Changing Rx Practices in the Treatment of Mental Illness: Impact on Forensic Evidence

"Off-label" prescription practices are increasingly used in the treatment of symptoms related to mental illness. In support of this conclusion, evidence is reviewed on the antidepressant treatment of numerous non-depressive disorders and on the antipsychotic drug treatment of non-psychotic disorders.

The impact of this evidence is discussed in light of the Daubert decision rendered by the U.S. Supreme Court. It is concluded that such practices challenge the application of the decision and increase the potential for testimonial error. Finally, a promising trend of identifying drugs by their neurochemical action is briefly discussed.

One trend that has revolutionized the face of Clinical Psychopharmacology is a breakdown in the disorder-specific nature of drug treatment (i.e., the use of antidepressant drugs to treat depression, antipsychotic drugs to treat psychosis, etc.). This "off-label" prescription practice has resulted in a clear broadening of the range of possible psychological symptoms that can be positively influenced by a specific class of drugs. The purpose of this article is to present examples of this practice in the published literature and suggest how this trend has impacted forensic considerations in the area of Clinical Psychopharmacology. Although no attempt has been made to exhaust coverage of this trend, an attempt has been made to present examples representative of different classes of drugs.

"Off-label" Prescription Use of Antidepressant Drugs

Although many classes of drugs originally designated for treatment of a specific mental disorder appear to demonstrate a broadening range of influence on other psychological symptoms, antidepressant drugs are particularly noteworthy. This may be due, in large measure, to the fact that there are several subtypes of antidepressant drugs with each subtype having its own profile.

The choice of antidepressant treatment of patients suffering from bipolar disorder (BD) finds support in two basic facts: (1) a significant number of patients with BD previously have been misdiagnosed as having unipolar major depressive disorder (MDD); and (2) patients with BD seem to have greater problems with depression than with mania (Ghaemi et al., 1999; El-Mallakh & Karippot, 2002). Evidence suggests that, although antidepressants (Tranylcypromine, Imipramine, Fluoxetine) have proven to be effective in the acute treatment of BD, they are also associated with a variety of adverse outcomes and a worsened course of bipolar illness (El-Mallakh & Karippot; Ghaemi et al.).

Although stimulants have been the standard choice for the treatment of Attention Deficit Hyperactivity Disorder (ADHD), there is an impressive body of literature also documenting the efficacy of tricyclic antidepressants (TCAs) in the treatment of such symptoms in more than 1000 subjects (Spencer et al., 1996). The TCA Desipramine is not only effective in treating children and adolescents with ADHD (Spencer et al., 2002), but in treating adult patients with ADHD symptoms as well (Wilens et al., 1996). In a somewhat similar manner, the effectiveness of the atypical antidepressant Bupropion compared favorably with that of Methylphenidate (the most popular stimulant choice) in the treatment of ADHD, with its benefits apparently not due to any specific antidepressant action. This conclusion is supported by the fact that Bupropion was an effective treatment for ADHD when there was an absence of comorbid MDD (Spencer et al., 2002; Barrickman et al., 1995).

Another example of the broadening influence of Bupropion beyond its antidepressant use has been its effectiveness in facilitating smoking cessation in double-blind, placebo-controlled trials (Hurt et al., 1997; Ahluwalia et al., 2002). However, the potential role of the drug's depressive action in these findings is less clear. In one case, a reduction in comorbid depression accompanied the smoking cessation produced by the drug (Ahluwalia et al.). However, in another case, no change in the effect of Bupropion on comorbid depression was observed to accompany its smoking cessation action (Hurt et al.). The role of drug-induced changes in comorbid depression in the smoking cessation produced by Bupropion may depend upon the specific level of depression. It may also be the case that the effectiveness of Bupropion in smoking cessation is due to the drug's unique action on norepinephrine-dopamine brain mechanisms associated with reinforcement properties and addiction (Hurt et al.).

Although anxiolytic drugs would appear to be the logical choice in the treatment of anxiety- related disorders such as General Anxiety Disorder (GAD), Panic Disorder (PD), Obsessive- Compulsive Disorder (OCD), and Post Traumatic Stress Disorder (PTSD), the positive influence of antidepressant drugs in the treatment of these disorders also has been well documented.

Both the antidepressant drugs Sertraline and Venlafaxine ER have proven, in double-blind, placebo-controlled trials, to be effective and well tolerated in the treatment of PTSD (Brady et al., 2000; Davidson et al., 2006a). Another recent study corroborated the findings with Venlafaxine ER (Davidson et al., 2006b) and is particularly noteworthy for the fact that the 6-month duration of the study is unique for double-blind, placebocontrolled evaluation of the drug's efficacy in treating PTSD.

Although the studies reviewed in this section clearly indicate the effectiveness of antidepressant drugs in treating clusters of non-depressive symptoms, one basic question remains-how do they produce such effects? Many of the studies do, in fact, have some comorbid depression present. However, it does not appear that the effectiveness of antidepressant drugs in such treatment contexts depends upon the result of an indirect action of these drugs on depression spilling over into the other symptom clusters. Rather, these antidepressant drugs appear to have effects specific to each of these disorders (i.e., anxiolytic effects, anti-OCD effects, etc.). Such a conclusion revolutionizes conventional labeling and traditional assumptions regarding antidepressant drugs.

"Off-label" Prescription Use of Antipsychotic Drugs

In addition to the above reviewed evidence for an expanded use of antidepressant drugs, there is also increasing evidence that antipsychotic drugs have a broadening range of influence on non-psychotic symptom clusters.

Although evidence suggests that typical antipsychotic drugs may have some effectiveness in the treatment of BD patients (Tohen et al., 2001), most of the findings in the literature have focused on the successful use of atypical antipsychotic drugs in such treatment. For example, the atypical antipsychotic drug Clozapine proved to be effective in the treatment of patients with either BD or schizoaffective disorder (bipolar subtype) for whom Lithium, anticonvulsants, or typical antipsychotics had been ineffective (Calabrese et al., 1996).

Other double-blind, placebo-controlled studies have demonstrated the effectiveness of the atypical antipsychotic drug Olanzapine in the treatment of acute mania symptoms in patients with BD (Tohen et al., 1999; Tohen et al., 2000). It is also of critical importance to note that there was no significant difference in the treatment advantage of Olanzapine for patients with or without psychotic features, ruling out the potential role of the drug's antipsychotic action as a contributing factor to the findings. Although in one of these studies (Tohen et al., 2000) the effectiveness of Olanzapine on bipolar mania was significant, there was some concern about the high placebo response rate (43%). However, the authors suggested that these findings may have been related to a large percentage of patients with rapid-cycling features and differences in trial duration in the placebo group (Tohen et al., 2000).

Further evidence in a double-blind, placebocontrolled study by Tohen et al. (2002) demonstrated that the addition of Olanzapine in BD patients (manic and mixed episodes) who had been inadequately responsive to more than 2 weeks of Lithium or Valproate therapy provided superior efficacy in the treatment of manic symptoms (Young Manic Rating Scale). While the patients in the Lithium or Valproate groups also showed some limited improvement in depressive symptoms (9.5%), patients in the Olanzapine cotherapy group showed significantly greater improvement (>50%). Although improvement in depressive symptoms was significantly greater in the cotherapy group, so were adverse events (somnolence, weight gains, etc.).

One study having a major impact in the medical literature is the double-blind, placebo-controlled study by Tohen et al. (2003), which examined the effects of Olanzapine and an Olanzapine-Fluoxetine combination in the treatment of BD. They found that Olanzapine was more effective than placebo, and that combined Olanzapine-Fluoxetine was more effective than Olanzapine or placebo in the treatment of BD. Although these findings would seem to suggest that Olanzapine therapy significantly improved depressive symptoms in patients with BD and that the Olanzapine-Fluoxetine combination had an even more robust antidepressant effect, these conclusions have more recently been challenged (Moreira- Almeida & Pietrobon, 2006). This challenge was based upon two primary characteristics of the study by Tohen et al. (2003): (1) Although the difference in the Olanzapine group was statistically significant, it is questionable whether this difference should be considered clinically meaningful (39% vs. 30% difference); and (2) Some of the improvement in depressive symptoms may simply represent side effects associated with Olanzapine. However, other evidence cited in this section suggests that the effectiveness of the drug in the treatment of BD is well established.

In addition to the effectiveness of Olanzapine in the treatment of BD, Keck et al. (2003a, 2003b) have demonstrated the effectiveness of the atypical antipsychotic drugs Aripiprazole and Ziprasidone in the double-blind, placebo-controlled treatment of bipolar patients (manic or mixed episodes). Aripiprazole was significantly superior to placebo in reducing acute mania in all primary and secondary efficacy variables (Keck et al., 2003a). Patients with schizophrenia and schizoaffective disorder were excluded from the study, again eliminating the potential influence of antipsychotic action of the drug as a contributing factor to the findings. In the other study (Keck et al., 2003b), Ziprasidone had significantly greater efficacy than placebo in the treatment of BD in patients with acute mania or mixed episodes. Again, patients with schizophrenia or schizoaffective disorder were excluded from the study. These latter findings appear to be quite impressive, but they are not without their critics. For example, Jagadheesan and Muirhead (2004) suggest that the findings by Keck et al. (2003a) on the effectiveness of Aripiprazole in treating bipolar patients are questionable for two reasons: (1) assessment variation due to the multi-center nature of data collection; and (2) the high attrition rate observed in both the Aripiprazole and placebo groups (58% and 79%, respectively). Although these concerns are deserving of mention, they do not appear to negate the evidence for the effectiveness of Aripiprazole in the treatment of BD.

The picture that appears to emerge as a result of reviewing the effectiveness of atypical antipsychotics in the treatment of BD suggests the following conclusion. These drugs are effective in the treatment of both the depressive and manic phases of BD and that such effectiveness is not due in any way (directly or indirectly) to the antipsychotic properties of the drug. It would seem appropriate to define these effects on BD as specifically antidepressive and antimanic.

In addition to its success in the treatment of BD, the atypical antipsychotic Olanzapine also has been found to be effective in the treatment of Borderline Personality Disorder (BPD). In a double-blind, placebo- controlled study of 6-month duration, Olanzapine proved to be a safe and effective agent in the treatment of women with BPD (Zanarini & Frankenburg, 2001). In a more recent double-blind, placebo-controlled study with a mixed sample of women and men, Olanzapine was again found to be significantly superior to placebo in the treatment of BPD (Bogenschutz & Nurnberg, 2004). In both of the above studies, weight gains (a potential problem with long-term management) were significant in the Olanzapine group.

An improvement in BPD symptoms has not been limited to Olanzapine but also has been demonstrated with the additional atypical antipsychotic drug Aripiprazole. In a double-blind, placebo-controlled study, Nickel et al. (2006) demonstrated that Aripiprazole was an effective and safe treatment for patients with BPD. No significant weight gains were observed, although other common side effects (headache, insomnia, nausea, etc.) were present.

The body of evidence supporting the broadening range of influence of atypical antipsychotic drugs on non-psychotic symptoms is quite impressive. Furthermore, the effectiveness of such drugs in improving non-psychotic clusters of symptoms may be due to their unique neurochemical action on dopamine and serotonin (DA/5-HT) systems (Schmidt et al., 2001), which may have a stabilizing effect. Dysfunction (destabilization) in these systems may underlie or cause the symptoms associated with a broader spectrum of non-psychotic features.

Forensic Impact of Findings

The value of an expert witness or the testimony offered by such a witness in a forensic contest clearly depends upon the reliability and clarity of evidence presented. That is to say, such testimony should be based on language that communicates information accurately. It would appear that the material reviewed in this article on the "off-label" prescription practices has, to some degree, weakened the accuracy of forensic evidence related to drug labeling in the treatment of mental illness.

The "off-labeling" prescription practices reviewed in the present article would appear to be particularly critical in view of the United States Supreme Court's ruling in the Daubert decision (Daubert v. Merrell Dow, 1993). The Daubert decision establishes a new set of criteria for courts to determine the admissibility of evidence. In addition to giving court judges a "gate keeping" function in determining the admissibility of evidence, the decision outlined four specific guiding principles for assessing the reliability of scientific evidence (Rast, 2006). These principles include testability, error rates, peer review, and general acceptance. Under Daubert, drug evidence is established by meeting operational definitions and scientific evidence, not labels. Accordingly, "off-label" prescription practices would appear to challenge the application of such principles, particularly in the case of error rate-actual or potential. For example, an expert witness using the outdated previous labeling practice in the process of giving testimony conveys inaccurate or, at the very least, incomplete information about the use and range of pharmacological action of such drugs in the treatment of mental illness. Forensic scientists offering testimony on the pharmacological treatment of mental illness need to be aware of the Daubert guidelines and adjust their testimony accordingly.

The "off-labeling" prescription practices discussed in this article have, in a very real sense, created a dilemma in how a practitioner identifies or labels a particular drug in the treatment of mental illness. Because original labels (anxiolytic, antidepressant, antipsychotic, etc.) appear to no longer communicate accurately the treatment choice or treatment effect of drugs, how should one identify such medication? While no definitive answer to this question has yet emerged in the literature, one trend does appear to be promising-the tendency to identify the clinical use of drugs by their mode of neurochemical action rather than by symptom labeling. Consistent with this trend, it may also be appropriate to reclassify drugs previously designated by their neurochemical action and symptom label (e.g., SSRI-antidepressants) simply by their neurochemical action without the designation of antidepressant (SSRIs). In a similar manner, it might also be appropriate to reclassify atypical antipsychotics as simply DA/5-HT stabilizers. It is recognized that this practice of re-labeling drugs used to treat mental illness by their neurochemical action is quite revolutionary and would require a considerable amount of re-orientation on the part of clinical practitioners.

Concluding Comments

There is little doubt that the "off-label" prescription practices outlined in this article have weakened the credibility and accuracy of testimonial evidence and decreased the value of such evidence in a forensic context. This coupled with the fact that the guidelines established by the Daubert decision have further specified the criteria that must be met in determining the admissibility of evidence should cause a potential forensic witness to re-examine his/her strategy.

At the very least, such practices should cause potential witnesses to become more sensitive to the labeling of drugs and what such labels communicate. In those instances where labels are prone to increase the probability of informational error, the potential witness should be forthright in his/her testimony, acknowledging the uncertainty of conclusions drawn from such language. The choice of words used to convey the degree of certainty (or uncertainty) associated with drug designations becomes extremely important in one's testimony. Honesty and forthrightness are still the best policy in such a situation. And it is important to remember that the game is still an adversary system of justice. Although serving as an expert witness in the drug treatment of mental illness can be an exciting and challenging role, thorough preparation by a potential witness will increase his/her credibility and allow the witness to speak forthrightly, without the threat of reprisal.