Effect size of different forms of ADHD treatment

The Effect size of a treatment is the value by which the symptoms improve.

The magnitude of the Effect sizes found is described using standard criteria.¹
One measure in which Effect size is measured is SMD (standardized mean difference, Cohen’s d): small = 0.20, medium = 0.50, large = 0.80.
A “small” effect is usually difficult to observe in an individual, but can be very important for public health if it is a general exposure affecting many individuals.
A “medium” / “moderate” effect should be perceptible to an attentive observer. From 0.5 clinical benefit.
A “large” / “strong” effect represents a clearly noticeable improvement.

Medication (especially stimulants) has the greatest effect in the treatment of ADHD.
ADHD medications have a special position within psychiatric medications. Compared to the usual Effect size of psychiatric medications, the Effect size of stimulants for ADHD is exceptionally high (AMP (SMD approx. 1.1 to 1.5) and MPH (approx. 0.9 to 1.1)).²³ . These are the highest Effect sizes found for psychiatric medications.⁴ The average Effect size of psychiatric drugs is 0.49 (SMD).⁵
Behavioral therapy alone is nowhere near as effective as medication alone, with an Effect size of 0.3² to 0.34⁶
A combination of medication and intensive behavioral interventions did not perform significantly better than medication alone in the MTA study of n = 579 children between the ages of 7 and 9.⁷⁸ For impulsive-aggressive symptoms and emotional disorders, medication and behavioral therapy were equally effective.⁹ A more recent comprehensive meta-analysis of k = 190 studies with n = 26,114 participants with ADHD-HI came to comparable results.¹⁰

It has been reported that medication during psychotherapy increases the learning effectiveness of psychotherapy.¹¹ As we understand it, in many cases they establish the ability to undergo therapy in the first place, as stimulants eliminate the dopamine deficit and thus restore the neurotrophic effect of dopamine, which is necessary for the ability to learn and supports the plasticity of the brain. ⇒ Neurophysiological correlates of learning problems in ADHD

Medication only works for ADHD for as long as it is given. Psychotherapy, environmental intervention and psychoeducation, on the other hand, have a long-term effect and continue to work beyond their specific application.

The Effect sizes mentioned in this article¹²¹³ generally refer to SMD values (Standard mean difference)¹⁴. SMD is a value that makes different studies comparable with each other. However, it always depends on what is being compared, as SMD represents a comparative value, so that a treatment method has a completely different SMD compared to placebo than compared to a “standard treatment” or compared to another treatment method. Nevertheless, the data as a whole gives an impression of the comparison of efficacy between the different drugs and forms of treatment.

When assessing studies on Effect sizes, a distinction must be made between unblinded and blinded evaluations.¹⁵ Unblinded evaluators are biased and overestimate treatment effects.¹⁶ In parent ratings, knowledge of the intervention increases parental tolerance of ADHD and/or their ability to cope with the negative effects rather than reducing symptoms.¹⁷ Even more so, parental involvement leads to a bias (“the effort must have been worth it”), which is why parent ratings should be viewed with particular caution. Teacher ratings are much more neutral in comparison.

Effect sizes in the headings indicate the range of studies / meta-study results, with the lowest and highest values omitted to limit bias due to outliers. Furthermore, values are only given in the headings if a sufficient number of studies or meta-analyses allow a reliable statement to be made.

• 1. Effectiveness of the treatment methods
  • 1.1. Comparison by Effect size (SMD)
    • 1.1.1. Drug treatment according to Effect size
      • 1.1.1.1. Stimulants: 0.8 to 1.5
        • 1.1.1.1.1. Amphetamine drugs: 1.1 to 1.5
        • 1.1.1.1.2. Methylphenidate: 0.9 to 1.1
        • 1.1.1.1.3. Mazindol
      • 1.1.1.2. Total non-stimulants: 0.52 to 0.71
        • 1.1.1.2.1. Guanfacine monotherapy: 0.67 to 0.76 for children and adolescents
        • 1.1.1.2.2. Combination therapy guanfacine with stimulants: plus 0.36 g compared to stimulants alone
        • 1.1.1.2.3. Clonidine: 0.38
        • 1.1.1.2.4. Atomoxetine: 0.63 children, 0.45 adults
        • 1.1.1.2.5. Bupropion: 0.32 to 0.46
        • 1.1.1.2.6. Modafinil: 0.62 to 0.76 KiJu
        • 1.1.1.2.7. Viloxazine
        • 1.1.1.2.8. Centanafadine
        • 1.1.1.2.9. Dasotraline
        • 1.1.1.2.10. Cannabinoids, Sativex
        • 1.1.1.2.11. SGA (second generation of antipsychotics) for hyperactivity in ASD
        • 1.1.1.2.12. Desipramine
        • 1.1.1.2.12. Tipepidine
        • 1.1.1.2.13. Nortriptyline
        • 1.1.1.2.14. Selegelin
    • 1.1.2. Effectiveness of non-drug therapies for ADHD according to Effect size
      • 1.1.2.1. Sports (endurance training): approx. 0.8 (0.62 to 1.96)
      • 1.1.2.2. Mindfulness-based behavioral therapy (MBCT)
      • 1.1.2.3. Cognitive behavioral therapy: 0.5 (0.3 to 0.85)
      • 1.1.2.4. Neurofeedback: 0.29 to 1.2
      • 1.1.2.5. Parent training: 0.31 to 0.68
      • 1.1.2.6. Repetitive transcranial magnetic stimulation (rTMS) 0.54
      • 1.1.2.7. Cognitive self-regulation training 0.54
      • 1.1.2.8. Behavioral training 0.54
      • 1.1.2.9. Food diets: 0.51
      • 1.1.2.10. Cognitive training 0.22 to 0.45
      • 1.1.2.11. Total psychological behavioral interventions: 0.35 for children and adolescents
      • 1.1.2.12. Social skills training 0.26 to 0.31
      • 1.1.2.13. Probiotics: 0.25
      • 1.1.2.14. Transcranial direct current stimulation (tDCS) 0.23
      • 1.1.2.15. Omega-3/-6 / PUFA supplementation: 0.11 to 0.17
      • 1.1.2.16. Zinc
      • 1.1.2.17. Acetyl-L-carnitine
      • 1.1.2.18. Phosphate dilyserine
      • 1.1.2.19. Iron
      • 1.1.2.20. St. John’s wort
      • 1.1.2.21. Saffron
      • 1.1.2.22. Wheat protein
      • 1.1.2.23. Antioxidants
      • 1.1.2.24. Teacher training
      • 1.1.2.25. Virtual reality interventions
      • 1.1.2.26. Online interventions
      • 1.1.2.27. Working memory training
      • 1.1.2.28. Digital therapy games
      • 1.1.2.29. School interventions
  • 1.2. Comparison after risk reduction
    • 1.2.1. Reducing the risk of psychiatric hospitalization
    • 1.2.2. Risk reduction of non-psychiatric hospitalization
    • 1.2.3. Risk reduction of suicidal behavior
    • 1.2.4. Risk reduction of incapacity for work
  • 1.3. Comparison according to odds ratio (OR)
    • 1.3.1. Drug interventions
    • 1.3.2. Minerals / vitamins / amino acids / herbal remedies
    • 1.3.3. Non-drug interventions
  • 1.4. Multimodal therapy
• 2. Efficacy latency of the treatment forms
• 3. Effectiveness of the forms of treatment
  • 3.1. Medication
  • 3.2. Psychotherapy
• 4. Duration of effectiveness of the treatment forms
  • 4.1. Early medication
  • 4.2. Short-term medication
  • 4.3. Long-term medication
  • 4.4. Multimodal treatment: non-drug therapy and medication
• 5. Compatibility
• 6. Comparative values: Effect size for other disorders
  • 6.1. Effect size of medication for depression: 0.30
  • 6.2. Effect size of psychotherapy for depression 0.22
  • 6.3. Effect size of medication for ASA
  • 6.4. NNT for other disorders

1. Effectiveness of the treatment methods¶

The different treatment options have varying degrees of effectiveness.
Depending on the question, a lower or higher SMD is “better”.
We have normalized the SMD of the studies in this review so that a higher value always represents greater efficacy in terms of treating ADHD symptoms.

1.1. Comparison by Effect size (SMD)¶

SMD (standardized mean difference, Cohen’s d) is one of the measures in which Effect size can be measured: small = 0.20, medium = 0.50, large = 0.80.
SMD 0.2: NNT = 8.93¹⁸
SMD 0.3: the value of the average person in the experimental group is greater than the value of 62% of the people in the control group¹⁹
SMD 0.5: 69% greater than in control group; NNT = 3.62¹⁸
SMD 0.6: at 73 % larger
SMD 0.75: at 77 % larger
SMD 0.8: NNT = 2.34¹⁸
SMD 1.0: around 70 % probability of a treatment benefit²⁰
SMD 2.0: almost 90 % probability of a treatment benefit²⁰

SMD is calculated as the mean value of the active substance group minus the mean value of the comparison group, whereby the result is divided by the pooled standard deviation of the groups. SMD is therefore a measure of effect size that reflects mean differences between two groups with the same group sizes and the same group variances. The effect is measured in comparison to a comparative measure (usually placebo). Therefore, SMD does not measure the overall improvement, but by how much the improvement was greater than the competitor (usually placebo). Unless we state otherwise, this refers to the Effect size after SMD compared to placebo,

1.1.1. Drug treatment according to Effect size¶

Reviews
A meta-meta-analysis (Umbrella study) found an Effect size for

• pharmacological treatment of ADHD from
  • 0.67 (parent rating)
  • 0.68 (teacher rating) and for
• psychological interventions from
  • 0.42 (parent rating) and
  • 0.25 (teacher rating).²¹

METASTUDIES

• 0.61 Meta-analysis of k = 49 studies with n = 7,685 children and adolescents on all ADHD medications in relation to ADHD symptoms overall²²
• 0.45 Meta-analysis of k = 44 studies with n = 9,952 participants on all medications for ADHD as a whole²³

1.1.1.1. Stimulants: 0.8 to 1.5¶

• Regardless of age
• Effect size of stimulants was reported to be 0.44 stronger than non-stimulants (expert rating).²³
• A meta-analysis of k = 15 RCTs with n = 4,648 children and/or adolescents aged 6 to 17 years with ADHD found an Effect size of;²⁴
  • 0.83 for stimulants
  • 0.58 for non-stimulants
• for children and adolescents
  • 0.88 on ADHD symptoms overall (meta-analysis, k = 12, n = 1,620)²²

1.1.1.1.1. Amphetamine drugs: 1.1 to 1.5¶

• Amphetamine medication regardless of age

  • Amphetamine drugs in total:
    • 1,1²⁵
    • 1,0⁴
    • 0.8 to 1.5²⁶
    • 0.90 in the expert assessment²⁷ as an update of an earlier meta-analysis with 0.72 at that time²⁸
  • According to symptoms:
    • For inattention: 1.5 (lisdexamfetamine medication, but only one examination)²⁵
    • For hyperactivity: approx. 1.2 (total amphetamine medication)²⁵
  • By active ingredient
    • Lisdexamfetamine (Vyvanse) 1.06 to 1.52
      • 1.34, 1.46 and 1.53 (3 studies on children)²²
      • 1,52²⁹
      • 1,28³⁰
      • 1,06²⁷
      • 0.89³¹ In a randomized double-blind study with n = 200 subjects, lisdexamfetamine and atomoxetine were compared in MPH nonresponders. Lisdexamfetamine was significantly more effective than atomoxetine in 2 of 6 categories and in the overall assessment. In terms of learning and school, lisdexamfetamine was 0.19 SMD better than atomoxetine.³²
    • D-Amphetamine immediate release
      • 1.24²⁹ (e.g. Attentin)
    • D-Amphetamine sustained release
      • 1.13²⁹ (not: lisdexamfetamine)
    • Mixed amphetamine salts immediate release
      • 1.34²⁹ 0.64³¹
      • 0,80²⁷
      • In a meta-analysis from the time before lisdexamfetamine was available, Adderall showed the best Effect size, significantly better than methylphenidate.³³
    • Mixed amphetamine salts sustained release (e.g. Adderall XR; mixture of dexamphetamine and levoamphetamine in a ratio of 3:1)
      • 0,77²⁹

• Children and young people:

  • Amphetamine drugs in total:
    • 1.16 (meta-analysis, k = 5, n = 757 children)²² Lisdexamfetamine scored significantly better than mixed amphetamine salts with 1.34, 1.46 and 1.53 (3 studies) compared to 0.76 and 0.77.
    • 1,02³⁴
    • 0.84 in the physician rating; k = 3, n = 813³⁵
    • 0.57 in the parent rating; k = 7, n = 1,247³⁵
    • 0.55 in the teacher rating; k = 5, n = 745³⁵
  • Lisdexamfetamine
    • 1.34, 1.46 and 1.53 (3 studies)²²
  • Mixed amphetamine salts (Adderall):
    • 0,85³⁶
    • 0.9 to 1.2 Adderall XR³⁷
  • Amphetamine suspension with extended release
    • 0.8 / 0.5 to 0.8 for up to 13 hours³⁸

• Adults:

  • 0.79³⁴ which is consistent with other publications³⁹⁴⁰ Amphetamine medications have an NNT of 1.6 in adults with ADHD.²⁶
  • Lisdexamfetamine:
    • 1.07 in European adults (lisdexamfetamine medication; meta-analysis of 22 studies)⁴¹
    • 0,97⁴²
    • 0,5⁴³

• Mixed amphetamine salts:

  • 0.76, 0.77 (2 studies)²²
  • 0.75 independent of age³⁶

• On Emotional Dysregulation approx. 0.41 (stimulants and ATX), compared to 0.8 on ADHD symptoms overall⁴⁴

1.1.1.1.2. Methylphenidate: 0.9 to 1.1¶

• Methylphenidate independent of age

  • MPH as a whole:
    • 0,9⁴⁵
    • Approx. 0.8⁴, 0.73²⁵
    • 0,75³⁰
    • 0,5³¹
    • According to less reliable source 1.3 to 1.69⁴⁶
  • According to symptoms:
    • Inattention
      • Approx. 0.8²⁵
      • 0.42 Sustained attention (metastudy, k = 29, n = 956)⁴⁷
    • Hyperactivity: on average approx. 1.0²⁵
    • Inhibition 0.40 (k = 25, n = 775)⁴⁷ * Methylphenidate with immediate release (unretarded)
    • Working memory 0.24 (k = 13, n = 559)⁴⁷
    • Emotional dysregulation: 0.34⁴⁸
  • Methylphenidate with immediate effect (immediate release)
    • 0,92²⁹
    • Ritalin, MPH Hexal: 1.01⁴⁹
  • Methylphenidate sustained release: 1.08⁴⁹
    • Concerta: 1.35⁴⁹
    • Medikinet retard: 0.95⁴⁹
    • Equasym: 1.09⁴⁹

• MPH in children and adolescents:

  • MPH in general
    • 0.87 in the teacher rating (meta-analysis of k = 5 studies, n = 668)⁵⁰
    • 0,78³⁴
    • 0.77 in the teacher rating (meta-analysis of k = 19 studies, n = 1,698)⁵⁰
    • 0,74⁵¹
    • 0.26 on executive memory compared to placebo (meta-analysis of k = 36 studies⁵²
    • 0.60 on non-executive memory vs. placebo (meta-analysis of k = 36 studies⁵²
    • 0.24 on reaction time compared to placebo (meta-analysis of k = 36 studies⁵²
    • 0.62 on reaction time variability vs. placebo (meta-analysis of k = 36 studies⁵²
    • 0.41 on response inhibition compared to placebo (meta-analysis of k = 36 studies⁵²
  • Concerta:
    • Children and young people: 1.0³⁷
  • Equasym retard:
    • Children and adolescents: 0.6 to 1.8³⁷
  • MPH immediate release; k = 7⁵³
    • Inattention
      • 0.96 in the parent rating
      • 0.98 in the teacher rating
    • Hyperactivity/impulsivity
    • 1.12 in the parent rating
    • 1.25 in the teacher rating
  • Medikinet retard:
    • In total:
      • OROS-MPH 0.9²⁹
      • MPH-MR 0.85²⁹
    • Children and adolescents: 0.9 to 1.0³⁷
    • Ritalin LA:
      • MPH-LA 0.95²⁹
      • Children and young people: 1.0³⁷
      • Adults: 0.49³⁴, 0.34⁴³
    • D-MPH: 0.76²⁹

Adults:

• 0.49 for total MPH (total means: without differentiation between immediate release and sustained release)⁵⁴
• 0.58 for immediate release MPH, found a meta-study of k = 18 RCTs in n = 2,045 adults⁵⁵
  • Dose level
    • 0.58 at an average dose of 57.4 mg/day
    • Each increase of 10 mg/dose resulted in an increase in Effect size of 0.12
  • Dosage regime
    • 0.40 solid doses
    • 0.53 Variable dosage
• OROS-MPH
  • 0,51⁵⁶
• Rertarded MPH compared to placebo⁵⁷
  • ADHD symptoms overall
    • 0.37 Self-assessment; k = 16, n = 3799
    • 0.42 Physician assessment; k = 18, n = 4183
    • 0.31 Peers and family members; k = 3, n = 1005
  • Quality of life
    • 0.15 Self-assessment; k = 6, n = 1888
  • Reduction in days missed from work 0.15; k = 1, n = 409
• Emotional dysregulation:
  • 0.41 (stimulants and ATX), compared to 0.8 on ADHD symptoms overall⁴⁴
  • 0.34 for MPH (without differentiation between immediate release and sustained release)⁵⁴

Meta-analyses comparing MPH with atomoxetine:
In comparison with atomoxetine, MPH proved barely superior in all its medication forms (0.07⁵⁸ to 0.09⁵⁹⁶⁰ .). With regard to inattention, MPH was minimally superior overall (0.13⁵⁸.).
However, OROS-MPH proved to be 0.31⁶⁰ to 0.32⁵⁹ more effective than atomoxetine.
The responder rate for MPH was 14% higher.⁵⁸

Meta-analyses comparing MPH with neurofeedback:
According to a meta-analysis of k = 18 RCTs with 778 participants, methylphenidate proved to be superior to neurofeedback with an Effect size of:⁶¹

• 0.59 on ADHD core symptoms
• 0.96 on inattention
• 0.47 on inhibition
  with a 40 % higher treatment discontinuation rate under MPH.

1.1.1.1.3. Mazindol¶

Mazindol, an alerting substance with stimulant properties, showed a high Effect size of 1.09 on ADHD core symptoms in adults with ADHD ⁴

Further studies should be awaited to confirm this.

1.1.1.2. Total non-stimulants: 0.52 to 0.71¶

Total non-stimulants

• Regardless of age
  • 0,71⁶²
• for children and adolescents
  • 0.52 than according to a meta-analysis of Effect size on ADHD symptoms overall (k = 37, n = 6,065)²²

Effect size of stimulants was reported to be 0.44 stronger than non-stimulants (expert rating).²³

1.1.1.2.1. Guanfacine monotherapy: 0.67 to 0.76 for children and adolescents¶

Guanfacine as monotherapy has an Effect size of 0.57 to 0.8.

• Children and young people:
  • 0,8²⁹
  • 0,76⁶³
  • 0.67 (meta-analysis)³⁴
  • 0.60 (approx.) by guafancin XR on ADHD total symptoms (meta-analysis, k = 6)²²

α-2 agonists (guanfacine and clonidine) in children and adolescents

• 0.59 on ADHD total symptoms (meta-analysis, k = 12, n = 2,276 children and adolescents) with α-2 agonist monotherapy (k = 9, n = 1,550):⁶⁴
  • 0.56 on hyperactivity/impulsivity (k = 9, n = 1 550)
  • 0.57 on inattention (k = 9, n = 1 550)
  • 0.44 on ODD symptoms (k = 9, n = 1 550)
  • fewer discontinuations of monotherapy due to ineffectiveness than with placebo
• 0.52 (approx.) on overall ADHD symptoms (meta-analysis, k = 11, n = 1,885 children and adolescents)²²

1.1.1.2.2. Combination therapy guanfacine with stimulants: plus 0.36 g compared to stimulants alone¶

A combination of MPH with guanfacine works better than MPH alone.

For augmenting additional treatment with α-2 agonists (k = 3, n = 726), there was an additional effect of (meta-analysis):⁶⁴
- 0.36 on overall ADHD symptoms
- 0.33 on hyperactivity/impulsivity
- 0.34 on inattention

A meta-analysis determined the overall Effect size of taking alpha-2 agonists (guanfacine, clonidine) in addition to stimulants in children and adolescents on ADHD symptoms²²

• 0.36 as additional effect size of alpha-2 agonists compared to stimulants alone (k = 5, n = 724)
• 0.64 and 0.34 were determined by the two studies for guanfacine XR (both statistically significant)
• 0.34, 0.30 and 0.16 were determined by the three studies for clonidine (only one of which was statistically significant).

One study found an improved effect of guanfacine plus MPH compared to guanfacine or MPH alone:⁶⁵

• Guanfacine alone: a reduction in symptoms of at least 50% in 68% of people with ADHD
• Methylphenidate alone: symptom reduction of at least 50% in 81% of people with ADHD
• Combination medication of MPH and guanfacine: symptom reduction of at least 50% in 91% of persons with ADHD)

1.1.1.2.3. Clonidine: 0.38¶

• Regardless of age:
  • 0.03 (meta-analysis)²⁹
• Children and young people:
  • 0.71 (meta-analysis)³⁴
  • 0.38 (approx.) on ADHD total symptoms (meta-analysis, k = 5)²²
• In children with ADHD and comorbid tics or Tourette’s syndrome
  • 0.40 in the parent evaluation (study with n = 68 participants)⁶⁶

For studies on alpha-2 agonists in general (guanfacine and clonidine), see Guanfacine monotherapy and Guanfacine combination medication with stimulants.

1.1.1.2.4. Atomoxetine: 0.63 children, 0.45 adults¶

• Atomoxetine independent of age:

  • 0.68 (meta-analysis)³⁰
  • Atomoxetine 0.68 (meta-analysis)³⁶

• Children and young people:

  • 0.7 (meta-analysis)³⁷
  • 0.63 (meta-analysis)²⁹
  • 0.56 (meta-analysis)³⁴
  • In a meta-analysis, ATX showed a 0.23 worse Effect size than MPH (k = 7; n = 1,611)²²

• Adults:

  • 0,46⁵⁶
  • 0,45³⁴
  • 0,24⁴³

• Emotional dysregulation

  • 0.41 (stimulants and ATX), compared to 0.8 on ADHD symptoms overall⁴⁴

• Atomoxetine has an NNT of 5 with regard to attention problems in ADHD²⁶

Noradrenaline reuptake inhibitors (atomoxetine and viloxazine) showed an Effect size of 0.55 on ADHD symptoms in general (meta-analysis, k = 28, n = 1,925).²²

1.1.1.2.5. Bupropion: 0.32 to 0.46¶

METASTUDY:

• Bupropion independent of age:
  • 0,32³⁰
  • 0,22²⁹
• Children and young people:
  • 0,96³⁴
• Adults:
  • 0,46³⁴

1.1.1.2.6. Modafinil: 0.62 to 0.76 KiJu¶

• Modafinil independent of age:
  • 0,65³⁶
  • 0.08 (individual study){Stuhec, Lukić, Locatelli (2018): Efficacy, Acceptability, and Tolerability of Lisdexamfetamine, Mixed Amphetamine Salts, Methylphenidate, and Modafinil in the Treatment of Attention-Deficit Hyperactivity Disorder in Adults: A Systematic Review and Meta-analysis. Ann Pharmacother. 2018 Aug 17:1060028018795703. doi: 10.1177/1060028018795703. n = 701}}
• Children and young people:
  • 0.71 to 0.77 compared to placebo (meta-analysis, k = 5 RCT)⁶⁷
  • 0.76 (meta-analysis, k = 4; n = 667)²² with 3 of the 4 studies reporting an Effect size of 0.42 to 0.54.
  • 0,62³⁴
  • 0.52 (meta-analysis)²⁹
• Adults:
  • Minus 0.16 (deterioration; meta-analysis)³⁴

1.1.1.2.7. Viloxazine¶

The Effect size of viloxazine was determined to be between 0.46 and 0.63 in various studies.⁴

1.1.1.2.8. Centanafadine¶

Manufacturer’s specification 0.6.⁴³
Independent studies must be awaited here.

1.1.1.2.9. Dasotraline¶

A 6-week RCT with 342 children aged 6-12 years found an Effect size of 0.48 at 4 mg/day. 2 mg/day was almost ineffective^{68 43}

A meta-analysis of k = 8 studies found the Effect size of dasotraline in children and adults with ADHD to be⁶⁹

• 0.35 on overall ADHD symptoms
• 0.27 on hyperactivity/impulsivity
• 0.33 on inattention

1.1.1.2.10. Cannabinoids, Sativex¶

There is only one small RCT on THC with n = 30 subjects. This found an Effect size of 0.2 compared to placebo for cognitive performance.⁷⁰⁷¹

1.1.1.2.11. SGA (second generation of antipsychotics) for hyperactivity in ASD¶

A meta-analysis examined the Effect size of the second generation of antipsychotics on the ADHD symptom of hyperactivity in children with ASD:⁷²
0.66 Stimulants
0.59 SGA

1.1.1.2.12. Desipramine¶

Desipramine improved the core symptoms of ADHD in children and adolescents⁶⁶

• 1.42 in the parent rating (k = 2, n = 99)
• 0.97 in the teacher rating (k = 2, n = 89
• in the physician rating (k = 2, n = 103 , no information in SMD)

• 0.90 in parent rating in children with ADHD and comorbid tics or Tourette syndrome (study with n = 68 participants)

Desipramine is no longer on the market in most countries due to its broad-spectrum effect and the resulting high level of side effects.

1.1.1.2.12. Tipepidine¶

0.38 in children and adolescents according to a single study (n = 51)²²

1.1.1.2.13. Nortriptyline¶

Improvement in core symptoms of ADHD in children and adolescents in physician ratings. Not specified in SMD⁶⁶

1.1.1.2.14. Selegelin¶

In two studies, Selegelin was no better than placebo in terms of improving ADHD symptoms.⁷³

1.1.2. Effectiveness of non-drug therapies for ADHD according to Effect size¶

Effect indicated in Effect size (SMD). Higher values are better.

A meta-meta-analysis found an Effect size for pharmacological treatment of ADHD of 0.67 (parent rating) and 0.68 (teacher rating) and for psychological interventions of 0.42 (parent rating) and 0.25 (teacher rating).⁷⁴

A meta-analysis of k = 32 published RCTs of behavioral interventions (all interventions aimed at increasing desirable and decreasing undesirable behaviors, i.e. classical contingency management, behavioral therapy (mainly by mediators such as parents or teachers) and cognitive behavioral therapy (such as verbal self-instruction, problem-solving strategies or social skills training) in children aged 3 to 18 years, found¹⁵

• unblinded
  • Improvements in the quality of education:
    • 0.68 Positive parenting behavior
    • 0.57 Negative parenting behavior
  • 0.37 Parental self-concept
  • 0.35 ADHD of the child
  • 0.26 Behavioral problems
  • 0.47 Social skills
  • 0.28 School performance
• blinded, only some results remained statistically significant:
  • 0.63 Positive parenting behavior
  • 0.43 Negative parenting behavior
  • 0.31 Behavioral problems

1.1.2.1. Sports (endurance training): approx. 0.8 (0.62 to 1.96)¶

In addition to Effect size, adherence to therapy should also be taken into account. People with ADHD should therefore do the sport that they enjoy the most.⁷⁵

Studies

• An aerobics group exercise program brought about improvements in participants with various disorders compared to the passive control group in terms of⁷⁶
  • Global ADHD symptom severity: 0.77
  • Depression; 0.68
  • Anxiety: 0.87
  • Sleep quality: 0.88

METASTUDIES

• 0.93 after a meta-analysis of 5 studies with a total of 144 test subjects.⁷⁷ Unfortunately, there was no comparison with the Effect size of medication.
• 0.65 for endurance training in children with ADHD⁷⁸
  • Attention: 0.84
  • Anxiety symptoms: 0.66
  • Executive functions: 0.58
  • Social disorders: 0.59
  • Hyperactivity: 0.56
  • Impulsiveness: 0.56
• Meta-analysis of k = 11 studies with n = 713 children:⁷⁹
  • Inhibition: 0.78
  • Twice 60 minutes or longer of open training per week showed the best effect Inhibition in children with ADHD
• 0.62 Cochrane meta-analysis of k = 35 studies with n = 1,356 subjects on the effect of sports training on
  • Depression
    • Long-term effect: 0.33
    • 0.18 when limited to the studies with high blinding. The comparative studies on the Effect size of psychotherapy (k = 7 studies, n = 189) or medication (k = 4 studies, n = 300) each found an identical SMD of endurance training.⁸⁰
• Meta-analysis of k = 23 studies with n = 535:⁸¹
  • Inattention: 0.604 to 0.715
  • Hyperactivity/impulsivity: 0.676
  • Emotional problems: 0.416
  • Behavioral problems: 0.347
• Meta-analysis of k = 15 RCT with n = 734 subjects:⁸²
  • Inattention: 0.60
  • Executive functions: 1.22
  • Motor skills: 0.67
  • Hyperactivity: 0.06
  • Depression: 0.72 (not statistically significant)
  • Social problems: 0.27 (not statistically significant)
  • Aggressive behavior: 0.24 (not statistically significant).
• Meta-analysis of k = 15 studies with n = 664 children and adolescents with ADHD aged between 6 and 18 years:⁸³
  • Executive functions: 0.61
  • Inhibition: 0.76
  • Cognitive flexibility 0.78.
• Meta-analysis of k = 24 with n = 914 children and adolescents with ADHD:⁸⁴
  • Inhibition: 0.50
  • Working memory: 0.50
  • Cognitive flexibility: 0.45
• Meta-analysis of k = 10 RCT with n = 474 children aged 6 to 12:⁸⁵
  • Attention: 0.48
  • Effect size was moderated by the type, frequency and duration of physical activity, not by the physical activity setting or the timing of the individual intervention.
• 0.29 (not statistically significant) according to a meta-analysis of k = 31 RCTs (5 of which related to ADHD) with n = 1,255 children and adolescents. In contrast, the meta-analysis found a significant effect for ASD (0.50), depression (0.68) and obesity (0.58)⁸⁶ This result is surprising, as the included studies consistently describe significant effects.

Reviews
A review analyzed 37 meta-analyses of 106 studies and found Hedges g:⁸⁷

• Inattention: 0.92
• Impulse control: 0.82
• Cognitive flexibility: 0.52

The evidence for the effectiveness of exercise on emotional, social and working memory performance was weak, and on hyperactivity and behavioral function was not significant.

A meta-analysis of k = 44 studies from 1983 to 2022 with n = 1,757 children and adolescents found an improvement compared to controlled alternatives (e.g. waiting list, no intervention, watching a video or sedentary attention control)⁷⁵

• Executive functions
  • of 1.15 through all types of physical activity
  • of 1.96 through open-ended skills activities that require participants to respond in a dynamically changing and externally controlled environment
• Inhibition
  • of 1.94 through activities with open skills
• Working memory
  • of 1.21 through closed skill activities dominated by aerobic exercise
• cognitive flexibility
  • of 1.44 through physical multi-component exercises
• Hyperactivity / impulsivity
  • of 1.60 through closed skill activities dominated by aerobic exercise
• Inattention
  • of 1.51 by closed skill activities dominated by aerobic exercise

A meta-analysis of k = 22 RCTs on children and adolescents with ADHD showed an Effect size for:⁸⁸

• permanent sports
  • 0.39 ADHD core symptoms
  • 0.32 Inattention
  • 0.68 Executive functions
• closed exercises
  • 0.83 ADHD core symptoms

1.1.2.2. Mindfulness-based behavioral therapy (MBCT)¶

People with ADHD who were additionally treated with MBCT (mindfulness-based cognitive therapy) showed a significantly greater reduction in ADHD symptoms [M difference = -3.44 (-5.75, -1.11), p = 0.004, d = 0.41]. This effect was maintained up to the 6-month follow-up. Of the people with ADHD who were additionally treated with MBCT, 27% showed a 30% reduction in ADHD-HI symptoms (p = 0.001), compared to only 4% of the people with ADHD who were not additionally treated with MBCT.⁸⁹
The effect largely persisted even after the end of therapy.

1.1.2.3. Cognitive behavioral therapy: 0.5 (0.3 to 0.85)¶

Cognitive behavioral therapy (CBT)

• Effect largely persists after the end of therapy
• Very long time to take effect

Studies:

• Effect on externally observed ADHD symptoms in adults

  • CBT compared to supportive therapy
    • 0,56⁹⁰
  • MBCT compared to waiting list
    • 0,85⁹¹
  • CBT compared to waiting list
    • 1,68⁹²
  • CBT compared to continued medication alone
    • 0,60⁹³⁹⁴

• Effect on self-observed ADHD symptoms in adults

  • CBT compared to supportive therapy: 0.16⁹⁵ with reference to 2 studies. Only one of these dealt with CBT⁹⁰, the second study examined DBT/skill training and also reported a good result⁹⁶.
  • CBT compared to waiting list: 0.84⁹⁵ with reference to 5 studies. However, only studies on MBCT^{9791 98} , internet-based CBT⁹⁹ and brief therapy¹⁰⁰ were included
  • 1,0⁹⁴

• CBT without concomitant medication:
  • 0,34¹⁰¹
  • 0,3²

METASTUDY:

• 0.70 according to a meta-analysis of k = 3 studies with a total of n = 107 subjects¹⁰²
• 0.79 compared to waiting lists in k = 4 studies with n = 160 subjects¹⁰³
• 0.43 compared to active control groups (e.g. psychoeducation, progressive muscle relaxation, encouragement) in k = 3 studies with n = 191 subjects¹⁰⁴

A meta-analysis found an effect of behavioral therapy on inattention in adults with ADHD of¹⁰⁵

• against waiting list or therapy as usual
  • in the external observation rating
    • 1.18 (0.44, 0.92, 0.95, 2.41; k = 4 studies with n = 124 adults)
  • in the self-assessment rating
    • 0.93 (0.18, 0.28, 0.68, 0.82, 0.92, 1.18, 1.22, 1.35, 1.71; k = 9 studies with n = 215 adults)
• against active controls (psychoeducation, neurofeedback, clinical management, cognitive training)
  • in the external observation rating
    • below 0.0 = slightly negative SMD = slight worsening (k = 4 studies with n = 229 adults, compared to psychoeducation, MPH or placebo)
    • 0.5 in k = 2 studies with a total of n = 49 adults, of which only one was statistically significant
  • in the self-assessment rating
    • below 0.0 = slightly negative SMD = slight deterioration (k = 8 studies with n = 298 adults)
    • 0.4 in k = 3 studies with n = 91 adults, of which only 1 was statistically significant

A large meta-analysis of k = 190 studies with n = 26,114 participants with ADHD-HI found that stimulants were more effective than behavioral therapy, cognitive training, or non-stimulants. Stimulants in combination with behavioral therapy appeared to be most effective.¹⁰⁶ See below under Effectiveness by OR.

A meta-analysis of k = 17 RCTs with n = 1,075 participants found an effect of cognitive interventions of:¹⁰⁷

• 0.39 on attention symptoms
  • Working memory training: no significant effect on attention symptoms
  • multiple cognitive training: 0.51 on attention symptoms
• 0.32 on executive functional behavior
  • Working memory training: no significant effect on executive functioning behavior
  • multiple cognitive training: 0.5 on executive functional behavior
• multiple cognitive training: 0.31 on hyperactivity/impulsivity

Overall, only treatment frequency had a significant effect on ADHD symptoms and executive behaviors.¹⁰⁷

1.1.2.4. Neurofeedback: 0.29 to 1.2¶

Studies without limitation to blinded observation:

• 0.49 to 0.68 after only 30 sessions⁴⁶
• 0,6¹⁰⁸¹⁰⁹

• 0,7¹¹⁰
• 1,2¹¹¹
• 1,2¹¹²
• 1,2¹¹³

METASTUDIES

• 0.49 (parent rating); k = 5, n = 246¹¹⁴
• 0.35 ADHD total symptoms meta-analysis with k = 13 studies on n = 520 subjects, unblinded:¹¹⁵ Further results in relation to individual symptoms:
  • 0.36 Inattention
  • 0.26 Hyperactivity/impulsivity

Studies with limitation to blinded observation according to Sonuga-Barke et al (2013):¹¹⁶

• 0,4¹⁰⁸
• 0,4¹¹⁰
• 0,3¹¹³
• 0,1¹¹⁷

METASTUDY:

• (also) unblinded studies:
  • 0,59¹¹⁶
  • 0.61 after a meta-analysis of k = 6 studies with a total of n = 203 subjects¹¹⁸ Unfortunately, there was no comparison of the Effect size of medication.
• Meta-analyses with only blinded studies
  • 0.30 (teacher rating, k = 5, n = 246)¹¹⁴
  • 0,29¹¹⁶
  • 0.30 Impulse control (not statistically significant; k = 13, n = 520)¹¹⁹
  • 0.13 Attention (not statistically significant; k = 13, n = 520)¹¹⁹

Children and young people:

• 0.44 on ADHD symptoms (statistically significant, meta-analysis, n = 12, n = 945)²²

According to symptoms:

• On attention: 0.8 to 1.2
  • 0.64 directly after treatment compared to untreated controls (MPH in contrast 1.08)¹²⁰
  • 0.80 at follow-up after 2 to 12 months (MPH in contrast 1.06)¹²⁰
  • 0,8¹²¹
  • 0.48 for inattention in EEG-NF compared to waiting list/TAU (k = 5; n = 279)¹²²
  • 0.46 (parent rating); k = 5, n = 246¹¹⁴
  • After 40 sessions: 1.2¹²³

• on hyperactivity/impulsivity: 0.5 to 0.61
  • 0.50 directly after treatment compared to untreated controls¹²⁰
  • 0.61 at follow-up after 2 to 12 months¹²⁰
  • 0.39 after 20 sessions; cannot be further improved by increasing the number of sessions¹²³
  • 0.03 for hyperactivity/impulsivity in EEG-NF compared to waiting list/TAU (k = 5; n = 279)¹²²
  • 0.34 (parent rating); k = 5, n = 246¹¹⁴

• On impulsivity: 0.68
  • After 20 sessions: 0.7¹²³
  • Cannot be further improved by increasing the number of sessions

In comparison:

• Neurofeedback is not as effective as MPH¹²⁴
  • In one study, MPH showed a symptom improvement of 46.9 % (SMD 2.03), while neurofeedback led to a symptom improvement of 26.7 % (SMD 0.89).¹²⁵
• Effect usually remains complete after the end of therapy

1.1.2.5. Parent training: 0.31 to 0.68¶

The results relate to the improvement of the children’s problematic behavior through parent training.
Parent training courses include, for example²²

• Educational and behavioral strategy programs for parents of preschool children
• Psychoeducation for families
• New Forest Parenting Package for parents of preschool children

METASTUDY:

• 0.86 for parent training in the group¹²⁶
• 0.68 in relation to ADHD (k = 11, n = 603 children from 33 to 144 months)¹²⁷
• 0.61 in relation to the ADHD of preschool children¹²⁸
• 0.42 to 0.53¹²⁹
• 0.40 for ADHD¹³⁰
• 0.31 in relation to children’s ADHD symptoms (k = 11 RCT, n = 1,078)²²
• 0.31, k = 4 studies¹³¹ although at the same time an Effect size of 0.85 was attributed to antidepressants, 0.35 to stimulants and 0.28 to multimodal treatment, which deviates so far from other findings that it raises questions about the robustness of the data.

Improvement of certain symptoms:

• externalizing behaviour of the child (e.g. breaking rules, oppositional behaviour or aggression):
  • 0.36 for externalizing symptoms¹³⁰
  • 0.32; not statistically significant (k = 3, n = 190)¹³²
• internalizing behaviour of the child (e.g. withdrawal or anxiety):
  • 0.48, statistically significant (k = 2, n = 142)¹³²
• 0.59 on comorbid behavioral problems¹²⁷
• 0.93 on parental self-esteem¹²⁷

1.1.2.6. Repetitive transcranial magnetic stimulation (rTMS) 0.54¶

Improvement in cognitive function: 0.54 compared to controls¹³³

A meta-analysis of k = 6 RCTs with n = 169 participants showed the Effect size for rTMS compared to sham treatment:¹³⁴ (although there was no statistical significance for superiority of treatment of the right PFC)

• on overall ADHD symptoms
  • 0.49 Treatment of right PFC
  • 0.24 Total group (treatment of left or right PFC)
  • 0.01 Treatment of left PFC
• on individual symptoms
  • 0.76 Inattention
  • 0.43 Impulsiveness
  • 0.0 Hyperactivity
  • 0.0 Depression

1.1.2.7. Cognitive self-regulation training 0.54¶

• Out-of-school 0.58, n = 5 studies¹³⁵ although an Effect size of 0.85 was attributed to antidepressants, 0.35 to stimulants and 0.28 to multimodal treatment. These figures are in no way consistent with the clinical evidence, which raises questions regarding the reliability of the data.
• School-based 0.49, n = 2 studies¹³⁵ although an Effect size of 0.85 was attributed to antidepressants, an Effect size of 0.35 to stimulants and an Effect size of 0.28 to multimodal treatment. These figures are in no way consistent with the clinical evidence, which raises questions regarding the reliability of the data.

1.1.2.8. Behavioral training 0.54¶

• Out-of-school 0.29, n = 1 study¹³⁵ although at the same time an Effect size of 0.85 was attributed to antidepressants, 0.35 to stimulants and 0.28 to multimodal treatment, which raises questions about the robustness of the data.
• School-based 0.50, n = 3 studies¹³⁵ although at the same time an Effect size of 0.85 was attributed to antidepressants, 0.35 to stimulants and 0.28 to multimodal treatment, which raises questions regarding the robustness of the data.

1.1.2.9. Food diets: 0.51¶

Effect only during faultless dietary compliance.
High values can be achieved under test conditions if food intolerance is present.
An elimination diet places very high demands on compliance in the elimination phase and, as a form of treatment, is a very serious intervention in lifestyle with a high potential for errors in daily life. It is barely feasible for children and has considerable social consequences.

Studies:

• Oligoantigenic diet:
  • 0,2¹³⁶
  • 0,2¹³⁷
• Avoidance of known antigens
  • 0,29¹³⁸
  • 0,6¹³⁹
  • 0,7¹⁴⁰
  • 1,9¹⁴¹
• Elimination diet
  • 3,58¹⁴²
  • 5,13¹⁴³
• Avoidance of food supplements / colorants: 0.2 (0.08 to 0.44)
  • 0.08 to 0.11 (teacher and observer rating)¹⁴⁴
  • 0.12 to 0.25¹³⁸
  • 0.21 to 0.283 (with the addition of smaller, less high-quality studies)¹⁴⁵
  • 0.21 to 0.44 in the parent rating¹⁴⁴
  • 0.42 when only blinded studies were used; of 8 studies, 6 showed no statistically significant results; 0.32 when all studies were included¹¹⁶

Meta-analyses of dietary interventions:

• 0.51 to 0.8¹⁴⁴
• 0.51 (0.2 to 1.9) if only the blinded studies were used; 1.48 (0.2 to 5.13) if all studies were included¹¹⁶

There is evidence that an elimination diet only works for certain people with ADHD (“subgroup”).¹⁴⁶
In our opinion, this is in the nature of things, as the elimination of foods can only bring about an improvement if there is a food intolerance. However, these are highly individual.

1.1.2.10. Cognitive training 0.22 to 0.45¶

• 0.45 after a meta-analysis of k = 4 studies with n = 159 subjects.¹⁴⁷

A meta-analysis examined the Effect size of cognitive training¹⁴⁸

• Treatment reports (unblinded):
  • 0.37 on ADHD in total
  • 0.47 on symptoms of inattention
  • 0.79 on ADHD overall through interventions targeting multiple neuropsychological deficits
• (presumably) blinded reports:
  • 0.20 on ADHD as a whole
  • 0.32 on symptoms of inattention

Children and young people:

• 0.37 on ADHD symptoms in children and adolescents (meta-analysis, k = 12, n = 655)²²
• 0.22 (meta-analysis)¹⁴⁹

One review found limited to minor symptom improvement with computerized cognitive training (CCT)¹⁵⁰
* 0.12 on ADHD total symptoms when restricted to the “probably blinded” results (PBLIND; k = 14 studies)
* 0.12 on hyperactivity/impulsivity symptoms when restricted to the “probably blinded” results (PBLIND; k = 14 studies)
* 0.17 on the symptoms of inattention when restricted to the “probably blinded” results (PBLIND; k = 14 studies)
* 0.40 on the symptoms of inattention during assessment immediately after training (indication of rater bias)
* 0.38 on working memory verbal
* 0.49 on working memory visual-spatial
* No improvement in attention, inhibition, reading or arithmetic*

Overall, there were improvements:¹⁵¹

• 0.52 verbal working memory
• 0.3 Executive functions in the parent rating
• There were no significant effects with regard to
  • Hyperactivity/impulsivity
  • School performance

1.1.2.11. Total psychological behavioral interventions: 0.35 for children and adolescents¶

A meta-analysis found an Effect size on ADHD symptoms of 0.35 (k = 14, n = 1,686) for the entirety of psychosocial behavioral interventions aimed at children and adolescents.²² When limited to RCTs, the value remained at 0.36. 8 of the 6 studies found no statistically significant effect.
The psychological behavioral interventions included comprised

• Skills training (e.g. executive function training, homework training or organizational skills training)
• Social skills training
• Executive function therapy for preschool children
• Driving program for young drivers
• sleep-focused intervention
• dialectical behavioral therapy
• cognitive behavioral therapy
• Attention training
• complex behavior modification intervention
• Behavioral counseling with school and home components
• Parent-child psychotherapy for mothers and their children with ADHD
• Mindfulness training
• Music therapy
• playful intervention
• dog-assisted therapy

With an Effect size of 0.007 (k = 3, n = 459), there was no improvement in school performance.²²

1.1.2.12. Social skills training 0.26 to 0.31¶

A Cochrane meta-analysis found a therapeutically unhelpful Effect size of social skills training for children and adolescents aged 5 to 17 in the teacher evaluation:¹⁵²

• ADHD total symptoms: 0.26 (k = 14 RCT with n = 1379)

• 0.11 Social competence (k = 11 RTC, n = 1,271)
• 0.02 Emotional competence (k = 2, n = 129)
• 0.06 General behavior (k = 8, n = 1,002)

Social skills training was included in this meta-analysis:

• Social skills training
• cognitive behavioral therapy
• multimodal behavioral/psychosocial therapy
• Treatment of children’s life and attention skills
• Life skills training
• the “Program for challenging horizons”
• verbal self-instruction
• metacognitive training (a variant of cognitive behavioral therapy)
• Behavioral therapy
• Treatment of behavior and social skills
• psychosocial treatment.

Skills training also includes leadership training, homework or organizational skills training.²²

• Meta-analysis of k = 3 studies¹³⁵
  • 0.31; however, antidepressants were attributed an Effect size of 0.85, stimulants an Effect size of 0.35 and multimodal treatment an Effect size of 0.28, which raises questions about the robustness of the data and suggests inflated Effect size results

1.1.2.13. Probiotics: 0.25¶

Probiotics improved ADHD compared to placebo:¹⁵³
0.25 Total ADHD symptoms
0.14 Inattention
0.08 Hyperactivity/impulsivity

1.1.2.14. Transcranial direct current stimulation (tDCS) 0.23¶

A review study of k = 102 studies found an Effect size of 0.23 for tDCS in ADHD.¹⁵⁴

1.1.2.15. Omega-3/-6 / PUFA supplementation: 0.11 to 0.17¶

There is no evidence that PUFAs (omega-3 and omega-6 fatty acids) are effective for ADHD.
Even in parent ratings, which are subject to a bias due to knowledge of the administration, the effects were below the minimum value of 0.2 that is meaningful for treatment¹⁵⁵

METASTUDY:

• 0.16 to 0.17 in the parent and teacher rating ¹⁴⁴
• Minus 0.08 in the parent rating meta-analysis of k = 16 studies with n = 1,116 participants. The effect on hyperactivity/impulsivity (0.08) was slightly better than that on inattention (minus 0.01).¹⁵⁶
• 0.16 on ADHD core symptoms Meta-analysis with k = 22 studies and n = 1,789 participants by omega-3 fatty acids¹⁵⁷
• 0.35 when limited to studies with a duration of at least 4 months compared to placebo.¹⁵⁷ Neither a high dosage of EPA nor a high EPA/DHA ratio improved ADHD symptoms.
• 0.06 to 0.17 on ADHD symptoms, meta-analysis of k = 31 studies with n = 1,755 patients:¹⁵⁸
  • ADHD core symptoms
    • 0.17 in the parent rating (k = 23)
    • 0.06 in the teacher rating (k = 10)
  • Behavioral difficulties
    • 0.02 in the parent rating (k = 7)
    • 0.04 in the teacher rating (k = 5)
  • Quality of life: no relevant effect (SMD = 0.01)

• 0.11 on ADHD symptoms. Meta-analysis of k = 7 studies with n = 719 children and adolescents²²

1.1.2.16. Zinc¶

A meta-analysis of k = 6 RCTs with n = 489 children found an Effect size on:¹⁵⁹

• 0.62 ADHD core symptoms
• 0.93 Hyperactivity (not statistically significant)
• 0.21 Inattention (not statistically significant)

An individual study (Bilici, 2004) reports an Effect size of 1.61 for children and adolescents. Unfortunately, such values are in no way consistent with empirical experience.²²

1.1.2.17. Acetyl-L-carnitine¶

A meta-analysis found two studies on this, one of which found a deterioration of 0.06 and the other an improvement of 0.21 in children and adolescents.
Overall, there is no detectable effect on ADHD symptoms.²²

1.1.2.18. Phosphate dilyserine¶

There is only one study on the effect of phosphate dilyserin on ADHD, which found an Effect size of 0.85 on ADHD symptoms in children and adolescents.²²

1.1.2.19. Iron¶

One study found an Effect size of 0.15 on ADHD symptoms in children and adolescents.²²

1.1.2.20. St. John’s wort¶

In this regard, a single study found a 0.22 worsening of ADHD symptoms in children and adolescents.²²

1.1.2.21. Saffron¶

A single study found an Effect size of 0.97 on ADHD symptoms in children and adolescents.²²
Another study (Baziar 2019) reported that saffron has a similar effect to methylphenidate.

1.1.2.22. Wheat protein¶

A single study found an Effect size of 1.07 on ADHD symptoms in children and adolescents.²²

1.1.2.23. Antioxidants¶

A meta-analysis found two studies on this, one of which found a deterioration of 0.14 and the other an improvement of 0.38 in children and adolescents.
Overall, there is no detectable effect on ADHD symptoms.²²

1.1.2.24. Teacher training¶

A meta-analysis of k = 22 studies found for teacher training:¹⁶⁰

• 0.71 to 0.78 Behavioral improvement of the children with ADHD as a consequence of the adapted teacher behavior

In addition, there was a strong improvement in the trained teachers’ knowledge about ADHD immediately after the training (1.96), which, however, decreased again by 1.21 within 3 months.

1.1.2.25. Virtual reality interventions¶

A meta-analysis of k = 4 studies with n = 125 people with ADHD found an Effect size for virtual reality-based interventions of¹⁶¹

• 1.38 on omission error (omission error)
• 1.50 for correct answers
• 1.07 on perceptual sensitivity
• 0.62 on commission error
• 0.67 on reaction time
• No effect on impulsivity

1.1.2.26. Online interventions¶

A meta-analysis of k = 6 RCTs with n = 261 subjects found an Effect size of online interventions in improving attention deficit and social functioning of adults and children with ADHD compared to the waiting list of 0.59.¹⁶²

1.1.2.27. Working memory training¶

A meta-analysis of k = 12 studies found working memory training to be effective in children and adults with and without ADHD:¹⁶³¹⁶⁴

• 0.37 on inattention in everyday life
• 0.62 on visual-spatial working memory tasks
• 0.41 on verbal working memory tasks

1.1.2.28. Digital therapy games¶

A meta-analysis of k = 20 studies found an Effect size of digital therapy games for ADHD:¹⁶⁵

• Inattention
  • 0.21 in the teacher rating
  • 0.28 in the parent rating
• Hyperactivity/impulsivity
  • minus 0.01 in the teacher rating
  • 0.16 in the parent rating

Most of the study results were not statistically significant.

1.1.2.29. School interventions¶

• 0.50 on ADHD symptoms; not statistically significant; meta-analysis of k = 5 studies with n = 822 children and adolescents²²
• 0.36 (not statistically significant) Meta-analysis of k = 39 studies on school interventions in low- and middle-income countries¹⁶⁶

1.2. Comparison after risk reduction¶

In a comparison after risk reduction, a lower hazard risk represents an improvement.

1.2.1. Reducing the risk of psychiatric hospitalization¶

Influence on the risk of psychiatric hospitalization¹⁶⁷

• Amphetamine (adjusted hazard ratio, aHR: 0.74 = reduction of 26%)
• Lisdexamfetamine (aHR: 0.80 = reduction by 20 %)
• ADHD medication polytherapy (aHR: 0.85 = reduction by 15 %)
• Dexamphetamine (aHR:0.88 = reduction by 12 %)
• Methylphenidate (aHR: 0.93 = reduction by 7 %)
• Modafinil: unchanged
• Atomoxetine: unchanged
• Clonidine: unchanged
• Guanfacine: unchanged

1.2.2. Risk reduction of non-psychiatric hospitalization¶

Influence on the risk of non-psychiatric hospitalization¹⁶⁷
Amphetamine, lisdexamfetamine, polytherapy (combination medication), dexamphetamine, methylphenidate and atomoxetine reduced the risk of non-psychiatric hospitalization.

1.2.3. Risk reduction of suicidal behavior¶

Influence on the risk of suicidal behavior¹⁶⁷

• Dexamphetamine (aHR: 0.69)
• Lisdexamfetamine (aHR: 0.76)
• Methylphenidate (aHR: 0.92)

1.2.4. Risk reduction of incapacity for work¶

Influence on the risk of incapacity for work¹⁶⁷

• Atomoxetine (aHR: 0.89)
  • particularly among adolescents and young adults aged 16 to 29 (aHR: 0.82)
    All other ADHD medications studied: not significant.

1.3. Comparison according to odds ratio (OR)¶

Overview according to Catalá-López et al (2017).¹⁰ The data compare the probability of treatment success compared to placebo. OR = 1 would be equivalent to placebo, an OR higher than 1 favors the treatment method.

Placebo: 1 (comparative value)

1.3.1. Drug interventions¶

Comparison according to odds ratio (OR)

• Total stimulants: 6.21

  • Amphetamine drugs: 7.45
  • Methylphenidate: 5.26

• Total non-stimulants: 3.95

  • Modafinil: 5.51
  • Atomoxetine: 3.63
  • Guanfacine: 3.29
  • Clonidine: 3.96
  • Bupropion: 2.41
  • Venlafaxine: 4.07
  • Reboxetine: 3.58
  • Antipsychotics: 1.36
  • Thioridazine: 1.04
  • Carbamazepine: 0.18

1.3.2. Minerals / vitamins / amino acids / herbal remedies¶

Comparison according to odds ratio (OR)

• Zinc: 2.42
• Polyunsaturated fatty acids (or PUFAs): 2.14
• Omega-3 and -3/6 fatty acids: 1.99
• L-carnitine: 1.20
• Amino acids: 1.19
• St. John’s wort (Hypericum perforatum): 1.00 (0.23-4.26)
• Ginkgo biloba: 0.21

1.3.3. Non-drug interventions¶

Comparison according to odds ratio (OR)

• Behavioral therapy: 2.97
• Neurofeedback: 1.96
• Parent training: 1.19
• Child, parent and/or teacher training: 2.73
• Cognitive training: 0.70
• Working memory training: 0.34

1.4. Multimodal therapy¶

Individual studies
In the “Multimodal Treatment Study of Children with ADHD” (MTA study) in the late 1990s, 579 children aged 7 to 10 were treated with medication, behavioral interventions or both for 14 months.¹⁶⁸ The MTA study only included group interventions (parent group training, child summer camp and school support) and no individual behavioral therapy.
Teachers and parents rated the reduction in symptoms with regard to the core symptoms of ADHD better in the children treated only with medication than in the children treated only with parent and behavioral training. The children who received medication and parent and behavioral training performed even slightly better than the children treated with medication alone, albeit with reduced medication requirements. When not only the core symptoms but all symptoms were considered, the children treated with medication and parent and behavioral training clearly performed best. In contrast, the effect of parent and behavioral training alone was lower than treatment with medication alone.⁸

METASTUDIES

• A comprehensive meta-analysis on the multimodal treatment of children and adolescents with ADHD found an Effect size for additional psychological treatment of
  • 0.36 on ADHD symptoms compared to medication alone (k = 7, n = 841, statistically marginally non-significant, low risk of bias)²²
    • The greatest Effect size was shown by a behavioral and social skills course for children and their parents and cognitive behavioral therapy for adolescents
  • 0.42 on broadband symptoms compared to medication alone (k = 3, n = 171, not statistically significant, high risk of bias)²²
    • Multimodal psychosocial treatment plus MPH versus MPH alone, group cognitive behavioral therapy plus MPH versus MPH, and individual cognitive behavioral therapy plus FDA-approved medications versus these medications alone were assessed.
• Another meta-analysis found that
  • 0.1 for parent and behavioral training alone versus no treatment
  • minus 0.4 if only the blinded raters were used, i.e. a deterioration compared to no parent and behavioral training.¹¹⁶

2. Efficacy latency of the treatment forms¶

By efficacy latency, we mean how long it takes for a treatment to have an effect.

• Medication
  • Stimulants: immediate effect, fully effective immediately when optimally adjusted
  • Noradrenaline reuptake inhibitors: 2-3 weeks flooding phase
  • Atomoxetine: several weeks to 6 months flooding phase
• Therapy
  • Behavioral therapy: several months for first steps, 3 years for adequate treatment effect
  • Neurofeedback: several months for first steps, 6 to 15 months for adequate treatment effect

3. Effectiveness of the forms of treatment¶

By effectiveness, we mean which symptoms the individual forms of treatment change.

3.1. Medication¶

• Stimulants:
  • Attention
  • Hyperactivity
  • Impulsivity (MPH more than amphetamine drugs)
  • Internal pressure
  • Emotional dysregulation
    • Mood swings / affect stability
    • Aggression/anxiety
    • Dysphoria (predominantly amphetamine medication, MPH less so)
• Non-stimulants:
  • Impulsiveness
    e.g. low doses of SSRIs
  • Emotional regulation
  • Attention
• Noradrenaline reuptake inhibitors:
  • Impulsiveness
  • Depression
  • Mood swings / affect stability
  • Hyperactivity

3.2. Psychotherapy¶

• Behavioral therapy:
  • Cognitive VT:
    Self-esteem, social behavior, stress reduction
    To a certain extent, changes in stress processing, including hormonal and immunological physical changes¹⁶⁹
  • Mindfulness-based VT:
    Attention, empathy, stress reduction
• Mindfulness training:
  Change in stress perception; change in stress processing in the CNS, impulsivity
  An improved perception of pleasant aspects causes immediate changes in the dopaminergic focusing and reinforcement system. Perceptions of the nature of an individual’s spatial environment and (permanently) a high social status have the same effect.¹⁷⁰
  However, it is unclear whether these changes (with the exception of social rank) have a lasting influence, which is a prerequisite for therapeutic use, or whether they are only activated during perception.
• Neurofeedback:
  Attention; impulsivity; hyperactivity, relaxation, sleep
• Environmental interventions:
  Elimination of stressors through elimination of stressors and greater understanding of the environment
• Psychoeducation:
  Elimination of stressors and better ability to regulate through greater understanding of the person with ADHD
  Increased self-esteem through the feeling of coming home, meeting and exchanging ideas with other people with ADHD

4. Duration of effectiveness of the treatment forms¶

4.1. Early medication¶

There is an indication that methylphenidate treatment with 2 mg / kg / day in very young rats caused a permanent reduction in dopamine transporters in the striatum (which would correspond to a permanent healing effect), while methylphenidate administration in somewhat older animals (“after puberty”) no longer had this effect.¹⁷¹ These results have not yet been reproduced.

No permanent reduction in DAT in the striatum is known in the treatment of humans with MPH, even in children before puberty.

Despite the immense importance of this question, no further studies are known to confirm the results. A dosage of twice 5 mg / kg / day in rats from the 7th day to the 35th day after birth resulted in short-term reductions in the DAT count, but these were no longer found on the 135th day of life.¹⁷²
There were no structural changes in the brain structures either immediately after the end of treatment on day 35 or on day 135.¹⁷² In view of the extreme dosage, this also proves that MPH is not very dangerous.

4.2. Short-term medication¶

The improvements with drug treatment end (at least with stimulants) immediately when the medication is stopped, with other drugs with a mirror effect after approx. 14 days at the latest.
The learning effects of neurofeedback and behavioral therapy are better with medication.
The effectiveness of non-drug therapy is long-lasting; however, sufficiently long and intensive therapy (6 months to 3 years) is required.
With neurofeedback, a continuation of treatment successes was observed 6 months after the end of treatment¹¹¹⁷³ Kühle observed that treatment successes persisted years later in some cases, but not in others.

4.3. Long-term medication¶

There are indications that longer-term medication could bring about a post-maturation of those brain structures that are affected by a developmental delay in ADHD.

In ADHD, dysfunctional executive functions are associated with a reduced amount of brain matter in the cortex.¹⁷⁴ In children with ADHD, the growth of brain matter in the cortex is significantly reduced, with the greatest delays in the PFC and ACC.¹⁷⁴
Adult people with ADHD who are treated with stimulants have a significantly larger brain mass in the relevant brain regions than adult people with ADHD who are not treated with stimulants.¹⁷⁵ This could indicate that treatment with stimulants can make up for or compensate for the developmental delay.¹⁷⁴
In people with ADHD who still showed the full ADHD symptoms as adults, no post-maturation (i.e. growth) of the brain mass in the relevant brain regions was recognizable.¹⁷⁶

4.4. Multimodal treatment: non-drug therapy and medication¶

There is evidence that dopaminergic ADHD medications - in the study particularly D-amphetamine medications (levodopa, which was also mentioned, is not suitable as an ADHD medication) - can increase neuroplasticity and thus increase the success of psychotherapy.¹⁷⁷

We are convinced that psychotherapeutic measures are significantly less effective for people with ADHD who are not medicated, as the ability to learn and absorb is massively reduced by ADHD itself. Since ADHD medication improves the impaired neuroplasticity in ADHD and thus often enough creates the ability to learn in the first place, we believe that prior medication is clearly recommended for successful psychotherapy.
Apart from this, we believe that psychotherapy makes little sense if the patient does not even know what the condition they are supposed to achieve through therapy feels like. This feeling can only be conveyed to the person with ADHD after a longer period (1 year or more) of properly adjusted medication.
This is particularly true for people with ADHD, as motivation in ADHD is altered in the direction of significantly increased intrinsic control. ADHD means that following extrinsic prompts is massively more difficult.

5. Compatibility¶

A higher odds ratio (OR) means poorer tolerability than placebo.

• Amphetamine drugs
  • In children (OR: 2.30)³⁴
  • In adults (OR: 3.26)³⁴
• Atomoxetine (OR: 2.33)³⁴
• Methylphenidate (OR: 2.39)³⁴
• Guanfacine in children and adults (OR: 2.64)³⁴
• Modafinil (OR: 4.01)³⁴

6. Comparative values: Effect size for other disorders¶

6.1. Effect size of medication for depression: 0.30¶

Effect size of antidepressants (SMD):¹⁷⁸
Amitriptyline: 0.48
Duloxetine: 0.37
Mirtazapine: 0.37
Venlafaxine: 0.33
Clomipramine: 0.33
Paroxetine: 0.32
Milnacipran: 0.30
Escitalopram: 0.29
Sertraline: 0.27
Vortioxetine: 0.28
Agomelatine: 0.26
Bupropion: 0.25
Citalopram: 0.24
Fluoxetine: 0.23

The average of the antidepressants mentioned is therefore 0.30.

Electroconvulsive therapy showed an Effect size on depression of 0.69 compared to ketamine.¹⁷⁹

6.2. Effect size of psychotherapy for depression 0.22¶

Effect size in SMD.
NNT: Number needed to treat. For 100% treatment success, NNT = 1. The higher the NNT, the worse the efficacy.

Effect size of psychotherapy for depression:¹⁸¹⁸⁰

• Behavioral activation: 0.82 (NNT = 2; 11 studies)
• Mindfulness-based cognitive behavioral therapy (MBCT): 0.73 (NNT = 3; 6 studies)
• Cognitive behavioral therapy: 0.71 (NNT = 3; 159 studies) to 0.79 (409 studies)¹⁸¹
• Interpersonal psychotherapy: 0.67 (NNT = 3; 22 studies)
• Problem-solving therapy: 0.48 (NNT = 4; 21 studies)
• Supportive therapy: 0.52 (NNT = 4; 17 studies)
• Psychodynamic therapy: 0.44 (NNT = 4; 8 studies)

The studies are subject to a strong bias. The example of cognitive behavioral therapy:
When only the higher-quality studies were taken into account, the Effect size decreased by 0.2.¹⁸²
When all studies that compared with waiting list were excluded, the Effect size decreased by 0.17.¹⁸
When all studies with a high risk of bias were excluded, the Effect size decreased by 0.32 to 0.39 (NNT = 5; 34 studies)¹⁸
If publication bias was also taken into account, the Effect size dropped to 0.34 (NNT = 5; 38 studies)¹⁸

If only the studies that fulfill all basic quality criteria are considered, an effect size of only 0.22 (instead of 0.74) remains.¹⁸⁰¹⁸³

6.3. Effect size of medication for ASA¶

According to a meta-analysis, ASA as a whole improved by

• Methylphenidate: 0.53 in the teacher rating (k = 2, n = 36)¹⁸⁴

A meta-analysis examined k = 125 RCTs with n = 7,450 children/adolescents and k = 18 RCTs with n = 1,104 adults in comparison to placebo:¹⁸⁵

Social communication problems:

• Children and young people
  • 0.27 Aripiprazole (k = 6)
  • 0.51 Methylphenidate (k = 3, n = 63)¹⁸⁴

Repetitive behavior:

• Children and young people
  • 0.60 Risperidone (k = 6)
  • 0.49 Atomoxetone (k = 3)
  • 0.35 Bumetanide (k = 4)
  • 0.34 Methylphenidate (k = 3, n = 69)¹⁸⁴
• Adults
  • 1.2 Fluoxetine (k = 1)
  • 1.0 Fluvoxamine (k = 1)
  • 0.97 Risperidone (k = 1)
  • 0.41 Oxytocin (k = 6)

Irritability:

• 0.90 Atypical antipsychotics compared to placebo; k = 12, n = 973¹⁸⁶
• 0.18 Neurohormones compared to placebo; k = 8, n = 466 participants¹⁸⁶
• 0.20 ADHD medication compared to placebo; k = 10 studies, n = 400 participants¹⁸⁶
• 0.06 Antidepressants compared to placebo; k = 3 studies, n = 267 participants¹⁸⁶

Aggression:

• 0.44 Atypical antipsychotics compared to placebo; k = 1, n = 77¹⁸⁶

Reduce self-harm:

• 1.43 Atypical antipsychotics compared to placebo; k = 1, n = 30¹⁸⁶
• 0.62 ADHD medication compared to placebo; k = 1 studies, n = 16 participants¹⁸⁶

ADHD symptoms in children and adolescents with ASD:

• Methylphenidate
  • Hyperactivity
    • 0.63 (parent rating) to 0.81 (teacher rating)¹⁸⁷
    • 0.78 (teacher evaluation)¹⁸⁸
  • Inattention 0.36 (parent rating) to 0.30 (teacher rating)¹⁸⁷
• Atomoxetine
  • Hyperactivity 0.49 (parent rating) to 0.43 (teacher rating)¹⁸⁷
  • Inattention 0.54 (parent rating) to 0.38 (teacher rating)¹⁸⁷

6.4. NNT for other disorders¶

Number needed to treat for various disorders:¹⁸

• Schizophrenia Response
  • Antipsychotics: NNT = 7
• Schizophrenia relapse prevention
  • Antipsychotics: NNT =N 3
• Depression - Response
  • SSRI: NNT = 7
• Depression - Relapse prevention
  • SSRI: NNT = 5
• Depression - Remission
  • Cognitive behavioral therapy vs. medication: NNT = 34
• Alcohol - Relapse
  • Acamprosate: NNT = 10
  • Naltrexone: NNT = 50