Annals of Pharmacotherapy
© The Author(s) 2018
Article reuse guidelines:
Attention-deficit hyperactivity disorder (ADHD) is a neuro-
developmental disorder characterized by the core symp-
toms of inattention, hyperactivity, and impulsivity.1 Its
worldwide prevalence in children and adolescents is
between 2.5% and 10% and less in adults.2,3 According to
the literature, 30% to 50% of children diagnosed in child-
hood continue to have symptoms into adulthood, which
means that many ADHD cases warrant treatment into adult-
hood with effective pharmacotherapy.4 ADHD is associated
with significantly increased mortality rates as was shown
by the Danish national study, where the authors found that
people diagnosed with ADHD in adulthood had a higher
mortality rate ratio.5
According to the newest British guidelines of British
Association for Psychopharmacology, the management of
ADHD consists of nonpharmacological and pharmacological
795703AOPXXX10.1177/1060028018795703Annals of PharmacotherapyStuhec et al
1Ormoz Psychiatric Hospital, Department of clinical pharmacy, Ormoz,
Slovenia, European Union
2University of Ljubljana, Faculty of Pharmacy, Slovenia, European Union
Matej Stuhec, Department of Clinical Pharmacy, Ormoz Psychiatric
Hospital, Ptujska cesta 33, SI-2270 Ormoz, Slovenia, European Union.
Efficacy, Acceptability, and Tolerability of
Lisdexamfetamine, Mixed Amphetamine
Salts, Methylphenidate, and Modafinil
in the Treatment of Attention-Deficit
Hyperactivity Disorder in Adults: A
Systematic Review and Meta-analysis
Matej Stuhec, MPharm, PhD1,2, Petar Lukić, MPharm2,
and Igor Locatelli, MPharm, PhD2
Objective: Psychostimulants are the first-line treatment in adults with attention-deficit hyperactivity disorder (ADHD).
This meta-analysis aimed to evaluate the efficacy, acceptability, and tolerability of lisdexamfetamine (LDX), mixed
amphetamine salts (MASs), modafinil (MDF), and methylphenidate (MPH) in comparison with placebo. Data Sources: We
systematically searched PubMed/MEDLINE and Clinicaltrials.gov in May 2016, along with CENTRAL and EU Clinical Trials
Register in February 2016, for the randomized, double-blind, placebo-controlled, parallel-group clinical trials conducted
on adults diagnosed with ADHD. Study Selection and Data Extraction: Substantial comorbidity, substance abuse or
dependence, and nonpharmacological interventions represented grounds for exclusion. Published reports were the sole
source for data extraction. Improvement in ADHD symptoms was the primary outcome. Random-effects model meta-
analysis was applied to calculate the standardized mean difference (SMD) with 95% CIs. Data Synthesis: The search
retrieved 701 records, of which 20 studies were eligible for analysis. High effect size (expressed as SMD) in reducing
ADHD symptoms was observed for LDX (−0.89; 95% CI = −1.09, −0.70), whereas MASs (−0.64; 95% CI = −0.83, −0.45)
and MPH (−0.50; 95% CI = −0.58, −0.41) reduced symptoms moderately compared with placebo. No efficacy was shown
for MDF (0.08; 95% CI; −0.18, 0.34). Relevance to Patient Care and Clinical Practice: In this meta-analysis, the
efficacy, tolerability, and acceptability of psychostimulants were compared with that for placebo. Five of the included trials
have not been evaluated in any of the previously published meta-analyses. Conclusions: The results suggest that LDX has
the largest effect size and has a promising potential for treating adults with ADHD.
ADHD, adults, psychostimulants, efficacy, tolerability, meta-analysis.
2 Annals of Pharmacotherapy 00(0)
options, including stimulants and nonstimulants. Although psy-
chotherapy (eg, cognitive behavioral therapy) is a first-line treat-
ment in children and adolescents with ADHD, pharmacotherapy
is the first-line treatment in adults with ADHD.6 According to
evidence-based pharmacotherapy, the effect sizes calculated
from meta-analyses are one of the most important outcomes,
which are clinically applicable to distinguish drug effective-
ness between 2 different medications. According to the guide-
lines and well-designed meta-analyses, stimulants are the
first-line treatment for ADHD in children, adolescents, and
adults, with their medium to high effect sizes, which have been
already calculated in many meta-analyses.1,6-9
Although large effect sizes of stimulants for ADHD
treatment were calculated in meta-analyses, most medica-
tions have not been studied in older adults (e.g., older than
50 years). Another significant limitation is a lack of meta-
analyses in adults with ADHD with the inclusion of the
According to the published meta-analyses in children
and adolescents, lisdexamfetamine (LDX) was the most
effective treatment strategy for ADHD with its large effect
size represented as standardized mean difference (SMD)
>1.0.1,6-9 However, there are no results on comparative effi-
cacy among different stimulants, including LDX, with the
latest randomized controlled trials (RCTs) until 2015 in
adults. In the literature, there are many different meta-anal-
yses available in adults,10-20 although many of them included
only methylphenidate (MPH) immediate release form and
similar forms of MPH excluding LDX and modafinil
(MDF).11,13,16,17,20 Also, there are only 4 meta-analyses
available that included LDX and compared it with other
stimulants or/and only placebo in adults.10,14,18,19 Meta-
analyses published by Mészáros et al19 and Faraone14
included only 1 RCT with LDX, which had a significant
impact on heterogeneity, and the results should be repli-
cated with the addition of the latest LDX RCTs. Meta-
analyses published by Castells et al10 included only
amphetamines with 7 RCTs, excluding MPH and MDF.
Meta-analyses published by Maneeton et al18 included only
LDX, with its 5 trials excluding MDF. Among these 4 meta-
analyses, the meta-analyses published by Mészáros et al,19
Faraone,14 and Castells et al10 were released in 2012 or
before and, therefore, with small numbers of LDX RCTs
available included in the meta-analysis.
The primary aim was to address efficacy (improvement
in ADHD symptoms, clinical global impression [CGI], and
executive function [EF] deficit), acceptability, and tolera-
bility of psychostimulants used in the treatment of ADHD
in adults using a meta-analysis. Additionally, we aimed to
upgrade previous meta-analyses with most recent RTCs
evaluating the efficacy and safety of LDX, mixed amphet-
amine salts (MASs), and MPH in different forms and to
include MDF in the comparison.
In the current systematic review with meta-analysis, we
applied a structured PICOS principle in defining the inclu-
sion/exclusion criteria as advised by the PRISMA guide-
lines.21 Only randomized, double-blind, placebo-controlled,
and parallel-group clinical trials that measured ADHD
symptoms were included in the systematic review. We set
the minimal duration of the studies at 2 weeks and a mini-
mum age of the included participants at 17 years. The diag-
nosis of ADHD had to be established in childhood or newly
discovered in adulthood and again confirmed with a vali-
dated diagnostic instrument before allocation to an inter-
Interventions of interest were stimulants of the central
nervous system, MASs, LDX, MPH, and MDF, applied
orally irrespective of the release form, dosing regimen, or
dosing method (fixed or titration). Studies in which one of
the interventions did not include placebo, or a nonpharma-
cological intervention was applied and compared with a
drug, were excluded. Both treatment-naïve patients and
those previously treated for ADHD were included.
We identified studies by using freely accessible databases:
Cochrane Central Register of Controlled Trials (last search:
May 6, 2016), PubMed/MEDLINE (last search: 1946 to
February 9, 2016), ClinicalTrials.gov (last search: February
9, 2016), and EU Clinical Trials Register (last search: May
6, 2016). We repeated searches several times between
December 17, 2014, and May 6, 2016. The search terms we
used in all our searches were as follows: (attention deficit
disorder with hyperactivity OR ADHD) AND adult AND
(methylphenidate OR dextroamphetamine OR amphet-
amine salts OR lisdexamfetamine OR atomoxetine OR
bupropion OR desipramine OR modafinil OR guanfacine
OR clonidine OR venlafaxine OR reboxetine OR dulox-
etine OR galanthamine) AND (double blind) AND
(Randomized Controlled Trial). Expansion of the search
terms to nonstimulants was aimed to identify any multiple
drug trials involving stimulants and nonstimulants that
may have not otherwise been detected. Moreover, the iden-
tified citations reflect the research activity within this field.
We did not apply restrictions regarding date, language, or
publishing status but excluded conference abstracts that
were not published as journal articles. We searched refer-
ence lists in relevant systematic reviews with meta-analy-
ses as well as those in hard copies of the studies assessed
for eligibility for any records not identified in the elec-
tronic databases. We did not contact study authors to iden-
tify any additional studies.
Stuhec et al 3
Data Extraction and Study Outcomes
Two reviewers (PL and IL) independently carried out data-
base search, reference list inspection, eligibility assessment,
and data collection according to predefined selection crite-
ria. Any disagreements were resolved by discussion or con-
sultation with the third reviewer (MS). Data extraction and
processing were performed in pen and stored electronically.
The data of interest were as follows: first author, publica-
tion year, drug(s) tested, study design, inclusion/exclusion
criteria, number of randomized participants, proportion of
previously treated participants, number of participants in
each study arm, mean age, mean age at initial diagnosis,
proportion of male participants, dosage regimen, mean
daily dose, dosing method (fixed, titration), study duration,
ADHD rating scale, primary outcomes, secondary out-
comes, and safety outcomes.
The outcomes of interest were decided on according to
their clinical relevance and applicability. The primary out-
come was the improvement in ADHD symptoms assessed
with a validated instrument for measuring current symp-
toms. Investigator rating scales were the preferred method
of symptoms assessment. We used self-reported scales
where the clinical scales were not applied or data unavail-
able. Secondary efficacy outcomes were CGI improvement
(CGI-I) and EF deficit improvement assessed with a vali-
dated instrument. The outcome retention in treatment
reflected the acceptability of the intervention and was
defined as a proportion of patients who completed the trial.
The outcome discontinuation as a result of adverse events
was a safety outcome reflecting tolerability of the interven-
tion and was defined as a proportion of the patients who
reported the adverse event as a cause of their dropping out
before the end point.
We collected mean changes (mean difference) from
baseline to end point with their SDs for continuous out-
comes (improvements in ADHD symptoms and EF deficit)
and the number of participants with events against the num-
ber of participants in each study arm for binary outcomes
(CGI, retention in treatment, discontinuation as a result of
an adverse event).
Assessment of the Risk of Bias
Two reviewers (PL and IL) independently assessed the risk
of bias utilizing the Cochrane Collaboration’s tool criteria.22
Differences were resolved through consensus. The tool
addresses 7 specific domains of bias—namely, sequence
generation, allocation concealment, blinding of participants
and personnel, blinding of outcome assessment, incomplete
outcome data, selective outcome reporting, and other
sources of bias. Descriptions of the methods and procedures
reported by the study authors provided support in assigning
the judgment relating to the risk of bias as “low risk,” “high
risk,” or “unclear risk.” The risk of bias was assessed at the
study level, except for the domains incomplete outcome
data and selective outcome reporting where only the pri-
mary outcome (improvement in ADHD symptoms) was
Statistical package Review Manager (RevMan) version
5.3 was used to calculate effect measures and data synthe-
sis.22 The treatment effect measure for continuous out-
comes is expressed as the SMD and presented as a point
estimate with 95% CI. Quantitative synthesis was con-
ducted by applying DerSimonian and Laird’s model of
random effects, as suggested by marked heterogeneity
observed in earlier reviews. The inverse variance method
was used for continuous variables (SMD), whereas the
Mantel-Haenszel method was used for binary outcomes
for which the treatment effect measure was expressed as
risk ratio (RR).22
Sensitivity and Subgroup Analysis
The 2 sensitivity analyses replicated the meta-analysis by
exclusion of studies assigned with 1 or more high risk of
bias judgements, and by expression of the treatment effect
in terms of the mean difference where symptom improve-
ment was assessed by an identical instrument. An attempt
was made at finding possible sources of heterogeneity by
performing the following subgroup analyses for primary
efficacy outcome only. The first juxtaposed the effect sizes
of 2 dosing methods—titration and fixed—and the second
compared 3 different release forms of MPH.
The search returned a total of 701 references of which 652
were excluded after study titles and abstracts were screened;
hence, 49 potentially relevant full-text articles were
acquired. Following an eligibility assessment, 20 studies
were included in the systematic review and 19 in the meta-
analysis. Figure 1 shows a detailed flowchart of the study
Two studies investigated MASs,23,24 3 LDX,25-27 and
14 MPH,28-41 whereas 1 trial focused on MDF.42 Five of the
included studies have not been found in the preceding rele-
vant meta-analyses.32,33,40-42 Table 1 summarizes the main
characteristics of the studies eligible for the meta-analysis.
A comprehensive summary table of the study reports is
noted in online Appendix 1 (Table S1; available online at
The primary efficacy outcome was an improvement in
ADHD symptoms. High effect size (expressed as SMD) in
reducing ADHD symptoms was observed for LDX (−0.89;
4 Annals of Pharmacotherapy 00(0)
95% CI = −1.09, −0.70; I2 = 0%), whereas MASs (−0.64;
95% CI = −0.83, −0.45; I2 = 0%) and MPH (−0.50; 95%
CI = −0.58, −0.4; I2 = 21%) reduced symptoms moder-
ately compared with placebo. No efficacy was shown for
MDF (0.08, 95% CI; −0.18, 0.34; 1 study). We excluded 1
study of MPH from the meta-analysis because summary
data could not be obtained.37 Figure 2 shows a forest plot of
study results related to improvement in ADHD symptoms.
An improvement in the CGI was a dichotomized vari-
able defined as grade ⩽2 or lower on the CGI-I scale. In 5
studies, however, this criterion was coupled with a reduc-
tion in scale scores by ⩾30%.25,27,29,30,38 Two studies of
Figure 1. Study selection flowchart.
Stuhec et al 5
MASs, 3 of LDX, and 12 of MPH were combined in this
meta-analysis, whereas 2 studies of MPH did not assess this
outcome.35,41 Compared with placebo, psychostimulants
appear more likely to improve global impression of ADHD
(expressed as RR)—MASs (2.27; 95% CI = 1.71, 3.00;
I2 = 0%), LDX (2.23; 95% CI = 1.76, 2.83; I2 = 0%), and
MPH (1.61; 95% CI = 1.41, 1.84; I2 = 54%)—whereas the
study of MDF did not demonstrate advantage over placebo
(0.93; 95% CI = 0.65, 1.32). Figure 3 shows a forest plot of
study results related to improvement in the CGI.
Only 1 trial per drug (altogether 4 studies) was observed
for evaluation of improvement in EF deficit; thus, the meta-
analysis was not applicable for this secondary outcome.
Stimulants demonstrated advantage over placebo (expressed
as SMD): MASs (−0.65; 95% CI = −0.90, −0.40),23 LDX
(−0.75; 95% CI = −1.08, −0.42),26 and MPH (−0.34; 95%
CI = −0.56, −0.13),32 except for MDF (−0.07; 95%
CI = −0.33, 0.19).42
The probability of retention in treatment (acceptability)
was statistically significantly higher with MASs (1.18; 95%
CI = 1.03, 1.35; I2 = 0%) than with placebo (expressed as
RR), whereas difference was insignificantly favourable for
LDX (1.04: 95% CI = 0.93, 1.15; I2 = 27%) and MPH
(0.97; 95% CI = 0.93, 1.01; I2 = 54%). The study of MDF
showed no advantage over placebo (0.78; 95% CI = 0.65,
0.94, 1 study). Figure 4 shows the corresponding forest plot.
Treatment tolerability was assessed in all eligible studies
(Figure 5). Statistically significantly higher probability
(expressed as RR) of discontinuation caused by treatment-
emergent adverse events compared with placebo was
Table 1. Summary of Study Characteristics.
Study Size (n)
Male) Rating Scale Used
Method Dose Range (mg)
Spencer et al, 200823 274/24.1 36.5/31 50.0 ADHD DSM-IV-TR Titration 12.5-75 1×/d 7
Weisler et al, 200624 255/22.0 39.2/33.8 60.1 ADHD DSM-IV-TR Fixed 20-60 1×/d 4
Adler et al, 200825 420/9.8 or 17.9 35.1/NDA 54.3 ADHD DSM-IV-TR Fixed 30-70 1×/d 4
Adler et al, 201326 161/NDA 34.6/5.4 52.2 ADHD DSM-IV-TR Titration 30-70 1×/d 10
Biederman et al, 201227 69/43.5 22.2/NDA 56.5 ADHD DSM-IV-TR Titration 30-70 1×/d 6
Adler et al, 200928 229/34.9 39.0/NDA 56.2 AISRS Titration 36-108 1×/d 7
Biederman et al, 200629 149/NDA 35.2/4.8 51.8 AISRS Titration 36-108 1×/d 6
Biederman et al, 201030 227/NDA 35.6/NDA 45.7 AISRS Titration 36-144 1×/d 6
Casas et al, 201331 279/NDA 35.7/31.7 52.3 CAARS-O: SV Fixed 54-72 1×/d 13
Goodman et al, 201732 357/NDA 35.7/NDA 53.8 AISRS Titration 18-72 1×/d 6
Huss et al, 201433 725/13.2 35.4/NDA 54.5 ADHD DSM-IV Fixed 40-80 1×/d 9
Kuperman et al, 200134 37/NDA 32.3/NDA 70.0 UPD ADHD DSM-III Titration Up to 0.9 mg/kg/d
divided into 3×/d
Medori et al, 200835 402/NDA 34.0/29.9 54.4 CAARS-O: SV Fixed 18, 36, 72 1×/d 5
Retz et al, 201236 162/33.3 37.4/NDA 46.9 WRAADDS Titration 10-120 divided into
Rösler et al, 200937 363/37.7 34.5/5.8 49.6 WRAADDS Titration 10-60 divided into
Spencer et al, 200538 146/8.9 38.0/NDA 58.2 AISRS Titration Up to 1.3 mg/kg/d
divided into 3×/d
Spencer et al, 200739 221/36.2 38.7/6.5 57.5 ADHD DSM-IV Fixed 20-40 1×/d 5
Takahashi et al, 201440 284/13.4 33.8/31.9 48.9 CAARS-O: SV Titration 18-72 1×/d 8
Weisler et al, 201241 430/NDA 33.9/21.3 57.1 ADHD DSM-IV Fixed 54 1×/d 6
Arnold et al, 201442 338/37.0 39.3/NDA 60.0 AISRS Fixed 255-510 1×/d 9
Abbreviations: ADHD, attention-deficit hyperactivity disorder; ADHD DSM-IV, ADHD rating scale based on DSM-IV; ADHD DSM-IV-TR, ADHD
rating scale based on DSM-IV-TR; AISRS, ADHD Adult Investigator Symptom Rating Scale; CAARS-O: SV, Conners’ Adult ADHD Rating Scale,
observer (clinician) report, screening form; DSM-III, Diagnostic and Statistical Manual of Mental Disorders, Third Edition; DSM-IV, Diagnostic and
Statistical Manual of Mental Disorders, Fourth Edition; DSM-IV-TR, Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision; LDX,
lisdexamfetamine; MASs, mixed amphetamine salts; MDF, modafinil; MPH, methylphenidate; NDA, no data available; UPD ADHD DSM-III, ADHD
rating scale based on DSM-III Rating Scale updated to DSM-IV; WRAADDS, Wender-Reimherr Adult-Attention-Deficit-Disorder Scale.
6 Annals of Pharmacotherapy 00(0)
observed for MASs (3.36; 95% CI = 1.48, 7.63; I2 = 0%),
MPH (2.63; 95% CI = 1.90, 3.64; I2 = 0%), and MDF
(3.21; 95% CI = 1.45, 7.09, 1 study); however, for LDX, a
borderline significance was observed (2.91; 95% CI =
0.98, 8.68; I2 = 0%, P = 0.055).
Significant heterogeneity was observed among the stud-
ies of MPH in the assessment of CGI and the retention in
treatment (acceptability). An I2 statistic of 54% was esti-
mated in both outcomes (Figures 3 and 4). Additionally, an
apparent weak overlapping was found in 2 studies,31,38
Figure 2. Mixed amphetamine salts, lisdexamfetamine, methylphenidate, and modafinil compared with placebo in attention-deficit
hyperactivity disorder symptom scale score reduction. Cochrane’s Risk of bias summary shows judgments for each study in each
domain of bias. For studies Biederman et al,29 2006, Biederman et al,30 2010, and Retz et al,36 2012, the end point scores were used
because mean changes from the baseline could not be obtained. (A) Random sequence generation (selection bias); (B) allocation
concealment (selection bias); (C) blinding of participants and personnel (performance bias); (D) blinding of outcome assessment
(detection bias); (E) incomplete outcome data (attrition bias); (F) selective reporting (reporting bias); (G) other bias. Low risk of bias
(plus), high risk of bias (minus), unclear risk of bias (question mark).
Stuhec et al 7
Figure 3. Mixed amphetamine salts, lisdexamfetamine, methylphenidate, and modafinil compared with placebo in illness improvement
on the clinical global impression scale.
8 Annals of Pharmacotherapy 00(0)
Figure 4. Mixed amphetamine salts, lisdexamfetamine, methylphenidate, and modafinil compared with placebo in overall retention in
Stuhec et al 9
Figure 5. Mixed amphetamine salts, lisdexamfetamine, methylphenidate, and modafinil compared with placebo in discontinuation
caused by treatment-emergent adverse events (tolerability).
10 Annals of Pharmacotherapy 00(0)
which also fell outside the 95% triangular confidence area
in the funnel plot. Although no heterogeneity was observed
for ADHD symptom improvement among the studies of
MPH (Figure 2; I2 = 21%), apparent poor overlapping
occurred in 2 studies and one of them was also an outlier in
the funnel plot.34,38 All funnel plots are presented in online
Appendix 2 (Figures S1 to S4).
A judgment of low risk of bias was assigned to 11 studies,
where a random component was described in the sequence
generation, and an adequate method was used to conceal
allocation. An unclear risk of bias judgment was assigned to
9 studies; 1 report provided insufficient information to per-
mit judgment of low risk or high risk, whereas 8 reports did
not describe or mention the randomization method used.
Blinding was intended in all included studies. However,
high risk of bias was assigned to a study with nonidentical
dosing regimens across the intervention groups together
with a single-blind lead-in phase.34 Low risk of attrition bias
was assigned to 9 studies with the drop-out rate <20% or
between 20% and 30%, provided the mixed-effect model for
repeated measures approach was applied. The discontinua-
tion rate above 30% or per-protocol efficacy analysis sup-
ported the judgment of high risk resulting from incomplete
outcome data in 7 studies.* Unclear risk resulting from the
disparity in data between 2 reports was found in 1 study,
whereas incompletely reported clinical scores were replaced
with self-report scale scores in the other. High risk of bias
was assigned to 4 studies that presented results only graphi-
cally. Because the mean changes were unavailable, we used
end point values where feasible.29,30,36,37 The pharmaceutical
industry funded all eligible studies. The risk of bias assess-
ment across studies is added to the forest plot presented in
A sensitivity analysis of the primary efficacy outcome
found that exclusion of the studies with the judgment of
high risk of bias did not result, in numerical terms at least,
in marked changes (data not shown). All studies of MASs
and LDX used the same symptom rating scale (ADHD
DSM-IV-TR, Table 1), and therefore, the mean difference
rating score was calculated on the same number of studies
as in the forest plot in Figure 2. Compared with placebo,
significant reduction of ADHD symptom scores was
observed for MASs (−7.69; 95% CI = −9.89, −5.49) and
LDX (−10.4; 95% CI = −12.6, −8.22).
The subgroup analysis was only applicable for studies
of MPH. The studies were split into 3 subgroups according
to the release form—namely, immediate release,34,38
extended release,33,36,37,39 and osmotic controlled oral
delivery system.28-32,35,40,41 No dissimilarity in efficacy
was observed between different release forms of MPH.
However, substantial heterogeneity in effect size within
the subgroups was found. Another analysis split the stud-
ies into 2 subgroups according to the dosing method used.
The 2 dosing methods did not show any advantage of one
over the other.
This article included RCTs of the available stimulants
(MPH, MASs, LDX, and MDF). A primary focus was on
how the treatment of interest compares with placebo,
regarding both efficacy and safety. This meta-analysis is the
first where MPH, MASs, LDX, and MDF were examined in
1 study. In addition to previous meta-analyses, our results
bring new data on comparative efficacy, acceptability, and
Results from our meta-analysis show that LDX is the
most effective treatment strategy, which is in line with pre-
viously published meta-analyses. In the meta-analysis pub-
lished by Castells et al,10 SMD for amphetamines was −0.72
(95% CI = −0.87, −0.57), but all the amphetamines were
combined in 1 group, including LDX and MASs.10 The sec-
ond meta-analysis published by Maneeton et al18 showed a
large effect size for LDX (−0.97; 95% CI = −1.15, −0.78).18
This meta-analysis includes new studies with LDX for
treatment of ADHD in adults. A prodrug form of LDX may
as well influence the monoamine release. However, a direct
comparison with other drugs has to be confirmed in head-
to-head RCTs in the adult population.15 Despite these limi-
tations and a lack of comparative RCTs, these results
demonstrate that LDX is a very effective medication in
ADHD treatment. Interestingly, the effect size for MASs
was much smaller than that for LDX. The effect size for
MPH compared with placebo was also lower than for LDX,
which has been shown in multiple previous meta-analyses
in adults, children, and adolescents. One of the main rea-
sons could be pooling all MPH RCTs into a heterogeneous
group regardless of their release form. This approach has
been seen in previous meta-analyses.1,6 However, more
RCTs, especially of amphetamines and MDF, are necessary
to corroborate these promising findings and arrive at defi-
nite conclusions. Heterogeneity in outcomes among MPH
studies was the highest among all the stimulants (I2 = 21%).
Larger heterogeneity in the MPH group could be related to
different MPH release forms.
In the meta-analysis of CGI-I score, all medications,
except MDF (1 trial), showed a higher probability of global
improvement compared with placebo. The RR values indi-
cate an apparent advantage of LDX and MASs over MPH.
The heterogeneity in all groups might not be relevant (under
30%), except for MPH (I2 = 54%). It could be in part
explained by a different definition of treatment improve-
ment among the MPH studies (CGI-I ⩽ 2 alone or coupled
with various cut-off scores). Another possible cause is that
*References 23, 27, 28, 31, 37, 38, 42
Stuhec et al 11
the improvement may not have been observed because of
the short duration of some RCTs. In 2 larger studies, a sub-
stantial discontinuation could be a likely contributor to a
marked heterogeneity. Therefore, interpretation of these
results requires extra caution.
Only 1 trial of each drug evaluated improvement in EF
deficit. EF plays an essential role in an adult’s life, and its
deficit is believed to be closely related to attention deficit
and ADHD-related quality of life.43 The largest effect size
in EF improvement was produced by LDX followed by
MASs, MPH, and MDF. Given the importance of EF in
adult life, further trials are essential to determine the effi-
cacy of stimulant use in adult ADHD.
In addtion to drug efficacy assessment, acceptability and
tolerability are also critical to successful treatment. Except
for MASs, amphetamines did not demonstrate a signifi-
cantly higher acceptability (retention in treatment) than pla-
cebo and was significantly lower for MDF. Insignificant
results along with marked heterogeneity may lead to flawed
interpretation of acceptability. The total RR in the meta-
analysis of discontinuation caused by adverse events was
higher than placebo for all stimulants; however, it was bor-
deline statistically significant for LDX. Incidence among
the studies, however, did not exceed 20%. There were some
differences in RRs between different dosing methods (fixed
vs titration); nevertheless, it is not essential for clinical
practice because titration to the optimal dose is advised
according to ADHD treatment guidelines.38 It should also
be noted that some of the adverse effects occur over a lon-
ger period, meaning that positive results need to be treated
with extra caution because the trials were of short
The quality and risk of bias assessment were an impor-
tant part of this meta-analysis. The majority of trials were
attributed at least 1 high risk of bias judgment (11/20).
Although the risk of bias was significant in many RCTs, a
sensitivity analysis did not find changes in the outcomes.
The findings should be observed together with the limita-
tions of this meta-analysis. Only stimulants were included,
and therefore, comparison of the efficacy with nonstimu-
lants (e.g., atomoxetine) was not possible. For such a com-
parison, additional meta-analyses are needed. The second
relevant limitation was that treatment-naïve patients were
not exclusively eligible, which means that some patients had
already been treated with different medications. This
approach increased the number of eligible RCTs for the
meta-analysis and provided results that more closely reflect
real clinical practice. Another important limitation concerns
comorbidity. Comorbidity of ADHD with psychiatric disor-
ders is substantial.6 Many RCTs included some comorbidity,
which might have led to a higher heterogeneity of the results.
However, such results better simulate real clinical practice
because ADHD is often treated simultaneously with
other mental conditions (e.g., anxiety disorders, depressive
symptoms).6 Additional limitation could originate from
ADHD subtype of the included patients that was reported in
only 10 trial reports. Even fewer, 7 trial reports considered
ADHD subscale scores as secondary outcomes. However,
results were reported in a manner that did not allow data
extraction. We, therefore, decided to analyze the ADHD
total scores only. Study duration is a significant limitation as
well. In real clinical practice, ADHD is often treated at least
for 1 year, or often longer; hence, these results could not be
extrapolated to chronic treatment. Long-term effects are par-
ticularly problematic with LDX because the available data
on this are scarce. Another significant limitation is heteroge-
neity, which was high in MPH, possibly connected with a
different formulation of MPH (immediate vs modified
release). When analyzing withdrawal resulting from adverse
events and all-cause treatment discontinuation, it is impor-
tant to consider that heterogeneity may be inevitable because
of study design and protocol requirements. Some trials can
have an open-label follow-on phase allowing patients to
drop out later on in the study.9 We did not contact authors of
any of the included studies to seek original trial protocols or
unreported data, or to clarify ambiguities. Finally, the indi-
rect network meta-analysis allows multiple pairwise com-
parisons between the drugs, which can overcome the
principal limitation of the standard meta-analysis—that is,
comparison of single drug with placebo.
Despite significant limitations, our findings have several
clinical implications and represent a considerable update to
previous meta-analyses, particularly with most recent RCTs
that have not previously been included. With the most
recent RCTs included, the results lend themselves to broader
implication, including the updates to the ADHD treatment
guidelines. Despite the importance of this and other meta-
analyses, other factors also need to be taken into account
when planning treatment with drugs, such as well-designed
clinical trials, treatment guidelines, drug tolerability, drug-
drug interactions, and numerous others. Because only a
minority of adults with ADHD receive adequate pharmaco-
logical treatment,45,46 the results of this study could also
support evidence-based decision regarding the pharmaco-
logical treatment option for patients with ADHD.
These results suggest that LDX has the largest effect size
and has a promising potential for treating adults with
ADHD. MASs have a larger effect size on EF improvement
than MPH. Five of the included trials were not evaluated in
any of the published meta-analyses, and therefore, this
research is a necessary addition to the previous assessments
of stimulants in the treatment of ADHD. Given that the
results of this meta-analysis did not find any favorable
effect for MDF, several new RCTs of MDF are warranted to
examine its value in the treatment of ADHD.
12 Annals of Pharmacotherapy 00(0)
Some results of this article were also presented as a poster at the
European College of Neuropsychopharmacology (ECNP) Congress
in Vienna in September 2016.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect
to the research, authorship, and/or publication of this article.
The authors received no financial support for the research, author-
ship, and/or publication of this article.
Supplemental material is available for this article online.
1. Stuhec M, Munda B, Svab V, Locatelli I. Comparative effi-
cacy and acceptability of atomoxetine, lisdexamfetamine,
bupropion and methylphenidate in treatment of attention
deficit hyperactivity disorder in children and adolescents:
a meta-analysis with focus on bupropion. J Affect Disord.
2. Polanczyk GV, Salum GA, Sugaya LS, Caye A, Rohde LA.
Annual research review: a meta-analysis of the worldwide prev-
alence of mental disorders in children and adolescents. J Child
Psychol Psychiatry. 2015;56:345-365. doi:10.1111/jcpp.12381
3. Štuhec M, Švab V, Locatelli I. Prevalence and incidence of
attention-deficit/hyperactivity disorder in Slovenian children
and adolescents: a database study from a national perspective.
Croat Med J. 2015;56:159-165. doi:10.3325/cmj.2015.56.159
4. Balint S, Czobor P, Meszaros A, Simon V, Bitter I.
Neuropsychological impairments in adult attention deficit
hyperactivity disorder: a literature review [in Hungarian].
Psychiatr Hung. 2008;23:324-335.
5. Dalsgaard S, Østergaard SD, Leckman JF, Mortensen PB,
Pedersen MG. Mortality in children, adolescents, and adults
with attention deficit hyperactivity disorder: a nationwide
cohort study. Lancet. 2015;385:2190-2196. doi:10.1016/
6. Bolea-Alamañac B, Nutt DJ, Adamou M, et al; British Association
for Psychopharmacology. Evidence-based guidelines for the
pharmacological management of attention deficit hyperac-
tivity disorder: update on recommendations from the British
Association for Psychopharmacology. J Psychopharmacol.
7. Faraone SV. Using meta-analysis to compare the efficacy
of medications for attention-deficit/hyperactivity disorder in
youths. P T. 2009;34:678-694.
8. Faraone SV, Biederman J, Spencer TJ, Aleardi M. Comparing
the efficacy of medications for ADHD using meta-analysis.
9. Roskell NS, Setyawan J, Zimovetz EA, Hodgkins P.
Systematic evidence synthesis of treatments for ADHD in
children and adolescents: indirect treatment comparisons of
lisdexamfetamine with methylphenidate and atomoxetine.
Curr Med Res Opin. 2014;30:1673-1685. doi:10.1185/0300
10. Castells X, Ramos-Quiroga JA, Bosch R, Nogueira M,
Casas M. Amphetamines for attention deficit hyperactivity
disorder (ADHD) in adults. Cochrane Database Syst Rev.
11. Castells X, Ramos-Quiroga JA, Rigau D, et al. Efficacy of
methylphenidate for adults with attention-deficit hyper-
activity disorder: a meta-regression analysis. CNS Drugs.
12. Cunill R, Castells X, Tobias A, Capellà D. Efficacy, safety
and variability in pharmacotherapy for adults with attention
deficit hyperactivity disorder: a meta-analysis and meta-
regression in over 9000 patients. Psychopharmacology (Berl).
13. Epstein T, Patsopoulos NA, Weiser M. Immediate-release
methylphenidate for attention deficit hyperactivity dis-
order (ADHD) in adults. Cochrane Database Syst Rev.
14. Faraone SV. Understanding the effect size of lisdexamfetamine
dimesylate for treating ADHD in children and adults. J Atten
Disord. 2012;16:128-137. doi:10.1177/1087054710379738
15. Faraone SV, Glatt SJ. A comparison of the efficacy of medi-
cations for adult attention-deficit/hyperactivity disorder using
meta-analysis of effect sizes. J Clin Psychiatry. 2010;71:754-
16. Faraone SV, Spencer T, Aleardi M, Pagano C, Biederman J.
Meta-analysis of the efficacy of methylphenidate for treating adult
attention-deficit/hyperactivity disorder. J Clin Psychopharmacol.
17. Koesters M, Becker T, Kilian R, Fegert JM, Weinmann S. Limits
of meta-analysis: methylphenidate in the treatment of adult
attention-deficit hyperactivity disorder. J Psychopharmacol.
18. Maneeton N, Maneeton B, Suttajit S, Reungyos J,
Srisurapanont M, Martin SD. Exploratory meta-analysis on
lisdexamfetamine versus placebo in adult ADHD. Drug Des
Devel Ther. 2014;8:1685-1693. doi:10.2147/DDDT.S68393
19. Mészáros A, Czobor P, Bálint S, Komlósi S, Simon V, Bitter I.
Pharmacotherapy of adult attention deficit hyperactivity disor-
der (ADHD): a meta-analysis. Int J Neuropsychopharmacol.
20. Peterson K, McDonagh MS, Fu R. Comparative benefits and
harms of competing medications for adults with attention-
deficit hyperactivity disorder: a systematic review and indi-
rect comparison meta-analysis. Psychopharmacology (Berl).
21. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA state-
ment for reporting systematic reviews and meta-analyses of
studies that evaluate health care interventions: explanation and
elaboration. J Clin Epidemiol. 2009;62:e1-e34. doi:10.1016/j.
22. Higgins JPT, Green S, eds. Cochrane Handbook for
Systematic Reviews of Interventions. Version 5.1.0. London,
England: Cochrane Collaboration; 2011. http://handbook-5-1.
cochrane.org/. Accessed August 6, 2018.
23. Spencer TJ, Adler LA, Weisler RH, Youcha SH. Triple-
bead mixed amphetamine salts (SPD465), a novel, enhanced
extended-release amphetamine formulation for the treat-
Stuhec et al 13
ment of adults with ADHD: a randomized, double-blind,
multicenter, placebo-controlled study. J Clin Psychiatry.
24. Weisler RH, Biederman J, Spencer TJ, et al. Mixed
amphetamine salts extended-release in the treatment of
adult ADHD: a randomized, controlled trial. CNS Spectr.
25. Adler L, Goodman DW, Kollins SH, et al; 303 Study Group.
Double-blind, placebo-controlled study of the efficacy
and safety of lisdexamfetamine dimesylate in adults with
attention-deficit/hyperactivity disorder. J Clin Psychiatry.
26. Adler LA, Dirks B, Deas PF, et al. Lisdexamfetamine dimesylate
in adults with attention-deficit/hyperactivity disorder who report
clinically significant impairment in executive function: results
from a randomized, double-blind, placebo-controlled study. J
Clin Psychiatry. 2013;74:694-702. doi:10.4088/JCP.12m08144
27. Biederman J, Fried R, Hammerness P, et al. The effects of
lisdexamfetamine dimesylate on the driving performance
of young adults with ADHD: a randomized, double-blind,
placebo-controlled study using a validated driving simulator
paradigm. J Psychiatr Res. 2012;46:484-491. doi:10.1016/j.
28. Adler LA, Zimmerman B, Starr HL, et al. Efficacy and
safety of OROS methylphenidate in adults with attention-
deficit/hyperactivity disorder. J Clin Psychopharmacol.
29. Biederman J, Mick E, Surman C, et al. A randomized, pla-
cebo-controlled trial of OROS methylphenidate in adults
with attention-deficit/hyperactivity disorder. Biol Psychiatry.
30. Biederman J, Mick E, Surman C, et al. A randomized, 3-phase,
34-week, double-blind, long-term efficacy study of osmotic-
release oral system-methylphenidate in adults with attention-
deficit/hyperactivity disorder. J Clin Psychopharmacol.
31. Casas M, Rösler M, Kooij SJJ, et al. Efficacy and safety
of prolonged-release OROS methylphenidate in adults with
attention deficit/hyperactivity disorder: a 13-week, random-
ized, double-blind, placebo-controlled, fixed-dose study.
World J Biol Psychiatry. 2013;14:268-281. doi:10.3109/156
32. Goodman DW, Starr HL, Ma Y-W, Rostain AL, Ascher S,
Armstrong RB. Randomized, 6-week, placebo-controlled study
of treatment for adult attention-deficit/hyperactivity disorder:
individualized dosing of osmotic-release oral system (OROS)
methylphenidate with a goal of symptom remission. J Clin
Psychiatry. 2017;78:105-114. doi:10.4088/JCP.15m10348
33. Huss M, Ginsberg Y, Tvedten T, et al. Methylphenidate
hydrochloride modified-release in adults with attention defi-
cit hyperactivity disorder: a randomized double-blind pla-
cebo-controlled trial. Adv Ther. 2014;31:44-65. doi:10.1007/
34. Kuperman S, Perry PJ, Gaffney GR, et al. Bupropion SR vs
methylphenidate vs placebo for attention deficit hyperactivity
disorder in adults. Ann Clin Psychiatry. 2001;13:129-134.
35. Medori R, Ramos-Quiroga JA, Casas M, et al. A random-
ized, placebo-controlled trial of three fixed dosages of
prolonged-release OROS methylphenidate in adults with
attention-deficit/hyperactivity disorder. Biol Psychiatry.
36. Retz W, Rösler M, Ose C, et al. Multiscale assessment of
treatment efficacy in adults with ADHD: a randomized pla-
cebo-controlled, multi-centre study with extended-release
methylphenidate. World J Biol Psychiatry. 2012;13:48-59.
37. Rösler M, Fischer R, Ammer R, Ose C, Retz W. A randomised,
placebo-controlled, 24-week, study of low-dose extended-
release methylphenidate in adults with attention-deficit/
hyperactivity disorder. Eur Arch Psychiatry Clin Neurosci.
38. Spencer T, Biederman J, Wilens T, et al. A large, double-
blind, randomized clinical trial of methylphenidate in the
treatment of adults with attention-deficit/hyperactivity dis-
order. Biol Psychiatry. 2005;57:456-463. doi:10.1016/j.bio-
39. Spencer TJ, Adler LA, McGough JJ, Muniz R, Jiang H,
Pestreich L; Adult ADHD Research Group. Efficacy and safety
of dexmethylphenidate extended-release capsules in adults
with attention-deficit/hyperactivity disorder. Biol Psychiatry.
40. Takahashi N, Koh T, Tominaga Y, Saito Y, Kashimoto Y,
Matsumura T. A randomized, double-blind, placebo-con-
trolled, parallel-group study to evaluate the efficacy and safety
of osmotic-controlled release oral delivery system methylphe-
nidate HCl in adults with attention-deficit/hyperactivity disor-
der in Japan. World J Biol Psychiatry. 2014;15:488-498. doi:1
41. Weisler RH, Pandina GJ, Daly EJ, Cooper K, Gassmann-Mayer
C; 31001074-ATT2001 Study Investigators. Randomized
clinical study of a histamine H3 receptor antagonist for the
treatment of adults with attention-deficit hyperactivity disor-
der. CNS Drugs. 2012;26:421-434. doi:10.2165/11631990-
42. Arnold VK, Feifel D, Earl CQ, Yang R, Adler LA. A 9-week,
randomized, double-blind, placebo-controlled, parallel-
group, dose-finding study to evaluate the efficacy and safety
of modafinil as treatment for adults with ADHD. J Atten
Disord. 2014;18:133-144. doi:10.1177/1087054712441969
43. Biederman J, Mick E, Fried R, Wilner N, Spencer TJ, Faraone
SV. Are stimulants effective in the treatment of executive
function deficits? Results from a randomized double blind
study of OROS-methylphenidate in adults with ADHD. Eur
Neuropsychopharmacol. 2011;21:508-515. doi:10.1016/j.
44. Banaschewski T, Coghill D, Santosh P, et al. Long-acting med-
ications for the hyperkinetic disorders: a systematic review and
European treatment guideline. Eur Child Adolesc Psychiatry.
45. Retz W, Retz-Junginger P, Thome J, Rösler M.
Pharmacological treatment of adult ADHD in Europe. World
J Biol Psychiatry. 2011;12(suppl 1):89-94. doi:10.3109/1562
46. Stuhec M, Locatelli I. Age-related pharmacotherapy of atten-
tion deficit hyperactivity disorder in Slovenia in children
and adolescents: a population-based study. Eur Psychiatry.