Journal of Clinical and Experimental Neuropsychology

ISSN: 1380-3395 (Print) 1744-411X (Online) Journal homepage: https://www.tandfonline.com/loi/ncen20

Comparative efficacy and safety of

methylphenidate and atomoxetine for attention-
deficit hyperactivity disorder in children and
adolescents: Meta-analysis based on head-to-head
trials

Qiang Liu, Hong Zhang, Qingqing Fang & Lili Qin

To cite this article: Qiang Liu, Hong Zhang, Qingqing Fang & Lili Qin (2017) Comparative efficacy
and safety of methylphenidate and atomoxetine for attention-deficit hyperactivity disorder in children
and adolescents: Meta-analysis based on head-to-head trials, Journal of Clinical and Experimental
Neuropsychology, 39:9, 854-865, DOI: 10.1080/13803395.2016.1273320

To link to this article: https://doi.org/10.1080/13803395.2016.1273320

Published online: 04 Jan 2017.

Submit your article to this journal

Article views: 525

View related articles

View Crossmark data

Citing articles: 4 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=ncen20
JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY, 2017
VOL. 39, NO. 9, 854–865
https://doi.org/10.1080/13803395.2016.1273320

Comparative efficacy and safety of methylphenidate and atomoxetine for

attention-deficit hyperactivity disorder in children and adolescents:
Meta-analysis based on head-to-head trials
Qiang Liua,b, Hong Zhangc, Qingqing Fangd and Lili Qine
a
Pediatrics Department, Shandong Provincial Hospital Affiliated to Shandong University, Shandong University, Jinan, China;
b
Newborn Department, Linyi People’s Hospital, Linyi, China; cDepartment of Pediatrics, Linyi Traditional Chinese Medical Hospital,
Linyi, China; dDepartment of Pediatrics, The People’s Hospital of Lanshan District, LinYi, China; eDepartment of Pediatrics,
Weishan County People’s Hospital, Jining, China

ABSTRACT ARTICLE HISTORY

Introduction: Comparative efficacy and safety are important issues for appropriate drug Received 31 August 2016
selection for attention-deficit hyperactivity disorder (ADHD) treatment. Therefore we Accepted 9 December 2016
conducted a meta-analysis, where we compared atomoxetine (ATX) and methylphenidate KEYWORDS
(MPH) for ADHD treatment in children and adolescents. Method: Literature retrieval was Adverse events;
conducted in relevant databases from their inception to April 2016 to select head-to-head atomoxetine;
trials that compared ATX and MPH in children and adolescents. Outcomes like response attention-deficit
rate, ADHD Rating Scale (ADHD–RS) score, and adverse events were compared between hyperactivity disorder;
ATX and MPH treatments. The standardized mean difference (SMD) and risk ratio (RR) guidelines;
with their corresponding 95% confidence intervals (CIs) were used as the effect size for methylphenidate; response
continuous data or dichotomous data, respectively. Results: Eleven eligible randomized- rate
controlled trials were included, and two of them were double-blind, while the remaining
were open-label. Compared to ATX, MPH showed a higher response rate (RR = 1.14, 95%
CI [1. 09, 1.20]), decreased inattention (SMD = −0.13, 95% CI [−0.25, −0.01]) and lower risk
of adverse events (drowsiness: RR = 0.17, 95% CI [0.11, 0.26; nausea: RR = 0.49; 95% CI
[0.29, 0.85; vomiting: RR = 0.41, 95% CI [0.27, 0.63]). However, MPH presented a higher
risk of insomnia than ATX (RR = 2.27, 95% CI [1.63, 3.15], p < .01). Conclusion: Results of
the meta-analysis add additional evidence of the effectiveness of both ATX and MPH and
suggest that MPH should be a first treatment option in most patients with ADHD.

Attention-deficit hyperactivity disorder (ADHD) is
one of the most common psychiatric disorders. The
prevalence rate of ADHD among children and ado-
lescents is different in different countries: above 8% or
less than 1% (Polanczyk, Salum, Sugaya, Caye, &
Rohde, 2015; Štuhec, Švab, & Locatelli, 2015).
ADHD is characterized by inattention, hyperactivity,
and impulsivity, and is often accompanied by learning
and cognitive disorders. A nationwide cohort study
indicates that the annual mortality rate of ADHD
patients is 5.85 per 10,000 individuals (Smith et al.,
2016).
Currently, major treatments for ADHD are phar-
macotherapy, behavior modification, and social psy-
chotherapy. Several standards have been established
to evaluate ADHD severity and efficacy for pharma-
cotherapy in clinical practices, such as the Clinical
Global Impression Scale (CGI), the Continuous
Performance Task (CPT), and the ADHD Rating
Scale (ADHD–RS; So, Noh, Kim, Ko, & Koh, 2002).
Stimulants such as methylphenidate (MPH) are
the medical treatments approved by the United
States Food and Drug Administration (FDA). It
has an excellent response rate. MPH blocks the
reuptake of dopamine (DA) transporter in the
presynaptic neuron (Madhusoodanan & Goia,
2016) and can reduce the density of DA carriers
in adults (Dresel et al., 2000). In addition, osmotic
controlled-release formulations of MPH (OROS-
MPH) have been designed to allow a long-term

CONTACT Lili Qin qinlili1978@163.com Department of Pediatrics, Weishan County People’s Hospital, Chenghou road, Jining,
Shandong, 277600, China
Qiang Liu and Hong Zhang contributed equally to this work.
© 2017 Informa UK Limited, trading as Taylor & Francis Group
 JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY
effect for ADHD management. Both immediate-
release MPH (IR-MPH) and OROS-MPH have
been found effective for alleviating the symptoms
of ADHD (Wilens et al., 2003). Despite this, non-
stimulant agents such as atomoxetine (ATX) are
supposed to be an alternative to the stimulants
(Wigal, 2009). Meanwhile, it should be noted that
although the efficacy and safety of MPH have been
demonstrated, its use in preschool-aged children is
still off-label (Wolraich et al., 2011).
ATX is a highly selective inhibitor of presynap-
tic norepinephrine transporter (inhibitor constant,
Ki 4.5 nM) with minimal affinity for other nora-
drenergic receptors or neurotransmitter transpor-
ters. Etiology of ADHD is the impaired
metabolism of catecholamine in cerebral cortex
and the decreased inhibition ability of dopaminer-
gic and noradrenergic activities. ATX has an inhi-
bitory role on DA reuptake (Štuhec, 2013) and
hardly has any abuse potential (Heil et al., 2002).
Although efficacy of MPH or ATX has been well
established compared with placebo, controversial
results exist in the comparison between the two
agents. It is found that MPH and ATX have compar-
able efficiency and adverse effects for ADHD treat-
ment (Garg, Arun, & Chavan, 2014). However,
another recent meta-analysis indicates that the effect
size of ATX (standardized mean difference,
SMD = −0.68, 95% confidence interval, CI [−0.76,
−0.59]) is slightly lower than that of MPH
(SMD = −0.75, 95% CI [−0.98, −0.52]) in reducing
ADHD symptoms, compared with placebo (Stuhec,
Munda, Svab, & Locatelli, 2015). Similarly, a study
states that OROS-MPH shows a more significant
improvement than ATX (Yildiz, Sismanlar, Memik,
Karakaya, & Agaoglu, 2011). In addition, immediate
release MPH has a higher clinical response than the
longer acting stimulants and nonstimulants (Peterson,
Mcdonagh, & Fu, 2008). On the other hand, the non-
stimulant medication may be used as a first-line treat-
ment for ADHD patients under certain circumstances
such as with comorbidity (Shier, Reichenbacher,
Ghuman, & Ghuman, 2013). Nevertheless, the evi-
dence-based guidelines for the pharmacological man-
agement of ADHD point out that nonstimulant
medications do not have high effect sizes even under
these conditions (Boleaalamañac et al., 2014).
Previously, a meta-analysis was performed to
evaluate the lisdexamfetamine treatment for
ADHD in children and adolescents, compared
with MPH and ATX. However, the comparison
between MPH and ATX was not considered
855
(Roskell, Setyawan, Zimovetz, & Hodgkins, 2014).
Then, several meta-analyses have compared the
efficiency of MPH and ATX (C. Bushe et al.,
2016; C. J. Bushe & Savill, 2013; Hanwella,
Senanayake, & de Silva, 2011; Hazell et al., 2011).
However, some limitations are presented. For
instance, none of them takes adverse events into
consideration. Bushe et al. mainly emphasize the
suicide-related events (C. J. Bushe & Savill, 2013).
Moreover, the outcome of response rate is not
mentioned (C. Bushe et al., 2016). Although
Bushe et al. have used network meta-analysis,
they only compared efficiencies between OROS-
MPH and ATX (C. Bushe et al., 2016). Hazell
et al. (2011) just included seven studies involving
1,368 patients, and Hanwella et al. (2011) included
nine randomized trials enrolling 2,762 participants.
The sample sizes in both of the two studies were
relatively small. Therefore, we performed this
meta-analysis considering the outcomes of
response rate and the adverse events. Moreover,
subgroup analyses stratified by MPH types and
different evaluation standards were conducted,
attempting to provide a comprehensive compari-
son of the efficiency and safety between MPH and
ATX for ADHD treatment.
Method
Search strategy
The literature retrieval was conducted in PubMed,
Embase, and Cochrane library up to April 2016,
with the key searching terms of “atomoxetine,”
“methylphenidate,” “attention deficit hyperactivity
disorder,” and “children.” The searching strategies
were (atomoxetine) AND (methylphenidate) AND
(attention-deficit hyperactivity disorder) OR
(ADHD) AND (child*) AND (random*).
Inclusion and exclusion criteria
Studies included in the present meta-analysis should
meet the following criteria: (a) The study was a
randomized-controlled trial (RCT) comparing the
efficiency and safety of MPH and ATX; (b) the
study was an English publication; (c) all the ADHD
cases were in accordance with the Diagnostic and
Statistical Manual of Mental Disorders (DSM–IV;
Poling et al., 2013); (d) the patients were children
under 18 years old; (e) the curative outcomes such as
response rate, ADHD–RS–IV total, ADHD–RS–IV
856 Q. LIU ET AL.
inattention, and ADHD–RS–IV hyperactivity/impul-
sivity were evaluated in this study; (f) the outcomes
of adverse events such as headache, insomnia, drow-
siness, anorexia, abdominal pain, nausea, and vomit-
ing were considered.
Exclusion criteria were: (a) non-RCT study; (b)
crossover trials; (c) duplicated publications; (d) the
study combined multiple therapies; (e) patients in
the studies had severe psychotic disorder, develop-
mental disorder, epilepsy, suicide/violence history,
and mental retardation, or had malignant tumor,
heart or kidney failure, and other serious diseases.
Data extraction and quality assessment
The following data information was extracted by
two investigators independently: the first author
name; publication time; study type; country; diag-
nostic criteria; follow-up time; the drug for inter-
vention; the case numbers, mean age, mean drug
dosage in each intervention group; curative evalua-
tion criteria and all the aforementioned outcomes.
Quality of the included RCTs was evaluated by
the Cochrane evaluation system (Higgins & Green,
2008). Disagreements were settled via discussion
with a third investigator.
Statistical analysis
The standardized mean difference (SMD) and risk
ratio (RR) with their corresponding 95% confidence
intervals (CIs) were used as a measure of the effect
size to calculate pooled results for continuous or
dichotomous data, respectively. Heterogeneity across
studies was examined by the Cochrane-based Q sta-
tistic and I2 test (J. Higgins, Thompson, Deeks, &
Altman, 2003), and a randomized-effects model was
applied if substantial heterogeneity was discovered
(p < .05, I2 > 50%); instead, a fixed-effects model
was selected when there lacked a significant hetero-
geneity (p > .05, I2 < 50%). For the outcome of
response rate, studies involving the OROS-MPH
and the therapeutic evaluation criteria of ADHD–
RS–IV Parents Version were extracted to further
perform subgroup analysis.
Publication bias and sensitive analysis
Egger’s test was used to examine whether publica-
tion bias existed among the included studies.
Sensitive analysis was also conducted through
eliminating a single study at one time, to detect
whether a reverse result would be generated after
this removal. The tool used for sensitive analysis
was Stata 12.0 (STATA, College Station, TX, USA).
A reverse result indicated instability of the meta-
analysis.
The software of RevMan 5.3 (Cochrane
Collaboration, http://ims.cochrane.org/revman) and
Stata 12.0 (STATA, College Station, TX, USA) were
utilized for statistical calculation.
Results
Study characteristics
Based on the predefined criteria, a total of 11 eligible
studies were included in our meta-analysis (Cetin,
Torun, & Taner, 2015; Garg et al., 2014; Kemner,
Starr, Ciccone, Hooper-Wood, & Crockett, 2005;
Kratochvil et al., 2002; Newcorn et al., 2008; Schulz
et al., 2012; Shang, Pan, Lin, Huang, & Gau, 2015;
Starr & Kemner, 2005; Su, Yang, Stein, Cao, &
Wang, 2016; Wang et al., 2007; Yildiz et al., 2011).
A detailed process of the study selection is presented
in Figure 1. The included studies were all RCTs: two
(Newcorn et al., 2008; Wang et al., 2007) double-
blind RCTs and nine open-label RCTs. Five studies
were conducted in the USA, while the remaining
were from Asian countries like China, India, and
Turkey. All the studies used DSM–IV as the diag-
nostic criteria for ADHD in children. In a majority
of the studies, patients were the mixed cases of
treating previously with stimulants with the treat-
ment naïve. Seven studies reported the comparison
of ATX versus OROS-MPH, and one study com-
pared ATX versus IR-MPH. The dosage of ATX was
varied from 0.5 to 36 mg kg–1 per day, and that of
MPH was from 0.45 to 32.8 mg kg–1 per day. Three
studies did not provide the specific dosage of either
MPH or ATX (Table 1). The follow-up duration was
varied from 2 to 12 weeks. The response rate was
defined as a 25% or a greater reduction from the
baseline ADHD–RS score, or an investigator-rated
CGI–Improvement (CGI–I) score of 2 or less
(“much improved” or “very much improved”).
Two studies did not use the ADHD–RS score to
determine the response rate: In the study of Cetin
(Cetin et al., 2015), the response rate was regarded
as a 40% reduction in Conners’s Comprehensive
Behavior Rating Scale–Teacher (CRS–T) scores
when compared to baseline values, while Yildiz
 JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY
Figure 1. Flow chart of the study selection.
et al. regarded CGI–Severity Scale (CGI–S) and
CGI–I as the standard (Yildiz et al., 2011).
Quality assessment
As most of the studies were open-label RCTs, there
was a high risk of bias on allocation concealment. In
the study of Cetin et al. (2015), they randomly allo-
cated the patients according to the odd numbers and
even numbers, which might account for the high risk
of bias on allocation concealment (Figure 2).
Nevertheless, risk of other bias was low, and the
overall quality of the included studies was high.
Outcomes
Response rate
Seven studies containing 2,845 cases (MPH: 1644;
ATX: 1201) reported the outcome of response rate.
857
We selected a fixed-effects model to calculate this
outcome due to the lack of significant heterogene-
ity (p = .27, I2 = 19%). The pooled result indicated
that MPH showed a higher response rate than
ATX for the treatment of ADHD (RR = 1.14,
95% CI [1.09, 1.20], p < .05; Figure 3a). Subgroup
analysis revealed that OROS-MPH achieved a
markedly higher response rate than ATX
(RR = 1.16, 95% CI [1.09, 1.23], p < .05;
Figure 3b); meanwhile, MPH that applied
ADHD–RS–IV criteria also achieved an increased
response rate, compared with ATX (RR = 1.17,
95% CI [1.10, 1.23], p < .05; Figure 3c). No obvious
publication bias was detected across the studies
based on Egger’s test (p = .4316).
ADHD–RS score
The ADHD–RS score included three items as total
score, hyperactivity/impulsivity, and inattention.
 858
Q. LIU ET AL.

Table 1. Characteristics of the included articles.

Diagnostic No. of patients Mean dose,
Author, year Study type Country criteria Participants Follow-up Intervention (M/F) Age, yearsa (mg kg–1 per daya) Efficacy assessment
Su 2016 Parallel, open-label, China DSM–IV Mix, 6–16 y 4w ATX 118 9.5 ± 1.9 36 ADHD–RS–IV,
RCT OROS-MPH 119 9.5 ± 1.9 0.5–1.2 CGI–ADHD–S
Shang 2015 Open-label, RCT Taiwan DSM–IV Drug-naive, 8w ATX 80 (70/10) 9.90 ± 2.78 NA ADHD–RS–IV Parents Version
7–16 y OROS-MPH 80 (70/10) 9.64 ± 2.42 NA
Cetin 2015 Open-label, RCT Turkey DSM–IV–TR Mix, 7–16 y 2w ATX 59 (45/14) 9.55 ± 2.71 1.14 ± 0.13 CRS–T
OROS-MPH 61 (53/8) 9.95 ± 2.02 0.73 ± 0.22
Garg 2014 Open-label, RCT India DSM–IV–TR Mix, 6–14 y 8w ATX 36 (29/7) 8.66 ± 2.44 0.61 ADHD–RS–IV Parents Version
IR-MPH 33 (27/6) 8.47 ± 2.22 0.45
Schulz 2012 Parallel group design USA DSM–IV Mix, 7–17 y 8w ATX 18 (15/3) 11.4 ± 3.0 NA ADHD–RS–IV
OROS-MPH 18 (15/3) 11.0 ± 2.4 NA
Starr 2005 Open-label, USA DSM–IV Mix, 6–12 y 3w ATX 58 (50/8) 9.1 ± 2.2 1.1 ± 0.4 ADHD–RS, CGI–S/CGI–I
multicenter, RCT OROS-MPH 125 (100/25) 8.6 ± 2.0 32.8 ± 10.9
Kemner 2005 Open-label, USA DSM–IV–TR Mix, 6–12 y 3w ATX 473 (352/121) 9.22 ± 2.12 0.5 ADHD–RS, CGI–I
multicenter, RCT OROS-MPH 850 (631/219) 8.77 ± 2.00 18
Kratochvil 2002 Open-label, RCT USA, Canada DSM–IV Mix, 7–15 y 10 w ATX 184 (167/17) 10.4 ± 2.1 1.40 ± 0.48 ADHD–RS–IV Parents Version,
MPH 44 (44/0) 10.4 ± 2.1 0.85 ± 0.53 CPRS–R
Newcorn 2008 Double-blind RCT USA DSM–IV Mix, 6–16 y 6w ATX 222 (172/50) 10.3 ± 2.2 1.45 ± 0.32 ADHD–RS–IV
OROS-MPH 220 (156/64) 10.2 ± 2.5 1.16 ± 0.55
Wang 2007 Double-blind RCT China, Korea, DSM–IV Mix, 6–16 y 8w ATX 164 (136/28) 9.4 ± 2.0 1.37 ADHD–RS–IV
Mexico MPH 166 (134/32) 9.9 ± 2.3 0.52
Yildiz 2011 Open-label, RCT Turkey DSM–IV–TR Mix, 8–14 y 12 w ATX 14 (13/1) 9.78 ± 1.36 NA CGI–S, CGI–I
OROS-MPH 11 (9/2) 10.16 ± 1.7 NA
Note. ATX = atomoxetine; OROS-MPH = oral osmotic methylphenidate; IR-MPH: immediate release methylphenidate; DSM–IV = the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition; DSM–IV–TR =
the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision; Mix = subjects could either have been treated previously with stimulants or be treatment naïve; ADHD = attention-deficit
hyperactivity disorder; CGI = Clinical Global Impression Scale; CGI–ADHD–S = CGI–ADHD–Severity scale; ADHD–RS–IV Parents Version = ADHD Rating Scale–IV Parents Version; CRS–T = Conners’s Comprehensive
Behavior Rating Scale–Teacher; CGI–S = CGI–Severity of Illness scale score; CGI–I = CGI–Improvement of Illness scale score; RCT = randomized controlled trial; y = years; w = weeks; m = months; M = male; F =
female; NA = not available; CPRS-R: Conners' Parent Rating Scale–Revised.
a
Data are presented as mean ± standard deviation.
 JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY 859

Figure 2. Quality assessments of the selected studies. (a) Bias risk of the identified studies; (b) sensitivity and specificity of
the included studies. “+” Indicates low risk of bias; “?” represents unclear risk of bias; “–”represents high risk of bias.

We did not detect any remarkable publication bias
among all the studies (Egger’s test: p > .05;
Table 2).
Based on the heterogeneity result, a rando-
mized-effects model was utilized to evaluate the
total score (heterogeneity: p = .0007, I2 = 74%),
while a fixed-effects model was selected for the
other two items (p > .05, I2 < 50%). From the
combined results, it was discovered that for total
score and hyperactivity/impulsivity, there were no
significant differences between ATX and MPH
treatments (p < .05; Figures 4a, 4b); while MPH
was significantly superior to ATX with a reduced
inattention in ADHD children (SMD = −0.13, 95%
CI [−0.25, −0.01], p = .03; Figure 4c).
and OROS-MPH (Table 2). As a result, MPH
dramatically decreased the risk of drowsiness
(somnolence; RR = 0.17, 95% CI [0.11, 0.26],
p < .05; Figure 5a), nausea (RR = 0.49, 95% CI
[0.29, 0.85], p < .05; Figure 5b), and vomiting
(RR = 0.41, 95% CI [0.27, 0.63], p < .05,
Figure 5c), compared with ATX. However, MPH
had a remarkably higher risk of insomnia than
ATX (RR = 2.27, 95% CI [1.63, 3.15], p < .05;
Figure 5d). No significant differences were
detected between the two treatments in the other
adverse events (p > .05). We did not observe any
evident publication bias among all the studies by
Egger’s test (p > .05).

Adverse events
The treatment-emergent adverse events including
abdominal pain, anorexia, drowsiness (somno-
lence), headache, insomnia, nausea, and vomiting
were assessed with the treatments of ATX, MPH,
Sensitivity analysis
No reverse result was discovered after the removal
of any single study, suggesting a stable result of the
present meta-analysis.
860 Q. LIU ET AL.

Figure 3. Comparison of the response rate between methylphenidate (MPH) and atomoxetine (ATX) agents. (a) Overall
analysis of MPH versus ATX; (b) subgroup analysis of osmotic controlled-release formulations of MPH (OROS-MPH) versus
ATX; (c) subgroup analysis of MPH (applied Attention-Deficit Hyperactivity Disorder Rating Scale, ADHD–RS–IV, criteria)
versus ATX. CI = confidence interval; M-H: Mantel Haenszel.
Table 2. Summary of outcomes.
Heterogeneity Effect size
No. of patients Model RR/
Outcomes Studies (MPH/ATX) p I2 (%) (F/R) SMD 95% CI p Egger’s test, p
Efficacy
Response rate 7 1525/1083 .29 17 F 1.16 [1.10, 1.22] <.01 .740
Subgroup ADHD–RS–IV 5 1066/613 .34 12 F 1.17 [1.10, 1.23] <.01 —
Subgroup OROS-MPH 5 1328/885 .37 8 F 1.18 [1.11, 1.25] <.01 —
ADHD–RS–IV total 6 1251/1044 .0007 74 R 0.07 [−0.12, 0.26] .49 .763
ADHD–RS–IV 5 537/678 .59 0 F −0.00 [−0.12, 0.12] .98 .070
hyperactivity/impulsivity
ADHD–RS–IV inattention 5 537/678 .26 25 F −0.13 [−0.25, −0.01] .03 .085
Adverse events
Abdominal pain 8 1523/1230 .31 15 F 0.82 [0.60, 1.12] .22 .788
Anorexia 9 1584/1289 .15 34 F 1.07 [0.90, 1.28] .44 .192
Drowsiness (somnolence) 9 1584/1289 .44 0 F 0.17 [0.11, 0.26] <.01 .749
Headache 9 1584/1289 .36 9 F 0.89 [0.70, 1.12] .33 .062
Insomnia 9 1584/1289 .99 0 F 2.24 [1.60, 3.12] <.01 .675
Nausea 8 1523/1230 .05 51 R 0.49 [0.29, 0.85] .01 .698
Vomiting 7 1398/1172 .75 0 F 0.41 [0.27, 0.63] <.01 .851
Note. ADHD–RS–IV = attention-deficit hyperactivity disorder Rating Scale–IV; ATX = atomoxetine; MPH = methylphenidate; OROS-MPH = oral
osmotic methylphenidate; F = fixed effect model; R = random effect model; RR = risk ratio; SMD = standardized mean difference; CI =
confidence interval.
 JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY 861

Figure 4. Comparison of the Attention-Deficit Hyperactivity Disorder Rating Scale (ADHD–RS) score between methylphe-
nidate (MPH) and atomoxetine (ATX) agents. (a) Total ADHD–RS–IV score of MPH versus ATX; (b) hyperactivity/impulsivity
of MPH versus ATX; (c) inattention of MPH versus ATX. CI = confidence interval.

Discussion remarkably greater than that of ATX (Childress &

Sallee, 2014). Similar to this finding, our present
Our meta-analysis included a total of 11 studies to
meta-analysis indicated that MPH, especially OROS-
compare efficiency between MPH and ATX for
MPH or MPH applied ADHD–RS–IV criteria,
ADHD treatment in children and adolescents. As a
showed a significantly higher response rate than
result, we found that MPH showed a significantly
ATX, favoring the superiority of MPH over ATX. By
increased response rate and reduced inattention for
further comparing our result to that of the previous
ADHD child patients compared to ATX. Moreover,
study, we found that most of our included studies
MPH showed a pronouncedly lower risk of drowsi-
were conducted in Asian countries (e.g., China,
ness (somnolence), nausea, and vomiting, but a
India, and Turkey), and patients in that previous
higher risk of insomnia, than ATX.
study were South Korean. Therefore, the advanta-
MPH is the most frequently used drug for ADHD
geous effect of MPH might be more applicable in
management. Reportedly, the response rate of MPH is
Asian populations. However, subgroup analysis stra-
higher than that of other psychiatric medications
tified by different populations should be considered.
(Kim, Sharma, & Ryan, 2015). Both of MPH and
With regard to the improvement on attentional
ATX have the mechanism to block DA transporter,
functions in children with ADHD, it is demonstrated
which could lead to the increase of synaptic catecho-
that OROS-MPH has pronouncedly improved the
lamines. Notably, the stimulants could stimulate addi-
scores of the neuropsychological executive function
tional secretion of monoamines from presynaptic
tests such as working memory and set shifting (Yang
neuron, and this might contribute to the high
et al., 2012). An event-related potential study finds
response rate of MPH in ADHD patients. A study
that MPH is more effective than ATX in the reduction
indicates that response rate of OROS-MPH is
of reaction time (Kratz et al., 2012), which might
862 Q. LIU ET AL.

Figure 5. Comparison of adverse events between methylphenidate (MPH) and atomoxetine (ATX) agents. (a) Drowsiness
of MPH versus ATX; (b) nausea of MPH versus ATX; (c) vomiting of MPH versus ATX; (d) insomnia of MPH versus ATX; M-H:
Mantel Haenszel.

contribute to the improvement of inattention. Our
combined results revealed that MPH achieved a sig-
nificantly lower inattention score than ATX, further
supporting the superiority of MPH over ATX.
Regarding the side effects, a study reports that ATX
is more likely to cause side effects such as nausea and
vomiting than IR-MPH (Cortese et al., 2013). The
plausible reason for lower risk of side effects of MPH
is the specific pharmacological property of MPH,
which could active the brainstem arousal system, cor-
tex, and subcortical structures, and could enhance the
dopaminergic firing, and this might facilitate the
 JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY
improvement of inattention. In our meta-analysis,
MPH showed a significantly lower risk of side effects,
such as drowsiness, nausea, and vomiting, than ATX,
suggesting that MPH was a better choice between the
two agents. However, our findings also indicated that
the risk of insomnia after MPH administration was
higher than that for ATX. Other investigators have
also confirmed that insomnia is more frequent in
patients receiving MPH than in those receiving ATX
(Rostain, 2007).
Although several meta-analyses have been pub-
lished on this subject, and some of them presented
the same result as that supporting the greater
effectiveness of MPH than ATX, our study has
more advantages than theirs. First, we selected
the eligible studies based on more rigorous criteria
that only the RCTs comparing ATX with MPH (or
OROS-MPH) were included. Additionally, we did
not include the study comparing ATX with the
standard current therapy (that only a portion of
patients applied MPH; Prasad et al., 2007), or the
cross-over study (Sangal, Owens, & Sangal, 2005),
which were both included in the previous two
meta-analyses (Hanwella et al., 2011). Moreover,
although discontinuation rate, which is correlated
with efficiency and toxicity of the medication
(Leucht et al., 2013), has been considered in some
of the previous meta-analyses (C. Bushe et al.,
2016), none of them evaluates the side effects.
These collectively suggest that our results are
more reliable.
Despite these advantages, there are also several
limitations. First, significant heterogeneity was dis-
covered for several outcomes, which might bring
in deviation of the combined results. The above
heterogeneity might be caused by different coun-
tries, patients under different conditions, and dif-
ferent study designs, and subgroup analyses
stratified by these factors should be further per-
formed. Second, the case numbers in MPH and
ATX were disproportionate in a few studies, which
might not provide exact data. Third, most of the
included studies were open-label but not blind.
Although they were RCTs, this design might
cause bias to some extent. Fourth, the follow-up
durations in these included studies were not the
same, but were varied from 2 to 12 weeks
(Table 1). The different follow-up time might
influence the precise efficiency of MPH and ATX.
Moreover, ATX has a long duration action, but we
noticed that almost all the studies had a small
duration less than 3 months, which might
863
prejudice the ATX’s function. Fifth, dose is an
important factor for the medicine evaluation
because the drug efficiency and adverse events
might be dose dependent. However, we did not
perform the subgroup analysis stratified by differ-
ent dose levels due to unavailable dose information
in several studies. In addition, patient conditions
were not the same in each study: Most of them
were mixed with naïve patients and the patients
who have previously received the treatment, while
a study enrolled the drug-naïve patients. Likewise,
we did not perform the subgroup analysis stratified
by patients with different medication treatments
since they could not be obtained from each study.
This would undoubtedly bring several deviations
in this meta-analysis. Last but not least, the sample
size was relatively small, so that more RCTs with
larger samples are required to support our
findings.
In conclusion, MPH is more effective than ATX
for ADHD in children and adolescents, with
higher response rate and reduced inattention, as
well as the lower risk of side effects except insom-
nia. However, more RCTs with larger samples are
warranted to confirm our findings.
Disclosure statement
No potential conflict of interest was reported by the
authors.
References
Boleaalamañac, B., Nutt, D. J., Adamou, M., Asherson,
P., Bazire, S., Coghill, D., . . . Santosh, P. (2014).
Evidence-based guidelines for the pharmacological
management of attention deficit hyperactivity disor-
der: Update on recommendations from the British
Association for Psychopharmacology. Journal of
Psychopharmacology, 28(3), 179–203. doi:10.1177/
0269881113519509
Bushe, C., Day, K., Reed, V., Karlsdotter, K., Berggren,
L., Pitcher, A., & Haynes, V. (2016). A network meta-
analysis of atomoxetine and osmotic release oral sys-
tem methylphenidate in the treatment of attention-
deficit/hyperactivity disorder in adult patients.
Journal of Psychopharmacology, 30(5), 444–458.
doi:10.1177/0269881116636105
Bushe, C. J., & Savill, N. C. (2013). Suicide related events
and attention deficit hyperactivity disorder treat-
ments in children and adolescents: A meta-analysis
of atomoxetine and methylphenidate comparator
clinical trials. Child and Adolescent Psychiatry and
Mental Health, 7(1), 19. doi:10.1186/1753-2000-7-19
864 Q. LIU ET AL.
Cetin, F. H., Torun, Y. T., & Taner, Y. I. (2015).
Atomoxetine versus oros methylphenidate in atten-
tion deficit hyperactivity disorder: A six-month fol-
low up study for efficacy and adverse effects. Turkiye
Klinikleri Journal of Medical Sciences, 35(2), 88–96.
doi:10.5336/medsci.2015-43336
Childress, A. C., & Sallee, F. R. (2014). Attention-deficit/
hyperactivity disorder with inadequate response to
stimulants: Approaches to management. CNS Drugs,
28(2), 121–129. doi:10.1007/s40263-013-0130-6
Cortese, S., Holtmann, M., Banaschewski, T., Buitelaar,
J., Coghill, D., Danckaerts, M., & Sergeant, J. (2013).
Practitioner review: Current best practice in the man-
agement of adverse events during treatment with
ADHD medications in children and adolescents.
Journal of Child Psychology and Psychiatry, 54(3),
227–246. doi:10.1111/jcpp.12036
Dresel, S., Krause, J., Krause, K.-H., LaFougere, C.,
Brinkbäumer, K., Kung, H. F., . . . Tatsch, K. (2000).
Attention deficit hyperactivity disorder: Binding of
[99mTc] TRODAT-1 to the dopamine transporter
before and after methylphenidate treatment.
European Journal of Nuclear Medicine, 27(10), 1518–
1524. doi:10.1007/s002590000330
Garg, J., Arun, P., & Chavan, B. S. (2014). Comparative
short term efficacy and tolerability of methylpheni-
date and atomoxetine in attention deficit hyperactiv-
ity disorder. Indian Journal of Pediatrics, 51(7), 550–
554. doi:10.1007/s13312-014-0445-5
Hanwella, R., Senanayake, M., & de Silva, V. (2011).
Comparative efficacy and acceptability of methylphe-
nidate and atomoxetine in treatment of attention
deficit hyperactivity disorder in children and adoles-
cents: A meta-analysis. BMC Psychiatry, 11(1), 176.
doi:10.1186/1471-244X-11-176
Hazell, P. L., Kohn, M. R., Dickson, R., Walton, R. J.,
Granger, R. E., & Wyk, G. W. (2011). Core ADHD
symptom improvement with atomoxetine versus
methylphenidate: A direct comparison meta-analysis.
Journal of Attention Disorders, 15(8), 674–683.
doi:10.1177/1087054710379737
Heil, S. H., Holmes, H. W., Bickel, W. K., Higgins, S. T.,
Badger, G. J., Laws, H. F., & Faries, D. E. (2002).
Comparison of the subject-rated, physiolog-ical and
psychomotor effects of atomoxetine and methyl-phe-
nidate in recreational drug users. Drug and Alcohol
Dependence, 67(2), 149–156. doi:10.1016/S0376-8716
(02)00053-4
Higgins, J., Thompson, S. G., Deeks, J. J., & Altman, D.
G. (2003). Measuring inconsistency in meta-analyses.
British Medical Journal, 327(7414), 557–560.
doi:10.1136/bmj.327.7414.557
Higgins, J. P. T., & Green, S. (2008). Cochrane handbook
for systematic reviews of interventions (Vol. 5). Wiley
Online Library: Hoboken, New Jersey, USA.
Kemner, J. E., Starr, H. L., Ciccone, P. E., Hooper-
Wood, C. G., & Crockett, R. S. (2005). Outcomes of
OROS methylphenidate compared with atomoxetine
in children with ADHD: A multicenter, randomized
prospective study. Advances in Therapy, 22(5), 498–
512. doi:10.1007/BF02849870
Kim, J.-W., Sharma, V., & Ryan, N. D. (2015).
Predicting methylphenidate response in ADHD
using machine learning approaches. International
Journal of Neuropsychopharmacology, 18(11), 1–7.
doi:10.1093/ijnp/pyv052
Kratochvil, C. J., Heiligenstein, J. H., Dittmann, R.,
Spencer, T. J., Biederman, J., Wernicke, J., . . .
Michelson, D. (2002). Atomoxetine and methylphe-
nidate treatment in children with ADHD: A prospec-
tive, randomized, open-label trial. Journal of the
American Academy of Child and Adolescent
Psychiatry, 41(7), 776–784. doi:10.1097/00004583-
200207000-00008
Kratz, O., Studer, P., Baack, J., Malcherek, S., Erbe, K., Moll,
G. H., & Heinrich, H. (2012). Differential effects of
methylphenidate and atomoxetine on attentional pro-
cesses in children with ADHD: An event-related poten-
tial study using the attention network test. Progress in
Neuro-Psychopharmacology and Biological Psychiatry, 37
(1), 81–89. doi:10.1016/j.pnpbp.2011.12.008
Leucht, S., Cipriani, A., Spineli, L., Mavridis, D., Örey,
D., Richter, F., & Geddes, J. R. (2013). Comparative
efficacy and tolerability of 15 antipsychotic drugs in
schizophrenia: A multiple-treatments meta-analysis.
The Lancet, 382(9896), 951–962. doi:10.1016/S0140-
6736(13)60733-3
Madhusoodanan, S., & Goia, D. (2016). Rapid resolution
of depressive symptoms with methylphenidate aug-
mentation of mirtazapine in an elderly depressed
hospitalized patient: A case report. Clinical Practice,
11(11), 283–288. doi:10.2217/cpr.14.32
Newcorn, J., Kratochvil, C., Allen, A., Casat, C., Ruff, D.,
Moore, R., & Michelson, D. (2008). Atomoxetine and
osmotically released methylphenidate for the treat-
ment of attention deficit hyperactivity disorder:
Acute comparison and differential response.
American Journal of Psychiatry, 165(6), 721–730.
doi:10.1176/appi.ajp.2007.05091676
Peterson, K., Mcdonagh, M. S., & Fu, R. (2008).
Comparative benefits and harms of competing med-
ications for adults with attention-deficit hyperactivity
disorder: A systematic review and indirect compari-
son meta-analysis. Psychopharmacology, 197(197), 1–
11. doi:10.1007/s00213-007-0996-4
Polanczyk, G. V., Salum, G. A., Sugaya, L. S., Caye, A., &
Rohde, L. A. (2015). Annual research review: A meta-
analysis of the worldwide prevalence of mental dis-
orders in children and adolescents. Journal of Child
Psychology & Psychiatry, 56(3), 345–365. doi:10.1111/
jcpp.12381
Poling, A., Methot, LL., Lesage MG. Diagnostic and
Statistical Manual of Mental Disorders (DSM-IV).
(2013). American Psychiatric Association: Virginia,
USA.
Prasad, S., Harpin, V., Poole, L., Zeitlin, H., Jamdar, S.,
& Puvanendran, K. (2007). A multi-centre, rando-
mised, open-label study of atomoxetine compared
with standard current therapy in UK children and
adolescents with attention-deficit/hyperactivity disor-
der (ADHD). Current Medical Research and Opinion,
23(2), 379–394. doi:10.1185/030079906X167309
 JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY
Roskell, N. S., Setyawan, J., Zimovetz, E. A., &
Hodgkins, P. (2014). Systematic evidence synthesis
of treatments for ADHD in children and adolescents:
Indirect treatment comparisons of lisdexamfetamine
with methylphenidate and atomoxetine. Current
Medical Research & Opinion, 30(8), 1673–1685.
doi:10.1185/03007995.2014.904772
Rostain, A. L. (2007). Sleep disturbances and ADHD
medications. Current Psychiatry Reports, 9(5), 399–
400. doi:10.1007/s11920-007-0051-5
Sangal, R. B., Owens, J. A., & Sangal, J. (2005). Patients
with attention-deficit/hyperactivity disorder without
observed apneic episodes in sleep or daytime sleepi-
ness have normal sleep on polysomnography. Sleep,
28(9), 1143–1148.
Schulz, K. P., Fan, J., Bedard, A. C., Clerkin, S. M., Ivanov,
I., Tang, C. Y., . . . Newcorn, J. H. (2012). Common and
unique therapeutic mechanisms of stimulant and non-
stimulant treatments for attention-deficit/hyperactivity
disorder. Archives of General Psychiatry, 69(9), 952–
961. doi:10.1001/archgenpsychiatry.2011.2053
Shang, C. Y., Pan, Y. L., Lin, H. Y., Huang, L. W., &
Gau, S. S. (2015). An open-label, randomized trial of
methylphenidate and atomoxetine treatment in chil-
dren with attention-deficit/hyperactivity disorder.
Journal of Child and Adolescent Psychopharmacology,
25(7), 566–573. doi:10.1089/cap.2015.0035
Shier, A. C., Reichenbacher, T., Ghuman, H. S., &
Ghuman, J. K. (2013). Pharmacological treatment of
attention deficit hyperactivity disorder in children
and adolescents: Clinical strategies. Journal of
Central Nervous System Disease, 5(5), 1–17.
Smith, S. D., Vitulano, L. A., Katsovich, L., Li, S., Moore,
C., Li, F., & Aktan, G. S. (2016). A randomized con-
trolled trial of an integrated brain, body, and social
intervention for children with ADHD. Journal of
Attention Disorders, 1087054716647490.
So, Y. K., Noh, J. S., Kim, Y. S., Ko, S. G., & Koh, Y. J.
(2002). The reliability and validity of Korean parent
and teacher ADHD rating scale. Journal of Korean
Neuropsychiatric Association, 41(2), 283–289.
Starr, H. L., & Kemner, J. (2005). Multicenter, rando-
mized, open-label study of OROS methylphenidate
versus atomoxetine: treatment outcomes in African-
American children with ADHD. Journal of the
national medical association, 97(10 Suppl), 11S–16S.
Štuhec, M. (2013). Duloxetine-induced life-threatening
long QT syndrome. Irish Journal of Medical Science,
182(3), 535–537. doi:10.1007/s11845-013-0925-y
Stuhec, M., Munda, B., Svab, V., & Locatelli, I. (2015).
Comparative efficacy and acceptability of atomoxetine,
lisdexamfetamine, bupropion and methylphenidate in
865
treatment of attention deficit hyperactivity disorder in
children and adolescents: A meta-analysis with focus
on bupropion. Journal of Affective Disorders, 178(2),
149–159. doi:10.1016/j.jad.2015.03.006
Štuhec, M., Švab, V., & Locatelli, I. (2015). Prevalence
and incidence of attention-deficit/hyperactivity disor-
der in Slovenian children and adolescents: A database
study from a national perspective. Croatian
Medical Journal, 56(2), 159–165. doi:10.3325/
cmj.2015.56.159
Su, Y., Yang, L., Stein, M. A., Cao, Q., & Wang, Y.
(2016). Osmotic release oral system methylphenidate
versus atomoxetine for the treatment of attention-
deficit/hyperactivity disorder in Chinese Youth: 8-
week comparative efficacy and 1-year follow-up.
Journal of Child and Adolescent Psychopharmacology,
26(4), 362–371. doi:10.1089/cap.2015.0031
Wang, Y., Zheng, Y., Du, Y., Song, D. H., Shin, Y.-J.,
Cho, S. C., & Gao, H. (2007). Atomoxetine versus
methylphenidate in paediatric outpatients with atten-
tion deficit hyperactivity disorder: A randomized,
double-blind comparison trial. Australasian
Psychiatry, 41(3), 222–230.
Wigal, S. B. (2009). Efficacy and safety limitations of
attention-deficit hyperactivity disorder pharmacother-
apy in children and adults. CNS Drugs, 23(Supplement
1), 21–31. doi:10.2165/00023210-200923000-00004
Wilens, T., Pelham, W., Stein, M., Conners, C. K.,
Abikoff, H., Atkins, M., & Palumbo, D. (2003).
ADHD treatment with once-daily OROS methylphe-
nidate: Interim 12-month results from a long-term
open-label study. Journal of the American Academy of
Child & Adolescent Psychiatry, 42(4), 424–433.
doi:10.1097/01.CHI.0000046814.95464.7D
Wolraich, M., Brown, L., Brown, R. T., Dupaul, G., Earls,
M., Feldman, H. M., . . . Perrin, J. (2011). ADHD:
Clinical practice guideline for the diagnosis, evaluation,
and treatment of attention-deficit/hyperactivity disor-
der in children and adolescents. Pediatrics, 128(5),
1007–1022. doi:10.1542/peds.2011-2654
Yang, L., Cao, Q., Shuai, L., Li, H., Chan, R. C. K., &
Wang, Y. (2012). Comparative study of OROS-MPH
and atomoxetine on executive function improvement
in ADHD: A randomized controlled trial. The
International Journal of Neuropsychopharmacology,
15(1), 15–26. doi:10.1017/S1461145711001490
Yildiz, O., Sismanlar, S. G., Memik, N. C., Karakaya,
I., & Agaoglu, B. (2011). Atomoxetine and methyl-
phenidate treatment in children with ADHD: The
efficacy, tolerability and effects on executive func-
tions. Child Psychiatry & Human Development, 42
(3), 257–269. doi:10.1007/s10578-010-0212-3