8

Non-Stimulant Medications and Non-Pharmacological Treatments

There are two things to note here: (1) there are very few drugs outside of the psychostimulants that can be used in ADHD; and (2) non- pharmacological or psychotherapeutic approaches have been recommended but are rarely specific to ADHD, e.g. parent training classes such as those supported by NICE. The lack of alternative drugs to methylphenidate and the fact that the likelihood of accessing the plethora of psychotherapeutic services is low necessitates that this is a comparatively short chapter!

Non-Stimulant Medications

The role of drugs other than the psychostimulants is limited. Certainly a patient may take more than one drug, and, given that comorbidity is the rule rather than the exception, many other drugs will be used to treat the other comorbid disorders.

There are a small number of drugs that are referred to as non-stimulants that are use in the treatment of ADHD. The most notable is atomoxetine. This is a relatively new drug in the treatment of ADHD and warrants closer analysis than the others. And as with methylphenidate and amphetamine, atomoxetine might help us understand the underlying pathology in ADHD.

Atomoxetine

Atomoxetine or Strattera is marketed by Eli Lilly & Co. and is the first non-stimulant approved for ADHD. One of the selling points is that it is not a controlled substance and is not addictive. Whether it works or not is another matter. Certainly atomoxetine takes time to become effective, and the dose needs to be systematically increased to avoid side-effects. Unlike methylphenidate, which in its immediate-release form takes 20–30 minutes to start working, atomoxetine takes three to four weeks, with the maximal effect seen as late as six to eight weeks. This is clearly a long time to wait to see if a drug is going to be effective.

Atomoxetine is a highly selective noradrenergic reuptake inhibitor [927–928]. It is similar to methylphenidate, without the DA component yet with a similar effect on NA. Amphetamine and methylphenidate both have effects on noradrenergic reuptake, but also affect DA reuptake. It has been suggested that the noradrenergic component may be the critical therapeutic target and the underling substrate of pathology [737, 929].

The increase in noradrenergic activity in response to atomoxetine was demonstrated in several regions of the brain that differentiated it from methylphenidate; most notable was the increase of NA and DA in the prefrontal cortex [930].

Owing to the comparatively recent introduction of atomoxetine, there is less information regarding its clinical efficacy and safety. Atomoxetine was first identified as an effective treatment for ADHD in a group of 22 adults with ADHD who participated in a double-blind placebo crossover study in which they acted as their own control [931].

The clinical efficacy of atomoxetine has been demonstrated in children and adults in numerous placebo comparison studies [932–938], and meta-analyses further support its use [939–940]. The effectiveness of atomoxetine has been shown to be long-term (8–24 months) without tolerance developing [941–942]. The action of atomoxetine on measures of behavioral inhibition has also revealed a positive effect [457, 943] which was associated with increased activation in the right inferior frontal cortex [944].

The effectiveness of atomoxetine over the psychostimulants has been questioned [945]. Prior to its use in the UK, there was only one study that compared atomoxetine to methylphenidate. Kratochvil et al. [946] found that there was no difference in therapeutic effect between methylphenidate and atomoxetine. This has since been supported by a multicenter study spanning China, Mexico, and South Korea [947]. However, these results are only on the bases of rating scales and the methylphenidate group was too small to detect differences. Kemner et al. [948] and Starr and Kemner [949] in a community-based open-label study, in which both the researchers and participants know which treatment is being administered, demonstrated that there was a greater improvement with methylphenidate (Concerta) compared to atomoxetine. A similar benefit for Adderall compared to atomoxetine was also found [950]. In a placebo-controlled double-blind study, methylphenidate was better at treating ADHD than atomoxetine, but both were better than placebo [951].

Given that until the emergence of atomoxetine it was only amphetamine and methylphenidate that were used in ADHD, a related question to the above studies is: what are the effects of previous exposure to psychostimulants? There is a high chance that patients will have experienced considerable pharmacological histories, which may affect response to atomoxetine. In a placebo-controlled double-blind study of over 500 children, among those with a history of stimulant treatment there was a response rate of 37 percent for atomoxetine compared to the 57 percent in patients who were stimulant-naïve at the start of the study [951].

Whilst one may begin to assume that atomoxetine is effective, but just not as effective as methylphenidate, we have to remember that approximately 25 percent of patients with ADHD fail to respond to methylphenidate and/or amphetamine. Atomoxetine therefore has a potentially important role to play in these individuals. In the Newcorn et al. study above [951], of those who did not respond to methylphenidate in the trial, 43 percent subsequently had improvements with atomoxetine. Two other studies have provided some support, but the methodology does not permit a definitive conclusion [952–953].

In general, atomoxetine is safe and well tolerated [954–955]. Unlike methylphenidate and amphetamine, it does not have an abuse profile [956–958] and has been seen to block the effects of amphetamine [959] but not cocaine [960–961]. This is good news in general and may well mean that it is a good drug to use in comorbid substance abuse cases [962]. However, in 2005, the US Food and Drug Administration (FDA) directed Eli Lilly & Co., the manufacturer, to revise product labeling to alert users and prescribers of the increased risk of suicidal thinking in children and adolescents. The packaging now contains the following:

Atomoxetine increased the risk of suicidal ideation in short-term studies in children or adolescents with Attention-Deficit/Hyperactivity Disorder (ADHD). Pooled analyses of short-term (6–18 weeks) placebo-controlled trials of STRATTERA in children and adolescents (a total of 12 trials involving over 2200 patients, including 11 trials in ADHD and 1 trial in enuresis) have revealed a greater risk of suicidal ideation early during treatment in those receiving STRATTERA compared to placebo. The average risk of suicidal ideation in patients receiving STRATTERA was 0.4% (5/1357 patients), compared to none in placebo-treated patients (851 patients). No suicides occurred in these trials.

In a meta-analysis of 14 trials with atomoxetine, no one committed suicide although there was more thought of it [963]. Some case studies point to suicidal thoughts as a problem with atomoxetine [964–965]. Atomoxetine is pharmacologically similar to antidepressant and approximately 1 in 50 children prescribed antidepressants have increased thoughts of suicide [966]. Strong conclusions about suicidal thoughts and the translation into suicidal actions needs a systematic investigation, which has as yet not been conducted [967].

It is still comparatively early days in the use of atomoxetine for ADHD. The addition of atomoxetine is clearly beneficial and should be welcomed. It provides an alternative to the psychostimulants and hope for some of those who do not respond to them. Just as important is the role atomoxetine has in the theoretical speculation surrounding ADHD neuropathology. Atomoxetine provides a new focus, away from DA and the striatum. Of course this may not herald great advances; after all, we are building a case that ADHD is a heterogeneous disorder with possibly numerous causal pathways. What is now needed is controlled randomized double-blind studies with comparisons against placebo and methylphenidate before concrete conclusions can be made.

Modafinil

Modafinil perhaps should not appear in this section because it is a psychostimulant that has been used for its arousal-inducing properties and treatment of narcolepsy. However, because it is unlike amphetamine structurally, neurochemically, and behaviorally, I shall consider it here. According to Minzenberg and Carter [968], modafinil has been demonstrated to directly bind and inhibit the DAT and noradrenaline transporter, NAT. However, modafinil’s pharmacological profile also leads to elevated DA, NA, serotonin, glutamate, and histamine levels, and decreased GABA levels. These effects are notable in the cortex, but minimal in various subcortical areas. Thus modafinil is a complex drug that does not work in the same regions of the brain as is suggested by methylphenidate and amphetamine. However, its arousal-inducing effects may be beneficial when one considers Rapport’s view of hyperactivity as a result of cortical under-arousal (see chapter 4).

In randomized double-blind and placebo-controlled trials, evaluation of ADHD after modafinil indicated improvements over that of placebo [969–975]. Modafinil improves performance on measures of behavioral inhibition in ADHD [974], healthy volunteers [976], and rats [977].

Despite a potential use of modafinil in ADHD, the American FDA declined to support its use in ADHD until more tests had been conducted. Their concern was a link with a sometimes fatal skin disease called Stevens–Johnson syndrome.¹

Clonidine and guanfacine

Clonidine (Catapres) and guanfacine (Tenex) are both agonists at the noradrenergic Alpha2 receptor. Guanfacine is a long-acting drug with a less problematic side-effect profile compared to clonidine.

These two drugs have been deployed for the treatment of ADHD and comorbid aggression and/or tics [978].

Early studies found a benefit of clonidine in treating ADHD when compared to placebo [979–980]. However, the effectiveness of clonidine was not seen across all dimensions and was dependent upon the measure being used [981]. A meta-analysis of clonidine in ADHD has found evidence to support its use, but only as a secondary medication after stimulant use [982]. Clonidine may well be used together with methylphenidate to address conduct problems in ADHD [983]. With the introduction of atomoxetine, the use of clonidine has decreased [984].

There are not many studies looking at the effectiveness of guanfacine, and many are not properly controlled. Open-label studies indicate that guanfacine is well tolerated and effective [985–990]. In a randomized double-blind placebo-controlled study, guanfacine was shown to be more effective than placebo in treating ADHD [991] and with comorbid tic disorder [992]. The most notable side-effect is sedation [993].

In the spontaneous hypertensive rat (SHR) model of ADHD, guanfacine has been shown to reduce over-activity and impulsivity whilst increasing sustained attention [994], and in the rat guanfacine can decrease neuronal activity in the striatum whilst increasing activity in the frontal cortex [995].

How Do These Drugs Fit into Theoretical Accounts of ADHD?

The short answer is that they do not fit well with some of the pre-existing theories on the neuropharmacology of ADHD. The psychological theories fare better as they do not always specify a neural system at fault. Robert Oades has reviewed the literature on NA (and serotonin) and supports a role of for it in attention by direct effects and interactions with DA [737, 926, 929]. Whilst there is some evidence to support such a view, that evidence is not as abundant as has been provided for a dopaminergic involvement in ADHD. Why might this be the case? Firstly, not until recently has there been a treatment for ADHD that has not had a dopaminergic action. Secondly, such ideas fly in the face of the existing wisdom on ADHD and are less likely to receive funding and therefore there is less work to publish, thereby maintaining the senior position adopted by the DA hypotheses. Thirdly, we might wish to consider that because atomoxetine does not appear to be as good as methylphenidate, then it is not as good a search tool for further understanding. One may consider that the teasing apart of the action of methylphenidate by using atomoxetine to act purely on NA systems compared to DA and NA systems is a valid endeavor. It may be that the heterogeneous nature of ADHD and the different symptoms are mediated by different neurochemicals.

Non-Pharmacological Treatments

This book is less concerned about behavioral interventions. After all, there is a whole industry dedicated to the publication of self-help manuals. The fact that there are so many self-help manuals is evidence of a failing in service provision. As we have seen in the MTA study (chapter 7), there are long-term benefits of behavioral interventions. Furthermore they may be extremely beneficial for some of the secondary symptoms of ADHD, such as poor social interactions and low self-esteem. Indeed patient-centered treatments such as social skills training can combat some of these problems extremely well!

Diller and Goldstein view the treatment of ADHD in a social/political context and state that

the controversy over treatments for ADHD is yet another reflection of the nature/nurture debate. With ADHD, researchers and leaders in the field of child psychiatry, psychology and pediatrics continue to fight a rear-guard battle against the legacy of a half a century of blaming mothers associated with the Freudian hegemony in our society. While remnants of the Freudian model remain viable, it is time to declare the battle over. However, insisting that the basis for behavior in children and adults is only biological and driven by heredity is simplistic, reductionistic and in fact does not fit the emerging research concerning gene/environment interaction.

[996] (p. 574)

We are not entirely at the mercy of our genes!

It is striking to note that psychosocial/behavioral interventions do not come under the same close scrutiny as drugs. Studies are unlikely to be placebo-controlled double-blind randomized trials for the likes of cognitive behavioral therapy (CBT), and for good reason. A review by Everitt and Wessley [23] points out that trials of psychological interventions can never be blind and treatment is dependent on the faith one has in one’s therapist. The therapist and patient are all aware of the treatment that is being received; there is no placebo comparison group. Furthermore, because the treatments are applied by different practitioners, there are the effects that some therapists are better than others – a confounding variable that interferes with the theoretical position of the intervention. And then there is the thorny issue of what aspect of the therapy is effective. Much may go on in a therapeutic session, but what exactly is effective remains an unknown.

A recent meta-analysis of 114 published studies using various behavioral treatments afound that there is a sufficient effect size of the interventions to warrant their use [997]. Similarly Toplak et al. [998] have found small effect sizes for cognitive and cognitive behavioral interventions, but argue that the methodologies are limited in many of the studies. Another meta-analysis compared behavioral/cognitive interventions with methylphenidate and did not find the behavioral/cognitive interventions to be as effective as the drug, but this did depend on the measure, with ADHD symptoms responding well to drugs and social behavior being equally affected by both interventions [999]. Like the drug therapies, psychotherapies and behavioral interventions are only as good as long as they are actively practiced. Furthermore, the “generalization of treatment effects across settings and over time – the overarching clinical objective of psychosocial interventions – remains an elusive goal” [1000] (p. 207).

One might assume that behavioral/cognitive or talking therapies are far removed from the neurobiological activities of the brain. Some have argued that they are just placebo effects. Certainly the main proponents of pharmacotherapy and psychotherapy are often seen to be at the opposite ends of a treatment spectrum. However, it is now becoming apparent that psychotherapies have a neurobiological underpinning [1001]. With the advent of neuroimaging techniques, changes that are brought about by therapy can now be seen in glorious color in neuroimages. Studies in depressed or anxious patients have seen several neural changes after psychotherapeutic interventions [1002–1005]. In an editorial published in 2006, Beitman and Viamontes state that

the psychotherapeutic process is an effective, structured, and neurobiologically definable method for decreasing the probability of the reactivation of old, maladaptive neural patterns, and catalyzing the creation and maintenance of a new set of adaptive synaptic patterns by which a patient can make sense of the world and respond effectively to its challenges.

[1006] (p. 220)

Of course there are limitations: the limitations of what is the therapeutic component and also the limitation of neuroimaging. This is interesting work and much remains to be done, and one is limited to human studies for some interventions as animal models are not available for a bit of motivational interviewing and CBT.

The two main interventions other than drug treatment are behavioral therapy and cognitive behavioral therapy. Despite the prominence of CBT in the treatment of psychiatric disorders, alone it has not been especially effective in ADHD, but does provide benefit when combined with behavioral therapies [1007].

In order to be effective, interventions need to be deployed consistently and across setting, and for the child that means at home and school. Therefore parent and teacher training in the management of ADHD is essential. The majority of training involves the use of the tenets of the behavioral therapies. A large body of literature has demonstrated the effectiveness of such an approach [1008–1009].

Behavioral therapy uses the principles of operant conditioning in which behaviors are modified by reinforcement. A behavior can be strengthened (i.e. a good behavior) by a positive reinforcer (i.e. a reward) or by a negative reinforcer (i.e. the escape or avoidance of something unpleasant). Punishment is different. Punishment is designed to reduce a particular behavior by imposing sanctions or aversive events (e.g. the old-fashioned smack).

This is entirely the essence of behavioral management; there are subtleties of delivery, but the main point is to reinforce desirable behavior and ignore undesirable behavior (easier said than done!). In chapter 9 we shall see that those with ADHD have reward deficiencies. That is not to say that behavioral techniques do not work at all, but rather they need to be different from standard delivery protocols. The person with ADHD will need larger and more powerful consequences and more prompts, cues, and reinforcement [1007].

Silver [1010] points out that the successful utilization of behavioral techniques requires a concerted effort from all those involved to deliver a program consistently and by using positive reinforcement. To this end, parent training is useful as the parents of ADHD children tend to have a more negative and directive parenting style [93]. However, one should also remember that parent style is not considered a causal factor in ADHD, though it may exacerbate some symptoms [1007].

In a review of parent training initiatives Sinclair [1007] identifies several common features:

1 Psychoeducation: understanding the disorder the child has.

2 Behavioral management and the ABC method: Antecedents – Behavior – Consequence (i.e. precipitating factors, the behavioral reaction and the consequence that reaction has to change the environment).

3 Acknowledging and responding to adaptive behaviors.

4 Rewards: immediate, novel and meaningful to the child.

5 Rewarding and ignoring: praise the good and ignore the bad.

6 Using positive skills and using behaviors that occur that can be used as a reinforcer, such as TV viewing.

7 Giving effective commands. These need to be short and specific, i.e. brush your teeth, not brush your teeth, get dressed for school, and pack your bag (too many instructions for the person with ADHD and a weak working memory).

8 Consistent behavioral rules.

9 Problem solving.

Cognitive behavioral therapy accounts for the complexities of behavior in a way in which the behavioral theories do not. CBT introduced the underlying cognitions/ideas that the person has as part of the treatment process. The point of CBT is to train the patient to recognize and manage the symptoms of ADHD. For example STOP-THINK-DO teaches children problem solving and self-monitoring, which should reduce impulsivity. Anger management is another form of CBT that is more commonly discussed and can be used in ADHD to deal with frustrations and resultant aggression.

In adults, group-delivered CBT has been shown to be moderately effective, with a treatment community being rated favorably by patients [1011]. The Young–Bramham [1012] method of intervention for adults uses several types of process to bring about change: CBT, psychoeducation, motivational interviewing (which is a method of facilitating change using empathy and highlighting the mismatch between behavior and goals), and cognitive remediation. Cognitive remediation is used mainly in brain injury and stroke and seeks to retrain the brain using neuropsychological tasks which target: attention and concentration; memory; planning; monitoring one’s work or behavior; and making adjustments based on feedback. Think along the lines of Brain Training. There is some evidence to support cognitive remediation in ADHD due to the similarities in some cases with brain injury and the behaviors they show [1013].

The Young–Bramham method attends to the core symptoms of ADHD and secondary problems such as social skills deficits. It is a modular program and addresses the following [1014]:

1 Coping with inattention and memory deficits.

2 Coping with impulsivity.

3 Time management.

4 Problem solving.

5 Social relationship skills.

6 Coping with feelings of anxiety.

7 Coping with feelings of anger and frustration.

8 Coping with feelings of depression.

9 Sleep problems.

10 Alcohol and drug misuse.

11 The future – learning to live with impulsivity and inattention.

Metacognitive therapy is based on the assumption that worry and rumination are processes leading to emotional disorder and are linked to incorrect beliefs about thinking and unhelpful self-regulation strategies. Recently Mary Solanto and colleagues have indicated that this may be effective in ADHD, especially on inattention and EF skills [1015].

Psychotherapeutic interventions clearly have a logical theoretical base upon which they are devised. However, unlike methylphenidate, which takes only up to 30 minutes to work, the psychotherapeutic strategies take considerably more time and effort before an effect is seen. The MTA study eventually highlighted the use of an effective and intense behavioral program for ADHD. But one has to remember that the cost of such interventions may well be considerably more than pharmacotherapy. Such costs depend on the economics that prevail. The NICE guidelines of 2008 now address the issue of parent training, although there are those who have said its recommendations need to be more specific and tailor-made towards ADHD (see chapter 10).

Alternative Treatments

The term “alternative treatment” brings to mind herbal remedies and homeopathic cures. However, an alternative treatment is a catch-all title for a diverse set of so-called treatments.

There are numerous interventions that do not fall within the mainstream of treatments for ADHD. You would be unlikely to be offered a course of neurolinguistic therapy (NLP) or neurofeedback. Whilst there are numerous interventions that are beyond the scope of this book, they are nevertheless interesting. One of the problems of deciding about these treatments is the limited evidence-based research and direction available. Some of the symptoms of ADHD are responsive to neurofeedback, a technique in which EEG signals are controlled by the individual [1016–1017]. Meditation techniques may also be of use, with measures of cognition as well as symptoms showing improvement, but clinical trials are required [1018]. A computerized progressive attentional training (CPAT) program composed of tasks that activate sustained attention, selective attention, orienting of attention, and executive attention has been seen to enhance academic abilities and parents’ rating of symptoms in ADHD children [1019].

Summary

The use of atomoxetine has changed the landscape of ADHD treatment from only the use of stimulants. It has opened up a dialogue to further enhance the understanding of ADHD, and of the brain in general. New drugs will be sure to come; the profits are too great for the pharmaceutical companies to want to miss out. It is important for the science of ADHD to have as many drugs available to treat ADHD; with many pharmacologically different drugs available, the common mechanisms will be more accurately detected. At present we are limited to methylphenidate, amphetamine, and atomoxetine. When it comes to the psychotherapies, it is less clear as to what the active ingredient is that is beneficial. CBT and parent/teacher training programs have been shown to be effective if delivered correctly. They have also been shown to be of benefit alongside drug treatment. It is interesting to note that a large body of the literature on non-pharmacological interventions/treatments acknowledges that the evidence is in need of systematic controlled studies along the lines of clinical trials. And even with such a large choice of potential alternatives to drugs, the patient/parent is unlikely to be offered much more than medication, despite guidelines.

Note

1 http://www.washingtonpost.com/wp-dyn/content/article/2006/03/23/AR2006032301688.html.