Complementary Therapies in Medicine (2011) 19, 216—227

available at www.sciencedirect.com

journal homepage: www.elsevierhealth.com/journals/ctim

Complementary medicines (herbal and nutritional

products) in the treatment of Attention Deficit
Hyperactivity Disorder (ADHD): A systematic review
of the evidence夽
Jerome Sarris a,b,∗, James Kean b, Isaac Schweitzer a, James Lake c

a
The University of Melbourne, Faculty of Medicine, Department of Psychiatry, Australia
b
Swinburne University of Technology, Centre for Human Psychopharmacology, Australia
c
Arizona Centre for Integrative Medicine, Tucson, AZ, United States

KEYWORDS Summary
ADHD; Overview: Complementary and Alternative Medicines (CAMs) are frequently given to children
Attention Deficit and adolescents for reputed benefits in the treatment of hyperkinetic and concentration dis-
Hyperactivity orders such as Attention Deficit Hyperactivity Disorder (ADHD). In such vulnerable populations
Disorder; high quality evidence is required to support such claims.
Paediatric; Aims: The aim of the paper is to assess the current evidence of herbal and nutritional inter-
Complementary ventions for ADHD using a systematic search of clinical trials meeting an acceptable standard
medicine; of evidence.
Herbal medicine; Methods: PubMed, PsycINFO, Cochrane Library and CINAHL were searched up to May 26th, 2011
Nutrition; for randomised, controlled clinical trials using CAM products as interventions to treat ADHD. A
Zinc; quality analysis using a purpose-designed scale, and an estimation of effect sizes (Cohen’s d)
Omega-3 where data were available, were also calculated.
Results: The review revealed that 16 studies met inclusion criteria, with predominant eviden-
tiary support found for zinc, iron, Pinus marinus (French maritime pine bark), and a Chinese
herbal formula (Ningdong); and mixed (mainly inconclusive) evidence for omega-3, and L-acetyl
carnitine. Current data suggest that Ginkgo biloba (ginkgo), and Hypercium perforatum (St.
John’s wort) are ineffective in treating ADHD.

夽 Dr Jerome Sarris is funded by an Australian National Health & Medical Research Council fellowship (NHMRC funding ID 628875), in a

strategic partnership with The University of Melbourne and the National Institute of Complementary Medicine at Swinburne University of
Technology.
∗ Corresponding author at: The University of Melbourne, Department of Psychiatry, The Melbourne Clinic, 2 Salisbury Street, Richmond,

Victoria 3121, Australia. Tel.: +61 03 9420 9350; fax: +61 03 9427 7558.
E-mail address: [email protected] (J. Sarris).

0965-2299/$ — see front matter © 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ctim.2011.06.007
CAM and ADHD 217

Conclusion: The research suggests only some CAMs may be beneficial in ADHD, thus clinicians
need to be aware of the current evidence. Promising candidates for future research include
Bacopa monniera (brahmi) and Piper methysticum (kava), providing potential efficacy in improv-
ing attentional and hyperkinetic disorders via a combination of cognitive enhancing and sedative
effects.
© 2011 Elsevier Ltd. All rights reserved.

Contents

Introduction.............................................................................................................. 217
Methods.................................................................................................................. 218
Results ................................................................................................................... 218
Overview of results ...................................................................................................... 218
Natural products ......................................................................................................... 218
Nutritional medicines .................................................................................................... 218
Herbal medicines ........................................................................................................ 219
Discussion ................................................................................................................ 222
Conflict of interest statement............................................................................................ 224
Appendix 1. Intervention search terms................................................................................. 225
References ............................................................................................................... 225

Introduction Diagnostic tools used to establish a diagnosis of atten-

tion deficit hyperactivity/impulsivity disorder involve the
The number of children diagnosed with Attention Deficit clear understanding that symptoms have been present and
Hyperactivity Disorder (ADHD) has grown markedly since persistent for a minimum of six months prior to the age
being recognised as a specific disorder in the 1970s. The of seven, are considered maladaptive, not consistent with
prevalence rate of ADHD within Western cultures is approxi- the child’s developmental level, and cannot be explained by
mately 5%, and remains the most common psychiatric illness other psychiatric or medical disorders.12 Symptoms relating
among young children, with an estimated 50% of these to ADHD include inattentiveness in the classroom or at home;
children retaining ADHD symptoms for the rest of their or an inability to carry out simple instructions or sustain task
lives.1,2 The economic consequences of ADHD persisting into attention which can result in careless mistakes in school
adulthood are significant, with one U.S. analysis finding an work and reduce motivation for subsequent participation.
average of 35 days of annual lost work performance, rep- Hyperactivity and/or impulsivity diagnosis include symp-
resenting 120 million days of annual lost work in the labor toms of restlessness that exist in the hands or feet during
force, equivalent to $19.5 billion lost human capital.3 times of sitting or sleeping and running, moving erratically
The aetiology of ADHD and the dysfunction of the or talking excessively which cause disruption in an other-
neuro-circuitry within prefrontal cortex have two potential wise calm environment.12 Comorbidity is common in ADHD,
theories; maturational lag,4 or developmental deviation.5,6 with strong links to oppositional defiance disorder and learn-
A lag in developmental maturation delineates that normal ing disorders in children; and major depressive disorder,
maturing of the prefrontal cortex is delayed and depend- anxiety disorders, social dysfunction and substance abuse
ing on the severity of symptoms of the child, will gradually in adults.13,14 Academic issues surrounding ADHD in child-
match the maturation level of normal peers.7 Develop- hood are linked to a higher drop-out rate from secondary
mental deviation has been found in electroencephalograph (high) school with fewer than 5% completing a university
(EEG) studies that have revealed that despite age changes, degree.15 A large proportion of ADHD adults are found to
ADHD symptoms and associated cognitive differences remain be unemployed15 with a significant number of those with
resilient as maturation continues, and is on an abnormal employment taking considerable amounts of unpaid leave.17
developmental path to peers.5,6 There has been extensive Conventional treatment options usually include either
research into the causes of ADHD including its high heritabil- in isolation or in combination, a pharmaceutical com-
ity and genetic influences that predispose a child to deficits ponent, a behavioural component, and a psychosocial
in dopamine and serotonin transmission.8 Other causes have component. Pharmacotherapies which inhibit the re-uptake
been attributed to harmful exposure to the foetus/child of noradrenline and dopamine such as the psychos-
in the prenatal, perinatal, postnatal and early childhood timulants methylphenidate and dextroamphetamine, and
phases.9 In utero exposure to excess alcohol, tobacco and non-stimulating pre-frontal cortex noradrenaline re-uptake
lead have been linked to an increased risk of ADHD,10 while inhibitor atomoxetine, are the standard Western treatment
studies on diet have found that ADHD symptoms may become of ADHD.2 Selective serotonin reuptake inhibitors (SSRIs) and
exacerbated when certain additives or food preservatives other antidepressants are also used with varying degrees
are consumed.11 of success. A third of ADHD patients who take stimu-
218 J. Sarris et al.

lants for ADHD report significant adverse effects including quality total out of ten. Quality assessment of the papers
anorexia, weight loss, abdominal pains, sleep disturbances, was independently rated to assess inter-rater reliability.
headaches, irritability, depressed mood and appetite,16—20 Effect sizes were reported in all placebo-controlled studies
with some reports of stimulant induced psychosis.21 where the results were significant (small clinical effect = 0.2,
Increasing apprehension regarding stimulant medication medium = clinical effect 0.5, large clinical effect = 0.8). We
and the ramifications of its use on children has led to the calculated an effect size as Cohen’s d28 by firstly subtract-
investigation and acknowledgment of alternative therapeu- ing the differences between the results on the assessment
tic medications.22 More than 50% of parents of children scale of the intervention and placebo, then dividing this by
diagnosed with ADHD treat their children’s symptoms using the pooled standard deviation at baseline.
one or more Complementary or Alternative Medicines (CAMs) Purpose-designed questionnaire:
including vitamins, but few disclose this to their child’s
paediatrician.23 Many individuals diagnosed with ADHD use 1. Was the study described as randomised?
CAMs alone or adjunctively with conventional pharmaco- 2. Was the randomisation protocol detailed and appropri-
logical treatments, with one study of 822 children with ate?
diagnosed or suspected ADHD revealing 12 percent had used 3. Was the study described as double-blind?
CAMs.24 Due to the prevalent use of CAMs by children with 4. Was the blinding process detailed and appropriate?
ADHD, evidence is required to support claims of efficacy, 5. Did the study have a control group?
especially as this is a vulnerable group. Previous publications 6. Was the control detailed and appropriate?
have explored this area, however to date no comprehensive 7. Was there an adequate exclusion criterion?
systematic review has been conducted, assessing the quality 8. Was the intervention used at a therapeutic dose?
of studies and the strength of their clinical effects (effect 9. Was there a description of withdrawals and dropouts?
sizes). Thus, the purpose of this paper is to present a sys- 10. Were the data clearly and adequately reported?
tematic review of CAM natural products used for treating
ADHD. Yes = 1 point; no = 0 points; total/10.

Methods Results

The electronic databases MEDLINE (PubMed), CINAHL, Overview of results

PsycINFO, and The Cochrane Library were accessed up to
May 26th, 2011 (see Fig. 1 for systematic review flowchart). Out of 2354 located potential studies in the area of CAMs
PubMed was searched using ADHD search terms in combi- and ADHD, 233 were found to be RCTs. Two hundred and
nation with specific CAMs (herbal and nutritional medicine) seventeen of these were eliminated, commonly due to irrel-
interventions (see Appendix 1 for intervention search term evance, methodological weakness (e.g. small sample, not
list). Papers that met the inclusion criteria were human ran- controlled, or randomised), or the study not having a pri-
domised controlled trials (RCTs) of sufficient methodological mary focus on attentional, behavioural or hyperactivity
rigor. outcomes. This left 16 clinical trials for inclusion. Results
Inclusion criteria: were coded under ‘‘Nutritional Medicines’’ and ‘‘Herbal
Medicines’’. Five nutritional interventions met criteria for
(1) Sample consisting of children or adolescents (aged inclusion: zinc, iron, omega-3, vitamin C, and acetyl-
5—17) or adults (aged 18—65) L-carnitine, while four herbal medicines were included:
(2) Primary diagnosis of ADHD, or marked level of attention Ginkgo biloba (ginkgo), Hypericum perforatum (St. John’s
or hyperactivity on recognised scale wort), Pinus marinus (French maritime pine bark), and Ning-
(3) Randomised and controlled design, or CAMs vs. a posi- dong Granule (traditional Chinese herbal formula).
tive control (e.g. a psychostimulant)
(4) Sample size ≥20 (10 if a cross-over study) Natural products
(5) Duration of intervention ≥1 week
(6) Have measurable outcomes on attention or hyperactiv- Nutritional medicines
ity scale Eleven studies using nutrients met criteria for inclusion.
(7) Rating on quality scale of 5 (see below for details of These had an overall quality rating of 7.6 (range 7—9),
scale) with six revealing a quality rating of nine out of ten (see
(8) Full paper in English Table 1). The average sample size and duration of the stud-
ies were 92 (range 23—400), and 18.5 weeks (range 6—52),
All other papers that did not meet these criteria were respectively. The intervention with the most research was
excluded. Each paper was analysed for methodological qual- found to be omega-3 (fish oil, DHA, or flaxseed oil). In
ity using a purpose-designed scale based on the Jadad our systematic review of the literature, four omega-3 RCTs
scale25 (as first used in Sarris and Byrne,26 and in subsequent were located that met inclusion criteria, with only one
reviews e.g. Pase et al.27 ). The Jadad scale uses three pri- producing significant positive results on the main primary
mary quality factors — randomisation, blinding and reported ADHD outcome measures. Two of the omega-3 studies,
withdrawals. The modified augmented version also assesses Stevens et al.29 (480 mg/day DHA; 80 mg/day EPA) and
other methodological factors — exclusion criteria, interven- Raz et al.30 (240 mg/day linoleic acid: LA, omega-6 and
tion used, control used, and data reporting to provide a 60 mg/day alpha-linolenic acid: ALA, omega-3) had some
CAM and ADHD 219

CAMs and ADHD

2354 hits

Eliminated Clinical trials searched

(n=2121) (n=233)

Eliminated (n=207) Relevant clinical trials

Not relevant (n=26)

Eliminated (n=10)
RCTs meeƟng inclusion criteria
Small sample (n=3), Not controlled
(n= 16)
or randomised (n=4), Replicated
study (n=1), Poor quality (n=2)

Figure 1 Systematic review flowchart.

methodological flaws that may have contributed to the neg- than use of the psychostimulant alone.35 A recent study
ative results, including the use of olive oil29 or vitamin by Arnold et al.36 did not however confirm the results
C30 as a placebo. This may have potentially contributed of the previous zinc studies. Zinc glycinate was randomly
to the better outcomes on the placebo groups. One study assigned to 52 children with ADHD for 13 weeks (8 weeks
using a DHA-predominant blend (510 mg/day; 100 mg/day monotherapy and then 5 weeks with added D-amphetamine).
EPA) as a predominantly adjunctive intervention to psychos- Results revealed on ADHD outcome scales that no signifi-
timulants, found no differential benefit of the supplement cant improvements occurred with zinc supplementation in
compared to controls.31 The ratio of DHA to EPA in this study either dose group (15 mg/day or 30 mg/day) over placebo or
might suggest as a possible reason for the lack of signifi- beyond D-amphetamine.
cant findings with some recent literature indicating that EPA In an RCT involving 23 children (with a small placebo
may be more beneficial in ameliorating ADHD symptoms.32 group (n = 5)), non-anaemic ADHD children with abnor-
Only one placebo-controlled trial using omega-3 with pos- mally low serum ferritin levels were randomised to oral
itive results was located. The Sinn and Bryan reporting iron (ferrous sulphate 80 mg/day) and showed progressive
more improvement on hyperactivity and inattention sub- improvements in ADHD symptoms over placebo.37 In a mul-
scales using long-chain polyunsaturated fatty acids (PUFAs) tisite study of 112 ADHD children randomised to placebo
(558 mg/day EPA; 174 mg/day DHA) compared to palm oil vs. acetyl-L carnitine (ALC: 1000 mg/day to 3000 mg/day
placebo in a sample of 132 children.33 No significant effect depending on weight of child), results revealed that chil-
however occurred on Conners Teacher Rating Scales. Strict dren with predominantly inattentive type ADHD experienced
exclusion criteria ensured no participants in the study were greater improvement over placebo (but there was no dif-
on stimulant medication or any additional omega-3 supple- ferential benefit on primary outcomes in children with
ments within the previous 3 months. combined type ADHD).38 In complex and poorly reported
Findings of studies using zinc in ADHD on various study, L-carnitine (100 mg/kg/day) and placebo was given
outcomes are mainly positive. In a large RCT (n = 400), to 24 Dutch boys over three crossover control periods was
children and adolescents randomised to a high dose of found to have some effects on various outcome measures,
zinc (150 mg/day) experienced significant improvements excepting the primary outcome.39 A novel study using ALC in
over placebo in hyperactivity and impulsivity (but not 51 children with ADHD and a genetic disorder (Fragile-X syn-
inattention).34 A high drop-out rate found in the study may drome), found that after one year of prescribed ALC, greater
however place limits on the significance of the findings. benefit on ADHD symptoms was found over placebo.40
This was due mainly to protocol violations (zinc: 25.7%;
placebo: 28.7%) rather than adverse reactions to the treat-
ment (zinc: 12.3%; placebo: 8.5%). Another study adding Herbal medicines
zinc (55 mg/day) to psychostimulants (1 mg/kg/day) in 44 Five studies using herbal medicines met criteria for inclu-
children resulted in a greater improvement in symptoms sion. These had an overall average quality rating of 8
 220
Table 1 CAM evidence in ADHD (nutrients).

Intervention First author Methodology Duration Result Effect size Quality/10 Comment
(weeks)

Zinc Arnold (2011) DB, RAN, PC 13 (8 At conclusion of the CPRS: 9 Adequate baseline zinc
Zinc group 1 (n = 20; controlled + 5 controlled phase, no high-dose serum level may have
15 mg/day) or zinc group MPH add-on) significant difference group limited results. Use of
2 (n = 8; 30 mg/day) vs. was found on primary (30 mg/day) glycinate zinc chelation
Placebo (n = 24) outcomes between zinc Inatt: −0.54b may be less effective
52 children and placebo. Addition of Hyp: −0.27b than sulphate chelation
Aged 6—14 MPH to zinc did not alter
result
Bilici (2004) DB, RAN, PC 12 Zinc was superior to ADHDS 9 Results of this large high
Zinc (n = 202) vs. placebo placebo in reducing both Hyp: 0.26a,** quality study encourages
(n = 198) hyperactive/impulsive Impul: 0.18a,* use of zinc, especially in
400 children and and impaired deficient populations
adolescents socialisation symptoms.
Aged 6—14 Did not reduce
inattention
Akhondzadeh DB, RAN, PC adjuvant 6 Significant result over ADHD-P-RS 9 Augmentation of MPH
(2004) study placebo after week 2 and 1.46a,* with zinc may be
MPH + zinc (n = 22) vs. at completion of study advised. No participants
MPH + placebo (n = 22) with a strong effect size. had received stimulant
44 children Both Parent and Teacher medication prior to the
Aged 5—11 ratings improved as the commencement of the
study progressed study
Iron Konofal (2008) DB, RAN, PC 12 Progressive significant ADHD-RS 9 Very specific population
Iron (n = 18) vs. placebo decrease on ADHD-RS Inatt: 0.92** group. Young population.
(n = 5) over placebo with a Hyp: 0.63** Small placebo group
23 children strong effect size; iron relative to zinc group
Aged between 5 and 8 also improved ADHD
with low ferritin levels symptoms on CGI-I
EFA supplement Stevens (2003) DB, RAN, PC 16 Clear benefit from ASQ-P 8 Specific classification
LC-PUFAs (n = 25) vs. LC-PUFAs was not 0.12b involving thirst/dry skin
placebo (n = 25) observed on major DBD signs may not be
50 children parent or teacher rating Hyp: −0.09b generalisable to broad
Aged 6—13 scales Inatt: −0.10b ADHD population. Olive
ADHD with thirst/dry oil as a placebo may

J. Sarris et al.
skin have confounded effects
Controls with few
thirst/dry skin symptoms
 CAM and ADHD
Raz (2009) DB, RAN, PC 7 No significant DSM-P 8 Placebo used contained
EFAs (n = 39) vs. placebo differences were found Inatt: −0.37b 1000 mg of vitamin C
(n = 39) between the groups Hyp: 0.15b which cannot be viewed
78 children Aged 7—13 as an inert control;
vitamin C has been
shown to improve the
outcome of ADHD
Sinn (2007) 3-Arm, DB, RAN, PC, CO 30 Significant results on CPRS 7 Significant results on
EFA (n = 36) vs. EFA + MV many subscales Hyp: 0.26* CPRS at week 15. Single
(n = 41) vs. placebo compared to placebo. No Inatt: 0.48** crossover (placebo to
(n = 27) effects found with (at week 15) EFA) in weeks 16—30
132 children (104 addition of reiterated these results.
completers) micronutrients High drop-out rate from
Aged 7—12 study mainly due to
non-compliance or
protocol violation
Voigt (2001) DB, RAN, PC, ADJ 16 No statistical significant CPRS 9 High quality design with
DHA (n = 27) vs. placebo improvement in any Non-significant strict inclusion criteria
(n = 27) objective or subjective data not unable to find
54 children measure of ADHD when provided significance with
Aged 6—12 DHA was given with DHA-only supplement.
pre-existing stable Supplement contained
psychostimulant no EPA
medication
Acetyl-L- Arnold (007) DB, RAN, PC 16 No effect on overall CTRS 8 High quality study with
carnitine ALC (n = 53) vs. placebo ADHD rating outcomes. Inatt: −0.18b non-significant results.
(n = 59) Superiority of ALC over Results of sample varied
112 children placebo in the according to geography
Aged 5—12 inattentive subsample of recruitment
Torrioli (2008) DB, RAN, PC 52 ALC decreased ADHD CGI-P 9 Well conducted complex
ALC (n = 24) vs. placebo symptoms on CGI 0.46* study. Specific group of
(n = 27) significantly over ADHD combined with
51 children (ADHD and placebo at completion Fragile-X-syndrome is
Fragile-X syndrome) after 52 weeks. This difficult to generalise to
Aged 6—13 results was not seen on normal ADHD population
CGI teacher’s rating

221
222 J. Sarris et al.

(range 7—9), with two revealing a maximum quality rat-

DB: double-blind; RAN: randomised; PC: placebo-controlled; POS-C: positive control; ADJ: adjunctive study; Att/Prob: attention problems; Hyp: hyperactivity; Inatt: inattention; Impul:
impulsivity; ADHDS: Attention Deficit Hyperactivity Disorder Scale; ADHD-P-RS: Attention Deficit Hyperactivity Disorder Rating Scale — Parent; ADHD-RS: Attention Deficit Hyperactivity
Disorder-Rating Scale; PTRS: Parent/Teacher Rating Scales; CPRS: Connors Parent Rating Scale; DSM-P: Parent Rating of DSM-IV (Diagnostic and Statistical Manual of Mental Disorders,
Fourth Edition) Symptom Criteria; CGI: Clinical Global Impression; CGI-P: Connor’s Global Index —Parent; ASQ-P: Connor’s Abbreviated Symptom Questionnaires —Parent; CPRS: Connor’s
Parent Rating Scale; CTRS: Connor’s Teacher Rating Scale; CBCL: Child Behaviour Checklist; MPH: methylphenidate; MV: multivitamin; LC-PUFAs: long-chain polyunsaturated fatty acids;
Only measured behaviour
of small sample of ADHD
effects of the treatment
ing of nine out of ten (see Table 2). Average sample size

The data provided did

multiple cross-overs).
and duration were 52 (range 24—61) and 7 weeks (range

not clearly indicate

vs. placebo (due to

4—9), respectively. A recent study by Salehi et al. (2009)
found that a ginkgo preparation (80—120 mg/day) had no
comparable benefit to methylphenidate in a sample of
fifty children.41 The time involved in this study (6 weeks)
Comment

may not have allowed for clinical effects of gingko to

males
reach full potential. Pycnogenol® is a French maritime pine
bark (FMPB) extract which has exhibited anti-oxidant and
anti-inflammatory properties. A study containing 61 ADHD
children aged 6—14 found that FMPB (1 mg/kg/day) was
Quality/10

able to ameliorate the negative symptoms associated with

ADHD including reduced hyperactivity, increased attention
and increased visual-motor coordination.42 Further investi-
8

gation of FMPB by Dvořáková43 found a significant effect on

symptoms of hyperactivity from a re-analysis of the sam-
CTRS and CBCL
Non-significant

ple. However, a small crossover study on an adult population

Effect size

(mean age of 42 years) found no significant effects of FMPB

provided
data not

(1 mg/0.5 kg/day) over 3 weeks administration compared to

placebo.44 It should be noted that neither the positive con-
trol methylphenidate nor FMPB outperformed placebo on
any ADHD rating scales. Dosage in this study was also much
higher than in previous studies. A rigorous study by Weber
rating scales after week
were detailed on ADHD

responders on CTRS vs.

8 (first control period).

et al. investigated St. John’s wort (900 mg/day) in the treat-

No significant effects

50% were classed as

ment of ADHD symptoms but was unable to find any positive

results after an 8 week placebo-controlled intervention.45
17% on placebo

A recent study by Li et al.46 evaluated the efficacy and

safety of a traditional Chinese herbal medicine prepara-
tion (Ningdong: NDG) (5 mg/kg/day) vs. methylphenidate
Result

(1 mg/kg/day) in 72 children with ADHD. The 8-week,

randomised, methylphenidate-controlled, doubled-blinded
trial, found that NDG significantly reduced ADHD symp-
24 (3 × 8 week

toms from baseline after an 8-week medication with fewer

side effects compared to methylphenidate. The study also
treatment
cross-over

showed the herbal formula to be safe and tolerable for ADHD

Duration

periods)
(weeks)

children as monitored by the blood, urine, and stool analysis

and liver and renal function. Interestingly the serum level
of homovanillic acid increased in the NDG group, although
the content of dopamine was not significantly altered during
a Effect size calculated as Cohen’s d via Pearson’s r calculation.

the study.
L-Carnitine (n = 13) vs.

b Effect size conducted on non-significant primary outcomes.

26 children (boys)
DB, RAN, PC, CO

placebo (n = 13)
Methodology

Discussion
Aged 6—13

EFA: essential fatty acids; ACL: acetyl-L-arnitine.

The findings of the systematic review revealed a mixture

of positive and inconclusive evidence from CAM in the
treatment of ADHD. The strength of this review is that
a rigorous systematic search criteria and quality analysis
Van Ouheusden

was conducted. As discussed in the introduction, this is

First author

the first systematic review to our knowledge on natural

products in the treatment of ADHD. A further strength is
(2002)

* Significant at p < 0.05.

that effect sizes were calculated to determine the clinical

** Significant at p < 0.01
Table 1 (Continued)

strength of the result. We however acknowledge a couple

of potential weaknesses with this review. Firstly, we only
reviewed studies in English, thereby some non-English RCTs
Intervention

were excluded involving three Chinese papers (three herbal

L-Carnitine

medicine studies)47—49 and one Russian paper (magnesium

plus B vitamin combination).50 In all studies, the results
favoured the CAM intervention. Secondly, while a systematic
review of the literature is a gold standard methodological
 CAM and ADHD
Table 2 CAM evidence in ADHD (herbal medicines).
Intervention First author Methodology Duration Results Effect size Quality/10 Comment
(weeks)
Ginkgo (Ginkgo Salehi (2009) DB, RAN, POS-C 6 GB has no PRS 8 Results demonstrate the
biloba) GB (n = 25) vs. MPH (n = 25) comparable Inatt: 0.95* strong relative clinical
50 children and efficacy in Hyp: 0.88** effect of a
adolescents comparison to (in favour of MPH) psychostimulant vs. GB.
Aged 6—14 MPH in treating All patients included were
ADHD. MPH was ADHD-combined type
significantly
more effective
on all outcomes
St. John’s wort Weber (2008) DB, RAN, PC 8 (+1 week for No significant ADHD-RS 9 High quality design.
(Hypericum SJW (n = 27) vs. placebo placebo run-in) difference was Hyp: −0.32,̂b Participants were allowed
perforatum) (n = 27) found between Inatt: −0.12,̂b to continue using
54 children Aged 6—17 groups alternative supplements
throughout study
Pycnogenol® Trebaticka DB, RAN, PC Pycnogenol® 4 Significant CAP 8 High quality study design.
French (2006) and (n = 44) vs. Placebo (n = 17) reduction in Hyp: 0.87** Notable relapse in
maritime bark Dvorakova 61 Children Aged 6—14 hyperactivity, Inatt: 1.09** symptoms after cessation
(Pinus (2007)a improvements in indicate a potential
marinus) attention and withdrawal effect
visual-motor
coordination
and
concentration.
Relapse of
symptoms noted
after cessation
Tenenbaum DB, RAN, PC, CO, POS-C 9 (3 × 3 week Neither MPH nor Barkley-ADHD 7 The use of a cross-over
(2002) Pycnogenol® vs. placebo treatments with Pycnogenol® Inatt: −0.52b design may have obscured
vs. MPH (n not detailed) 1 week washout outperformed Hyp: −0.04b the effect. Positive results
24 adults between) placebo of the Trebaticka (2006)
Aged 24—53 study (children sample)
not replicated in this adult
sample study
TC herbal Li (2011) DB, RAN, POS-C 8 An equivocal TARS 9 While no inert control was
formula ND (n = 36) vs. MPH (n = 36) effect was found 0.59* used in this study, it
(Ningdong: 72 children between ND and appears that Ningdong
NDG) Aged 6—13 MPH on Teacher Granule had efficacy,
and Parent albeit with less side
ADHD rating effects (excepting
scales at week 8 hypersomnia)
DB: double blind; RAN: randomised; PC: placebo controlled; POS-C: positive control (e.g. psychostimulant); CO: cross-over; CAP: child attention problems (teacher rated); Hyp: hyper-
activity; Inatt: inattention; Impul: impulsivity; PRS: Parent Rating Scale; Barkley-ADHD: Barkley’s ADHD Rating Scale; TARS: Teacher ADHD Rating Scale; MPH: methylphenidate; SJW: St.
John’s wort; GB: Ginkgo biloba; TC: traditional Chinese; NDG: Ningdong Granule; NA: not applicable.
a An additional analysis conducted and published.
b Refers to Effect size conducted on non-significant primary outcomes.
* Significant at p < 0.05.

223
** Significant at p < 0.01.
224 J. Sarris et al.

technique, such an approach may neglect studies that alter on cognition,55 this activity theoretically due to reuptake
the landscape of the conclusions. inhibition of noradrenaline in the pre-frontal cortex and
Omega-3 studies reveal primarily unsupportive evidence, GABA-ergic effects.55 Safety considerations regarding the
although one recent large study provided some positive use of kava in children and adolescents however would
results on parent-rated measures. Sinn and Bryan33 were need to be strongly considered (especially regarding the
able to find significance with a large population of ADHD potential effect on the liver).56,57 Brahmi has emerging
children using a common behavioural observational mea- evidence as a cognitive enhancer which is beneficial in
sure and utilised the same population age range as the improving various outcomes of mental performance.58,59
other omega-3 studies did. The high drop-out rate from This may be achieved via cholinesterase inhibition and
this study poses an issue when looking to validate such antioxidant effects. Antioxidant effects may potentially
a design, but it should be noted that the drop-out rate be beneficial in ADHD as evidence by the beneficial
was evenly spread between active and non-active groups effects of pine bark, which due to the oligomeric
and that the drop-out rate was more often due to non- proanthocynadin compounds provides a strong antioxidant
compliance than to adverse events. Raz et al.30 used a activity.42
greater dosage of omega-6 (240 mg per capsule) over omega- It is worth noting that five studies were found to use
3 (60 mg per capsule) in their study which may be why some form of Continuous Performance Test (CPT). A CPT
significance was not reached. As well as the issue of is an objective neurophysiological measure of attention,
dosage, vitamin C was used for placebo which has signifi- impulsivity and inhibition that removes the subjectivity
cant antioxidant properties and may have also play a role found in behavioural measures.60 When used in this con-
in the lack of findings within the study. Stevens et al.29 text CPT provides reaction times of ADHD children in
were unable to find significance when they used omega- response to stimuli and is interpreted as a measure of atten-
3 (LC-PUFAs) supplementation in children with ADHD and tion or activation processes.61 All five studies using CPT
skin/thirst problems. The specificity of the population may found no significant effects of the tested CAM product over
not be generalisable to a broad ADHD population, and so placebo. These results could indicate that CPTs are sensi-
may have acted as a confounding variable in this case. tive enough to only pick up strong effects, such as from
The use of olive oil as a placebo may mask the beneficial psychostimulants.61
clinical effects of essential fatty acids because an active One potential application of technology to enhance
constituent of olive oil is converted into oleamide which research of CAMs in ADHD is via the use of neuroimag-
is known to affect brain function.51 Additionally, the short ing technologies. These techniques are valuable to observe
durations and low doses of essential fatty acids used in effects of substances on brain waves (electroencephalog-
some studies may not be adequate to result in long-term raphy: EEG), cerebral blood flow (fMRI), and activation of
changes in neuronal membrane structure required for clin- brain function (PET).62 Future application of these tech-
ical improvement. The dosage issue has been explored by niques may reveal biological evidence of effects of CAMs,
a small open-label study (n = 9) in which ADHD children uncovering a physiological effect before it manifests as
were supplemented with high dose EPA/DHA concentrates a psychological change on the attention or hyperactivity
(16.2 g/day) while continuing on stimulant medications.51 symptoms.
Most children were rated by a blinded psychiatrist as having In respect to the results of our systematic review inform-
significant improvements in both inattention and hyperac- ing clinical practice, currently there is no clear picture about
tivity that correlated with reductions in the AA:EPA ratio which, if any, CAMs can be recommended for use in treat-
at the end of 8-week treatment period. Large prospec- ing ADHD. The CAM natural products reviewed provide a
tive trials are needed to replicate these findings. Voigt mixture of results, with the most promising concerning min-
et al.31 used a supplement of DHA of 2415 mg per week erals zinc and iron, and the antioxidant botanical French
in a strictly psychostimulant treated population. The oil maritime pine bark. Mineral status and deficiency should
used in this study had no traces of EPA in it at all, which always be a clinical consideration when treating children
may call into question the balance of DHA and EPA needed and adolescents with ADHD, however beyond addressing
in ADHD treatment.52 Current evidence does not support deficiency, it may be unlikely that a greater effect will
the use of EFA in ADHD as a stand-alone treatment, and occur from supplementation in those with a good diet. While
future studies should focus on its use only in deficient the current omega-3 data are not supportive of its use in
samples. ADHD, future studies using higher dose preparations may
The minerals studied (iron and zinc) displayed mainly reveal better effects, and regardless, can still be advised in
positive evidence of effect on reducing ADHD symptoms. cases of deficiency. Herbal medicines while presently under-
This beneficial effect could be potentially occurring due to researched in this area may yet provide novel treatments of
addressing deficiency, as mineral deficiency is common is ADHD. Interventions involving combinations of herbal and
Western child and adolescent populations.53 Results with L- nutritional medicines to address mineral deficiency, provide
acetyl carnitine were positive in two out of three studies, antioxidant and GABA-ergic effects, and those that mod-
this being potentially due to its effect on the metabolism ulate prefrontal cortex activity may be of benefit in this
and transportation of fatty acids.54 population.
While the herbal medicines St. John’s wort and ginkgo
monotherapies did not show positive results, other botani-
cals still may provide a beneficial effect. Future potential Conflict of interest statement
studies could involve kava (Piper methysticum) or brahmi
(Bacopa monniera). Kava has demonstrated positive effects None declared.
CAM and ADHD 225

Appendix 1. Intervention search terms

Disorders Major interventions Specific interventions

ADHD Nutritional medicine Ginkgo biloba

ADD Nutraceutical Bacopa monniera
Attention Deficit Hyperactivity Disorder Nootropic Bacopa
Attention deficit disorder Ayurvedic medicine Ginkgo
Hyperkinetic syndrome Herbal medicine Vitamins
Oppositional defiance disorder Botanical medicine Minerals
Learning disorders Natural medicine Zinc
Traditional Chinese medicine Magnesium
Complementary medicine Iron
CAM Pycnogenol
Omega-3
Essential fatty acids
Polyunsaturated essential fatty acids
Panax ginseng
Panax quinquefolium
Ginseng

References year-old and 8/9-year-old children in the community: a

randomised, double-blinded, placebo-controlled trial. Lancet
1. Wolraich ML, Wibbelsman CJ, Brown TE, Evans SW, Gotlieb 2007;370(9598):1560—7.
EM, Knight JR, et al. Attention-deficit/hyperactivity disorder 12. American Psychiatric Association. Diagnostic and statistical
among adolescents: a review of the diagnosis, treatment, and manual of mental disorders. 4th ‘Text Revision’ ed. Arlington:
clinical implications. Pediatrics 2005;115(6):1734—46. American Psychiatric Association; 2000.
2. Biederman J. Attention-deficit/hyperactivity disorder: a 13. Newcorn JH, Weiss M, Stein MA. The complexity of ADHD: diag-
selective overview. Biological Psychiatry 2005;57(July nosis and treatment of the adult patient with comorbidities.
(11)):1215—20. CNS Spectrums 2007;12.(8 Suppl. 12).
3. Kessler RC, Adler L, Ames M, Barkley RA, Birnbaum H, Green- 14. Biederman J, Faraone SV, Spencer T, Wilens T, Norman D,
berg P, et al. The prevalence and effects of adult attention Lapey KA, et al. Patterns of psychiatric comorbidity, cogni-
deficit/hyperactivity disorder on work performance in a nation- tion, and psychosocial functioning in adults with attention
ally representative sample of workers. Journal of Occupational deficit hyperactivity disorder. American Journal of Psychiatry
and Environmental Medicine 2005;47(6):565—72. 1993;150(12):1792—8.
4. Lazzaro I, Gordon E, Whitmont S, Plahn M, Li W, Clarke S, 15. Cimera R. Making ADHD a gift: teaching Superman how to fly,
et al. Quantified EEG activity in adolescent attention deficit vol. 16. Lanham: Scarecrow Press Inc.; 2002.
hyperactivity disorder. Clinical EEG Electroencephalography 16. Schachter HM, Pham B, King J, Langford S, Moher D. How
1998;29(1):37—42. efficacious and safe is short-acting methylphenidate for the
5. Chabot RJ, Serfontein G. Quantitative electroencephalo- treatment of attention-deficit disorder in children and adoles-
graphic profiles of children with attention deficit disorder. cents? A Meta-analysis. Canadian Medical Association Journal
Biological Psychiatry 1996;40(10):951—63. 2001;165:1475—88.
6. Clarke AR, Barry RJ, McCarthy R, Selikowitz M. Elec- 17. Hutchins P. The Clinician’s guide to psychotropic prescribing
troencephalogram differences in two subtypes of in children & adolescents. Child and Adolescent Mental Health
Attention-Deficit/Hyperactivity Disorder. Psychophysiology Statewide Network; 2004.
2001;38(2):212—21. 18. Sonuga-Barke EJS, Coghill D, Wigal T, Debacker M, Swanson J.
7. El-Sayed E, Larsson JO, Persson HE, Santosh PJ, Rydelius PA. Adverse reactions to methylphenidate treatment for attention-
‘‘Maturational lag’’ hypothesis of attention deficit hyperac- deficit/hyperactivity disorder: structure and associations with
tivity disorder: an update. Acta Paediatrica, International clinical characteristics and symptom control. Journal of Child
Journal of Paediatrics 2003;92(7):776—84. and Adolescent Psychopharmacology 2009;19(6):683—90.
8. Wallis D, Russell HF, Muenke M. Review: Genetics of attention 19. Tonge BJ. Common child and adolescent psychiatric problems
deficit/hyperactivity disorder. Journal of Pediatric Psychology and their management in the community. Medical Journal of
2008;33(10):1085—99. Australia 1998;168(5):241—8.
9. Swanson JM, Kinsbourne M, Nigg J, Lanphear B, Stefanatos 20. Efron D, Jarman F, Barker M. Methylphenidate versus
GA, Volkow N, et al. Etiologic subtypes of attention- dexamphetamine in children with attention deficit hyper-
deficit/hyperactivity disorder: brain imaging, molecular activity disorder: a double-blind, crossover trial. Pediatrics
genetic and environmental factors and the dopamine hypoth- 1997;100(6):662—6.
esis. Neuropsychology Review 2007;17(1):39—59. 21. Berman SM, Kuczenski R, McCracken JT, London ED. Poten-
10. Curtis LT, Patel K. Nutritional and environmental approaches tial adverse effects of amphetamine treatment on brain and
to preventing and treating autism and attention deficit hyper- behavior: a review. Molecular Psychiatry 2009;14(2):123—
activity disorder (ADHD): a review. Journal of Alternative and 42.
Complementary Medicine 2008;14(1):79—85. 22. Stubberfield TG, Wray JA, Parry TS. Utilization of alternative
11. McCann D, Barrett A, Cooper A, Crumpler D, Dalen L, Grimshaw therapies in attention-deficit hyperactivity disorder. Journal of
K, et al. Food additives and hyperactive behaviour in 3- Paediatrics and Child Health 1999;35(5):450—3.
226 J. Sarris et al.

23. Chan E, Rappaport LA, Kemper KJ. Complementary and alter- order. Prostaglandins Leukotrienes and Essential Fatty Acids
native therapies in childhood attention and hyperactivity 2002;67(1):33—8.
problems. Journal of Developmental and Behavioral Pediatrics 40. Torrioli MG, Vernacotola S, Peruzzi L, Tabolacci E, Mila M,
2003;24(1):4—8. Militerni R, et al. A double-blind, parallel, multicenter compar-
24. Bussing R, Zima BT, Gary FA, Garvan CW. Use of com- ison of L-acetylcarnitine with placebo on the attention deficit
plementary and alternative medicine for symptoms of hyperactivity disorder in fragile X syndrome boys. American
attention-deficit hyperactivity disorder. Psychiatric Services Journal of Medical Genetics, Part A 2008;146(7):803—12.
2002;53(9):1096—102. 41. Salehi B, Imani R, Mohammadi MR, Fallah J, Mohammadi
25. Jadad A, Moore A, Carroll D, Jenkinson D, Gavaghan D, McQuay M, Ghanizadeh A, et al. Ginkgo biloba for Attention-
H. Assessing the quality of reports of randomised clinical trials: Deficit/Hyperactivity Disorder in children and adolescents:
is blinding necessary? Controlled Clinical Trials 1996;17:1—12. a double blind, randomized controlled trial. Progress
26. Sarris J, Byrne GJ. A systematic review of insomnia in Neuro-Psychopharmacology and Biological Psychiatry
and complementary medicine. Sleep Medicine Reviews 2010;34(1):76—80.
2011;15(2):99—106. 42. Trebatická J, Kopasová S, Hradečná Z, Činovský K, Škodáček I,
27. Pase M, Grima N, Sarris J. The effects of dietary and nutri- Šuba J, et al. Treatment of ADHD with French maritime pine
ent interventions on arterial stiffness: a systematic review. bark extract, Pycnogenol® . European Child and Adolescent
American Journal Of Clinical Nutrition 2011;93(2):446—54. Psychiatry 2006;15(6):329—35.
28. Cohen J. Statistical power analyses for the behavioral sci- 43. Dvořáková M, Ježová D, Blažíček P, Trebatická J, Škodáček I,
ences, vol. 2. Hillsdale: Erlbaum; 1988. Šuba J, et al. Urinary catecholamines in children with atten-
29. Stevens L, Zhang W, Peck L, Kuczek T, Grevstad N, Mahon tion deficit hyperactivity disorder (ADHD): modulation by a
A, et al. EFA supplementation in children with inatten- polyphenolic extract from pine bark (Pycnogenol® ). Nutri-
tion, hyperactivity, and other disruptive behaviors. Lipids tional Neuroscience 2007;10(3—4):151—7.
2003;38(10):1007—21. 44. Tenenbaum S, Paull JC, Sparrow EP, Dodd DK, Green L. An
30. Raz R, Carasso RL, Yehuda S. The influence of short-chain essen- experimental comparison of Pycnogenol® and methylphenidate
tial fatty acids on children with attention-deficit/hyperactivity in adults with Attention-Deficit/Hyperactivity Disorder (ADHD).
disorder: a double-blind placebo-controlled study. Journal of Attention Disorders 2002;6(2):49—59.
Journal of Child and Adolescent Psychopharmacology 45. Weber W, Vander Stoep A, McCarty RL, Weiss NS, Biederman
2009;19(2):167—77. J, McClellan J. Hypericum perforatum (St John’s Wort) for
31. Voigt RG, Llorente AM, Jensen CL, Fraley JK, Berretta MC, attention-deficit/hyperactivity disorder in children and ado-
Heird WC. A randomized, double-blind, placebo-controlled lescents: a randomized controlled trial. JAMA - Journal of the
trial of docosahexaenoic acid supplementation in children with American Medical Association 2008;299(22):2633—41.
attention-deficit/hyperactivity disorder. Journal of Pediatrics 46. Li J, Li Z, Wang S, Wang SZ, Qi FH, Zhao L, et al. Ningdong gran-
2001;139(2):189—96. ule: a complementary and alternative therapy in the treatment
32. Hirayama S, Hamazaki T, Terasawa K. Effect of docosa- of attention deficit/hyperactivity disorder. Psychopharmacol-
hexaenoic acid-containing food administration on symptoms ogy (Berl) 2011.
of attention-deficit/hyperactivity disorder — a placebo- 47. Chen J, Chen YY, Wang XM. Clinical study on treat-
controlled double-blind study. European Journal of Clinical ment of children attention deficit hyperactivity disorder by
Nutrition 2004;58(3):467—73. Jiangqian granule. Zhongguo Zhong Xi Yi Jie He Za Zhi 2002
33. Sinn N, Bryan J. Effect of supplementation with polyunsatu- Apr;22(4):258—60.
rated fatty acids and micronutrients on learning and behavior 48. Chen YH, Huang B, Zhao X. Clinical and experimental study
problems associated with child ADHD. Journal of Developmen- on treatment of childhood hyperkinetic syndrome with Yizhi-
tal and Behavioral Pediatrics 2007;28(2):82—91. dan. Zhongguo Zhong Xi Yi Jie He Za Zhi 2001;21(January
34. Bilici M, Yildirim F, Kandil S, Bekaroglu M, Yildirmis S, Deger O, (1)):19—21.
et al. Double-blind, placebo-controlled study of zinc sulfate 49. Ding GA, Yu GH, Chen SF. Assessment on effect of treatment
in the treatment of attention deficit hyperactivity disorder. for childhood hyperkinetic syndrome by combined therapy of
Progress in Neuro-Psychopharmacology and Biological Psychi- Yizhi mixture and Ritalin. Zhongguo Zhong Xi Yi Jie He Za Zhi
atry 2004;28(1):181—90. 2002;22(April (4)):255—7.
35. Akhondzadeh S, Mohammadi MR, Khademi M. Zinc sulfate 50. Nogovitsina OR, Levitina EV. Effect of MAGNE-B6 on the clinical
as an adjunct to methylphenidate for the treatment of and biochemical manifestations of the syndrome of attention
attention deficit hyperactivity disorder in children: a double deficit and hyperactivity in children. Eksperimental’naia
blind and randomized trial [ISRCTN64132371]. BMC Psychiatry i klinicheskaia farmakologiia 2006;69(January—February
2004;4. (1)):74—7.
36. Arnold LE, Disilvestro RA, Bozzolo D, Bozzolo H, Crowl L, 51. Richardson AJ, Puri BK. A randomized double-blind, placebo-
Fernandez S, et al. Zinc for attention-deficit/hyperactivity dis- controlled study of the effects of supplementation with
order: placebo-controlled double-blind pilot trial alone and highly unsaturated fatty acids on ADHD-related symptoms
combined with amphetamine. Journal of Child and Adolescent in children with specific learning difficulties. Progress
Psychopharmacology 2011;21(February (1)):1—19. in Neuro-Psychopharmacology and Biological Psychiatry
37. Konofal E, Lecendreux M, Deron J, Marchand M, Cortese S, 2002;26(2):233—9.
Zaïm M, et al. Effects of iron supplementation on attention 52. Arterburn LM, Boswell KD, Koskelo EK, Kassner SL, Kelly C,
deficit hyperactivity disorder in children. Pediatric Neurology Kyle DJ. A combined subchronic (90-day) toxicity and neu-
2008;38(1):20—6. rotoxicity study of a single-cell source of docosahexaenoic
38. Arnold LE, Amato A, Bozzolo H, Hollway J, Cook A, acid triglyceride (DHASCO® oil). Food and Chemical Toxicology
Ramadan Y, et al. Acetyl-L-carnitine (ALC) in attention- 2008;38(1):35—49.
deficit/hyperactivity disorder: a multi-site, placebo-controlled 53. Adamson P. Vitamin and mineral deficiency. A global progress
pilot trial. Journal of Child and Adolescent Psychopharmacol- report. Micronutrient Initiative and UN Children’s Fund,
ogy 2007;17(6):791—801. UNICEF; 2004.
39. Van Oudheusden LJ, Scholte HR. Efficacy of carnitine in the 54. Malaguarnera M, Cammalleri L, Gargante MP, Vacante M,
treatment of children with attention-deficit hyperactivity dis- Colonna V, Motta M. L-Carnitine treatment reduces severity of
CAM and ADHD 227

physical and mental fatigue and increases cognitive functions double-blind placebo-controlled randomized trial. Phytother-
in centenarians: a randomized and controlled clinical trial. apy Research 2008;22(December (12)):1629—34.
American Journal of Clinical Nutrition 2007;86(6):1738—44. 59. Tripathi YB, Chaurasia S, Tripathi E, Upadhyay A, Dubey
55. Sarris J, La Porte E, Schweitzer I. Kava: A Comprehensive GP. Bacopa monniera Linn. as an antioxidant: mechanism of
Review of Efficacy, Safety, and Psychopharmacology. Aust N Z action. Indian Journal of Experimental Biology 1996;34(June
J Psychiatry 2011;45(1):27—35. (6)):523—6.
56. Sarris J, Adams J, Wardle J. Time for a reassessment 60. Rosvold HE, Mirsky AF, Sarason I, Bransome Jr ED, Beck LH.
of Kava in anxiety? Complementary Therapies in Medicine A continuous performance test of brain damage. Journal of
2009;17:121—2. Consulting Psychology 1956;20(5):343—50.
57. Teschke R. Kava hepatotoxicity — a clinical review. Annals of 61. Levy F, Hobbes G. Discrimination of attention deficit hyperac-
Hepatology 9; in press. tivity disorder by the continuous performance test. Journal of
58. Stough C, Downey LA, Lloyd J, Silber B, Redman S, Hutchison Paediatrics and Child Health 1997;33(5):384—7.
C, et al. Examining the nootropic effects of a special extract 62. Bandettini PA. What’s new in neuroimaging methods? Annals of
of Bacopa monniera on human cognitive functioning: 90 day the New York Academy of Sciences 2009;1156(March):260—93.