Fitzpatrick, M.S.W.
Executive Director

Ken Duckworth, M.D.
Medical Director

[1] AHRQ report at page 54.
[2] AHRQ report at page 74.
[3] AHRQ report at page 66.
[4] AHRQ report at page 43.
[5] AHRQ report at page 43.
[6] Note discussion in AHRQ report on page 59.
[7] AHRQ report at page 13.
[8] AHRQ report at page 52.
[9]Hensley, P, Nurnberg, H:  Formulary Restriction of Selective Serotonin Reuptake Inhibitors for Depression, Pharmacoeconomics 2001; Nurnberg, H, Thompson, P, Hensley, P: Antidepressant Medication Change in a Clinical Treatment Setting:  A Comparison of the Effectiveness of Selective Serotonin Reuptake Inhibitors, J Clin Psychiatry, September 1999.